SentiSight SDK questions

General information about Neurotechnology and its software

Re: [SentiSight] Opening camera in sentisight

Postby vaidasz » Oct 22, 2012 15:24

patiljagdish wrote:Hello,
I am new to sentisight and sorry for asking this question.
How to read and display from camera ?
Are there any simple tutorials to learn from scratch ?


Hello,

Please check samples provided with SDK on how to read and display from the camera. There is also a tutorial provided with SDK that shows how to get the frames from the camera.
Tutorials provided with SDK show how you could learn and recognize the objects.
Vaidas Z
Neurotechnology
vaidasz
Neurotechnology
 
Posts: 559
Joined: Apr 30, 2010 06:57
Location: Vilnius, Lithuania

How to save learn models as image in SentiSightSampleCS

Postby patiljagdish » Oct 26, 2012 04:27

Hello ,
I am new to sentisight.
In the given SentisightSampleCS, i want to save all the learn image in some folder.
After learning the image or video using mouse, i want to create the bounding rectangle around the part of image or video selected and save it. Currently it won't save the learn image and pass it to database after clicking the Save model button.
Can anyone help me ?

Admin: merged to "SentiSight SDK questions"
patiljagdish
 
Posts: 15
Joined: Oct 22, 2012 10:33

How to save the learn image in given SentisightSampleCS

Postby patiljagdish » Oct 26, 2012 05:20

Hello,
I am new to Sentisight. In the given sample SentisightSampleCS after learning the image or video file by using mouse, how can i can save that learn model in some folder as image.
I want to create the bounding rectangle around each and every learn model to save it .
Can anyone help me ?
patiljagdish
 
Posts: 15
Joined: Oct 22, 2012 10:33

Re: How to save learn models as image in SentiSightSampleCS

Postby Martynas » Oct 26, 2012 08:39

patiljagdish wrote:Hello ,
I am new to sentisight.
In the given SentisightSampleCS, i want to save all the learn image in some folder.
After learning the image or video using mouse, i want to create the bounding rectangle around the part of image or video selected and save it. Currently it won't save the learn image and pass it to database after clicking the Save model button.
Can anyone help me ?

patiljagdish wrote:Hello,
I am new to Sentisight. In the given sample SentisightSampleCS after learning the image or video file by using mouse, how can i can save that learn model in some folder as image.
I want to create the bounding rectangle around each and every learn model to save it .
Can anyone help me ?

Hi,

such functionality is not implemented in the Sample. So if you need such functionality you need to implement it by yourself.
Best regards,

Martynas
Software Consultant
Neurotechnology
Martynas
Neurotechnology
 
Posts: 2195
Joined: Sep 19, 2008 10:02
Location: Vilnius, Lithuania

Re: How to save learn models as image in SentiSightSampleCS

Postby patiljagdish » Oct 26, 2012 08:44

Yes, i know this functionality is not there in sample. What i want to know is in which format the learn model is pass to database and frankly speaking i was not able to understand where it happen.
patiljagdish
 
Posts: 15
Joined: Oct 22, 2012 10:33

Re: How to save learn models as image in SentiSightSampleCS

Postby patiljagdish » Oct 29, 2012 03:45

Yes i know that, but i was not able to identify how and where the image(model) is passed to database and in which format.
It will be very helpful if you guide me.

Thanks
patiljagdish
 
Posts: 15
Joined: Oct 22, 2012 10:33

Re: How to save learn models as image in SentiSightSampleCS

Postby patiljagdish » Oct 29, 2012 06:09

I know that, but i am not able to understand where in source code sentisight take image of model and passed to database and also how. Can we save model after after converting to bitmap(code in which model is passed to database) ?
patiljagdish
 
Posts: 15
Joined: Oct 22, 2012 10:33

Re: How to save learn models as image in SentiSightSampleCS

Postby Martynas » Oct 29, 2012 08:39

patiljagdish wrote:Yes, i know this functionality is not there in sample. What i want to know is in which format the learn model is pass to database and frankly speaking i was not able to understand where it happen.

patiljagdish wrote:Yes i know that, but i was not able to identify how and where the image(model) is passed to database and in which format.
It will be very helpful if you guide me.

Thanks

patiljagdish wrote:I know that, but i am not able to understand where in source code sentisight take image of model and passed to database and also how. Can we save model after after converting to bitmap(code in which model is passed to database) ?

first of all i would like to ask you to read the documentation of the SentiSight SDK, section "2.3 Main SentiSight API functions". Here you will find that the SentiSight engine is using models, but not images for recognition. Creation of model from image is called "Learning".
It dope snot matter, what format of image you will be suing, as it will need to create the NImage from your image (bmp, jpg and etc.) and then use the SEEngine.Learning.AddToModel method to learn the model.
If you have a database with already learned models, then you will need to load the to SEEngine at the start up of it.
If you need to save images in the database too, then you can implement it by yourself. Just use standard tools of your programming language to save image in the database.
Best regards,

Martynas
Software Consultant
Neurotechnology
Martynas
Neurotechnology
 
Posts: 2195
Joined: Sep 19, 2008 10:02
Location: Vilnius, Lithuania

How to retrieve the images added to model [SentiSight]

Postby patiljagdish » Nov 02, 2012 05:11

Hello,
I want to know how to retrieve the images added to model, while learning process. I want those images in any standard format(.png , .jpg, etc..).

Admin: merged to "SentiSight SDK questions"
patiljagdish
 
Posts: 15
Joined: Oct 22, 2012 10:33

Re: How to retrieve the images added to model [SentiSight]

Postby Martynas » Nov 05, 2012 08:23

patiljagdish wrote:Hello,
I want to know how to retrieve the images added to model, while learning process. I want those images in any standard format(.png , .jpg, etc..).

Hello,

model does not contain any images. so if you need to have images, then you should save them separately.
Best regards,

Martynas
Software Consultant
Neurotechnology
Martynas
Neurotechnology
 
Posts: 2195
Joined: Sep 19, 2008 10:02
Location: Vilnius, Lithuania

Re: SentiSight SDK questions

Postby benjimitsu » Dec 05, 2012 10:18

Hi all,
This is my very first time of using Sentisight. I want to investigate the C++ demo code for Sentisight recognition using Visual Studio 2008. I have already installed wxwidget 2.8.12. All sub-projects in wx.dsw have been compiled successfully. Then, I go for compiling the SentiSightSampleWX. The include and library I have configured as:
Code: Select all
C:\wxWidgets-2.8.12\include
C:\wxWidgets-2.8.12\include\msvc
C:\wxWidgets-2.8.12\lib\vc_lib

However, there are lots of error that I cannot figure out. Plz help me. The errors as listed below.
Code: Select all
1>------ Build started: Project: SentiSightSampleWX, Configuration: Debug Win32 ------
1>Linking...
1>msvcprtd.lib(MSVCP90D.dll) : error LNK2005: "void __cdecl std::_Debug_message(wchar_t const *,wchar_t const *,unsigned int)" (?_Debug_message@std@@YAXPB_W0I@Z) already defined in libcpmtd.lib(stdthrow.obj)
1>msvcprtd.lib(MSVCP90D.dll) : error LNK2005: "public: __thiscall std::_Lockit::~_Lockit(void)" (??1_Lockit@std@@QAE@XZ) already defined in libcpmtd.lib(xlock.obj)
1>msvcprtd.lib(MSVCP90D.dll) : error LNK2005: "public: __thiscall std::_Lockit::_Lockit(int)" (??0_Lockit@std@@QAE@H@Z) already defined in libcpmtd.lib(xlock.obj)
1>msvcprtd.lib(MSVCP90D.dll) : error LNK2005: "public: void __thiscall std::_Container_base_secure::_Orphan_all(void)const " (?_Orphan_all@_Container_base_secure@std@@QBEXXZ) already defined in CameraConfigurationDialog.obj
1>msvcprtd.lib(MSVCP90D.dll) : error LNK2005: "public: __thiscall std::_Container_base_secure::~_Container_base_secure(void)" (??1_Container_base_secure@std@@QAE@XZ) already defined in CameraConfigurationDialog.obj
1>msvcprtd.lib(MSVCP90D.dll) : error LNK2005: "public: __thiscall std::basic_string<char,struct std::char_traits<char>,class std::allocator<char> >::~basic_string<char,struct std::char_traits<char>,class std::allocator<char> >(void)" (??1?$basic_string@DU?$char_traits@D@std@@V?$allocator@D@2@@std@@QAE@XZ) already defined in CameraConfigurationDialog.obj
1>msvcprtd.lib(MSVCP90D.dll) : error LNK2005: "public: __thiscall std::basic_string<char,struct std::char_traits<char>,class std::allocator<char> >::basic_string<char,struct std::char_traits<char>,class std::allocator<char> >(char const *)" (??0?$basic_string@DU?$char_traits@D@std@@V?$allocator@D@2@@std@@QAE@PBD@Z) already defined in CameraConfigurationDialog.obj
1>msvcprtd.lib(MSVCP90D.dll) : error LNK2005: "public: __thiscall std::basic_string<char,struct std::char_traits<char>,class std::allocator<char> >::basic_string<char,struct std::char_traits<char>,class std::allocator<char> >(class std::basic_string<char,struct std::char_traits<char>,class std::allocator<char> > const &)" (??0?$basic_string@DU?$char_traits@D@std@@V?$allocator@D@2@@std@@QAE@ABV01@@Z) already defined in CameraConfigurationDialog.obj
1>msvcprtd.lib(MSVCP90D.dll) : error LNK2005: "public: char const * __thiscall std::basic_string<char,struct std::char_traits<char>,class std::allocator<char> >::c_str(void)const " (?c_str@?$basic_string@DU?$char_traits@D@std@@V?$allocator@D@2@@std@@QBEPBDXZ) already defined in CameraConfigurationDialog.obj
1>msvcprtd.lib(MSVCP90D.dll) : error LNK2005: "public: __thiscall std::_Container_base_secure::_Container_base_secure(void)" (??0_Container_base_secure@std@@QAE@XZ) already defined in CameraConfigurationDialog.obj
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _free already defined in LIBCMTD.lib(dbgfree.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _wcscmp already defined in LIBCMTD.lib(wcscmp.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _wcslen already defined in LIBCMTD.lib(wcslen.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _malloc already defined in LIBCMTD.lib(dbgmalloc.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _towlower already defined in LIBCMTD.lib(towlower.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _iswspace already defined in LIBCMTD.lib(_wctype.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _realloc already defined in LIBCMTD.lib(dbgrealloc.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _wcsncmp already defined in LIBCMTD.lib(wcsncmp.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _iswalnum already defined in LIBCMTD.lib(_wctype.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: __wcsnicmp already defined in LIBCMTD.lib(wcsnicmp.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: __swprintf already defined in LIBCMTD.lib(swprintf.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _sprintf already defined in LIBCMTD.lib(sprintf.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _memmove already defined in LIBCMTD.lib(memmove.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _sscanf already defined in LIBCMTD.lib(sscanf.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _tolower already defined in LIBCMTD.lib(tolower.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _isspace already defined in LIBCMTD.lib(_ctype.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _strchr already defined in LIBCMTD.lib(strchr.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: "public: virtual __thiscall std::exception::~exception(void)" (??1exception@std@@UAE@XZ) already defined in LIBCMTD.lib(stdexcpt.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: "public: __thiscall std::exception::exception(void)" (??0exception@std@@QAE@XZ) already defined in LIBCMTD.lib(stdexcpt.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: "public: __thiscall std::exception::exception(class std::exception const &)" (??0exception@std@@QAE@ABV01@@Z) already defined in LIBCMTD.lib(stdexcpt.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: __invalid_parameter already defined in LIBCMTD.lib(invarg.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: __CrtDbgReportW already defined in LIBCMTD.lib(dbgrptw.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: "public: __thiscall std::exception::exception(char const * const &)" (??0exception@std@@QAE@ABQBD@Z) already defined in LIBCMTD.lib(stdexcpt.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _memmove_s already defined in LIBCMTD.lib(memmove_s.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _iswalpha already defined in LIBCMTD.lib(_wctype.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _iswdigit already defined in LIBCMTD.lib(_wctype.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _wcsftime already defined in LIBCMTD.lib(wcsftime.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: __wassert already defined in LIBCMTD.lib(wassert.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _mbstowcs already defined in LIBCMTD.lib(mbstowcs.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _wcstol already defined in LIBCMTD.lib(wcstol.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _wcstoul already defined in LIBCMTD.lib(wcstol.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: __errno already defined in LIBCMTD.lib(dosmap.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: __vswprintf_p already defined in LIBCMTD.lib(vswprnc.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _exit already defined in LIBCMTD.lib(crt0dat.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: __open_osfhandle already defined in LIBCMTD.lib(osfinfo.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: __get_osfhandle already defined in LIBCMTD.lib(osfinfo.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: __fileno already defined in LIBCMTD.lib(fileno.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _abort already defined in LIBCMTD.lib(abort.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: ___iob_func already defined in LIBCMTD.lib(_file.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _fflush already defined in LIBCMTD.lib(fflush.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _fputwc already defined in LIBCMTD.lib(fputwc.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _isalnum already defined in LIBCMTD.lib(_ctype.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _isalpha already defined in LIBCMTD.lib(_ctype.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _isdigit already defined in LIBCMTD.lib(_ctype.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _setlocale already defined in LIBCMTD.lib(setlocal.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: __gmtime64 already defined in LIBCMTD.lib(gmtime64.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: __localtime64 already defined in LIBCMTD.lib(loctim64.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: __wfopen already defined in LIBCMTD.lib(wfopen.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _fclose already defined in LIBCMTD.lib(fclose.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _fread already defined in LIBCMTD.lib(fread.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _fwrite already defined in LIBCMTD.lib(fwrite.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: __close already defined in LIBCMTD.lib(close.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: __read already defined in LIBCMTD.lib(read.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: __write already defined in LIBCMTD.lib(write.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: __commit already defined in LIBCMTD.lib(commit.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: __lseeki64 already defined in LIBCMTD.lib(lseeki64.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _wcstombs already defined in LIBCMTD.lib(wcstombs.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _fprintf already defined in LIBCMTD.lib(fprintf.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: __CrtSetDbgFlag already defined in LIBCMTD.lib(dbgheap.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _atoi already defined in LIBCMTD.lib(atox.obj)
1>MSVCRTD.lib(ti_inst.obj) : error LNK2005: "private: __thiscall type_info::type_info(class type_info const &)" (??0type_info@@AAE@ABV0@@Z) already defined in LIBCMTD.lib(typinfo.obj)
1>MSVCRTD.lib(ti_inst.obj) : error LNK2005: "private: class type_info & __thiscall type_info::operator=(class type_info const &)" (??4type_info@@AAEAAV0@ABV0@@Z) already defined in LIBCMTD.lib(typinfo.obj)
1>LINK : warning LNK4098: defaultlib 'MSVCRTD' conflicts with use of other libs; use /NODEFAULTLIB:library
1>LINK : warning LNK4098: defaultlib 'LIBCMT' conflicts with use of other libs; use /NODEFAULTLIB:library
1>wxsqlite3.obj : error LNK2001: unresolved external symbol "public: void __thiscall wxStringData::Free(void)" (?Free@wxStringData@@QAEXXZ)
1>SEView.obj : error LNK2001: unresolved external symbol "public: void __thiscall wxStringData::Free(void)" (?Free@wxStringData@@QAEXXZ)
1>TransformTool.obj : error LNK2001: unresolved external symbol "public: void __thiscall wxStringData::Free(void)" (?Free@wxStringData@@QAEXXZ)
1>VideoLoader.obj : error LNK2001: unresolved external symbol "public: void __thiscall wxStringData::Free(void)" (?Free@wxStringData@@QAEXXZ)
1>wxSampleConfig.obj : error LNK2019: unresolved external symbol "public: void __thiscall wxStringData::Free(void)" (?Free@wxStringData@@QAEXXZ) referenced in function "public: virtual void * __thiscall Neurotec::wxNBadImageFormatException::`scalar deleting destructor'(unsigned int)" (??_GwxNBadImageFormatException@Neurotec@@UAEPAXI@Z)
1>RangeValidator.obj : error LNK2001: unresolved external symbol "public: void __thiscall wxStringData::Free(void)" (?Free@wxStringData@@QAEXXZ)
1>RecognizeTask.obj : error LNK2001: unresolved external symbol "public: void __thiscall wxStringData::Free(void)" (?Free@wxStringData@@QAEXXZ)
1>SESample.obj : error LNK2001: unresolved external symbol "public: void __thiscall wxStringData::Free(void)" (?Free@wxStringData@@QAEXXZ)
1>SESampleFrm.obj : error LNK2001: unresolved external symbol "public: void __thiscall wxStringData::Free(void)" (?Free@wxStringData@@QAEXXZ)
1>LearnTask.obj : error LNK2001: unresolved external symbol "public: void __thiscall wxStringData::Free(void)" (?Free@wxStringData@@QAEXXZ)
1>ObjectSelection.obj : error LNK2001: unresolved external symbol "public: void __thiscall wxStringData::Free(void)" (?Free@wxStringData@@QAEXXZ)
1>OptionsDlg.obj : error LNK2001: unresolved external symbol "public: void __thiscall wxStringData::Free(void)" (?Free@wxStringData@@QAEXXZ)
1>Precompiled.obj : error LNK2001: unresolved external symbol "public: void __thiscall wxStringData::Free(void)" (?Free@wxStringData@@QAEXXZ)
1>FrameCapturerThread.obj : error LNK2001: unresolved external symbol "public: void __thiscall wxStringData::Free(void)" (?Free@wxStringData@@QAEXXZ)
1>ImageDirectoryLoader.obj : error LNK2001: unresolved external symbol "public: void __thiscall wxStringData::Free(void)" (?Free@wxStringData@@QAEXXZ)
1>ImageFileLoader.obj : error LNK2001: unresolved external symbol "public: void __thiscall wxStringData::Free(void)" (?Free@wxStringData@@QAEXXZ)
1>ImageSelections.obj : error LNK2001: unresolved external symbol "public: void __thiscall wxStringData::Free(void)" (?Free@wxStringData@@QAEXXZ)
1>CameraConfigurationDialog.obj : error LNK2001: unresolved external symbol "public: void __thiscall wxStringData::Free(void)" (?Free@wxStringData@@QAEXXZ)
1>Database.obj : error LNK2001: unresolved external symbol "public: void __thiscall wxStringData::Free(void)" (?Free@wxStringData@@QAEXXZ)
1>DrawSelectionTool.obj : error LNK2001: unresolved external symbol "public: void __thiscall wxStringData::Free(void)" (?Free@wxStringData@@QAEXXZ)
1>FrameCapturer.obj : error LNK2001: unresolved external symbol "public: void __thiscall wxStringData::Free(void)" (?Free@wxStringData@@QAEXXZ)
1>..\..\..\..\Bin\Win32_x86\SentiSightSampleWX.exe : fatal error LNK1120: 1 unresolved externals
1>Build log was saved at "file://c:\Program Files\Neurotechnology\SentiSight 3.1 SDK Trial\Samples\SentiSight\CPP\SentiSightSampleWX\Obj\Win32\Debug\BuildLog.htm"
1>SentiSightSampleWX - 94 error(s), 2 warning(s)
========== Build: 0 succeeded, 1 failed, 0 up-to-date, 0 skipped ==========

Thanks all.
benjimitsu
 
Posts: 3
Joined: Dec 05, 2012 02:05

Re: SentiSight SDK questions

Postby Martynas » Dec 06, 2012 12:28

benjimitsu wrote:Hi all,
This is my very first time of using Sentisight. I want to investigate the C++ demo code for Sentisight recognition using Visual Studio 2008. I have already installed wxwidget 2.8.12. All sub-projects in wx.dsw have been compiled successfully. Then, I go for compiling the SentiSightSampleWX. The include and library I have configured as:
Code: Select all
C:\wxWidgets-2.8.12\include
C:\wxWidgets-2.8.12\include\msvc
C:\wxWidgets-2.8.12\lib\vc_lib

However, there are lots of error that I cannot figure out. Plz help me. The errors as listed below.
Code: Select all
1>------ Build started: Project: SentiSightSampleWX, Configuration: Debug Win32 ------
1>Linking...
1>msvcprtd.lib(MSVCP90D.dll) : error LNK2005: "void __cdecl std::_Debug_message(wchar_t const *,wchar_t const *,unsigned int)" (?_Debug_message@std@@YAXPB_W0I@Z) already defined in libcpmtd.lib(stdthrow.obj)
1>msvcprtd.lib(MSVCP90D.dll) : error LNK2005: "public: __thiscall std::_Lockit::~_Lockit(void)" (??1_Lockit@std@@QAE@XZ) already defined in libcpmtd.lib(xlock.obj)
1>msvcprtd.lib(MSVCP90D.dll) : error LNK2005: "public: __thiscall std::_Lockit::_Lockit(int)" (??0_Lockit@std@@QAE@H@Z) already defined in libcpmtd.lib(xlock.obj)
1>msvcprtd.lib(MSVCP90D.dll) : error LNK2005: "public: void __thiscall std::_Container_base_secure::_Orphan_all(void)const " (?_Orphan_all@_Container_base_secure@std@@QBEXXZ) already defined in CameraConfigurationDialog.obj
1>msvcprtd.lib(MSVCP90D.dll) : error LNK2005: "public: __thiscall std::_Container_base_secure::~_Container_base_secure(void)" (??1_Container_base_secure@std@@QAE@XZ) already defined in CameraConfigurationDialog.obj
1>msvcprtd.lib(MSVCP90D.dll) : error LNK2005: "public: __thiscall std::basic_string<char,struct std::char_traits<char>,class std::allocator<char> >::~basic_string<char,struct std::char_traits<char>,class std::allocator<char> >(void)" (??1?$basic_string@DU?$char_traits@D@std@@V?$allocator@D@2@@std@@QAE@XZ) already defined in CameraConfigurationDialog.obj
1>msvcprtd.lib(MSVCP90D.dll) : error LNK2005: "public: __thiscall std::basic_string<char,struct std::char_traits<char>,class std::allocator<char> >::basic_string<char,struct std::char_traits<char>,class std::allocator<char> >(char const *)" (??0?$basic_string@DU?$char_traits@D@std@@V?$allocator@D@2@@std@@QAE@PBD@Z) already defined in CameraConfigurationDialog.obj
1>msvcprtd.lib(MSVCP90D.dll) : error LNK2005: "public: __thiscall std::basic_string<char,struct std::char_traits<char>,class std::allocator<char> >::basic_string<char,struct std::char_traits<char>,class std::allocator<char> >(class std::basic_string<char,struct std::char_traits<char>,class std::allocator<char> > const &)" (??0?$basic_string@DU?$char_traits@D@std@@V?$allocator@D@2@@std@@QAE@ABV01@@Z) already defined in CameraConfigurationDialog.obj
1>msvcprtd.lib(MSVCP90D.dll) : error LNK2005: "public: char const * __thiscall std::basic_string<char,struct std::char_traits<char>,class std::allocator<char> >::c_str(void)const " (?c_str@?$basic_string@DU?$char_traits@D@std@@V?$allocator@D@2@@std@@QBEPBDXZ) already defined in CameraConfigurationDialog.obj
1>msvcprtd.lib(MSVCP90D.dll) : error LNK2005: "public: __thiscall std::_Container_base_secure::_Container_base_secure(void)" (??0_Container_base_secure@std@@QAE@XZ) already defined in CameraConfigurationDialog.obj
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _free already defined in LIBCMTD.lib(dbgfree.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _wcscmp already defined in LIBCMTD.lib(wcscmp.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _wcslen already defined in LIBCMTD.lib(wcslen.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _malloc already defined in LIBCMTD.lib(dbgmalloc.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _towlower already defined in LIBCMTD.lib(towlower.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _iswspace already defined in LIBCMTD.lib(_wctype.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _realloc already defined in LIBCMTD.lib(dbgrealloc.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _wcsncmp already defined in LIBCMTD.lib(wcsncmp.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _iswalnum already defined in LIBCMTD.lib(_wctype.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: __wcsnicmp already defined in LIBCMTD.lib(wcsnicmp.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: __swprintf already defined in LIBCMTD.lib(swprintf.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _sprintf already defined in LIBCMTD.lib(sprintf.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _memmove already defined in LIBCMTD.lib(memmove.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _sscanf already defined in LIBCMTD.lib(sscanf.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _tolower already defined in LIBCMTD.lib(tolower.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _isspace already defined in LIBCMTD.lib(_ctype.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _strchr already defined in LIBCMTD.lib(strchr.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: "public: virtual __thiscall std::exception::~exception(void)" (??1exception@std@@UAE@XZ) already defined in LIBCMTD.lib(stdexcpt.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: "public: __thiscall std::exception::exception(void)" (??0exception@std@@QAE@XZ) already defined in LIBCMTD.lib(stdexcpt.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: "public: __thiscall std::exception::exception(class std::exception const &)" (??0exception@std@@QAE@ABV01@@Z) already defined in LIBCMTD.lib(stdexcpt.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: __invalid_parameter already defined in LIBCMTD.lib(invarg.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: __CrtDbgReportW already defined in LIBCMTD.lib(dbgrptw.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: "public: __thiscall std::exception::exception(char const * const &)" (??0exception@std@@QAE@ABQBD@Z) already defined in LIBCMTD.lib(stdexcpt.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _memmove_s already defined in LIBCMTD.lib(memmove_s.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _iswalpha already defined in LIBCMTD.lib(_wctype.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _iswdigit already defined in LIBCMTD.lib(_wctype.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _wcsftime already defined in LIBCMTD.lib(wcsftime.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: __wassert already defined in LIBCMTD.lib(wassert.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _mbstowcs already defined in LIBCMTD.lib(mbstowcs.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _wcstol already defined in LIBCMTD.lib(wcstol.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _wcstoul already defined in LIBCMTD.lib(wcstol.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: __errno already defined in LIBCMTD.lib(dosmap.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: __vswprintf_p already defined in LIBCMTD.lib(vswprnc.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _exit already defined in LIBCMTD.lib(crt0dat.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: __open_osfhandle already defined in LIBCMTD.lib(osfinfo.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: __get_osfhandle already defined in LIBCMTD.lib(osfinfo.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: __fileno already defined in LIBCMTD.lib(fileno.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _abort already defined in LIBCMTD.lib(abort.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: ___iob_func already defined in LIBCMTD.lib(_file.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _fflush already defined in LIBCMTD.lib(fflush.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _fputwc already defined in LIBCMTD.lib(fputwc.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _isalnum already defined in LIBCMTD.lib(_ctype.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _isalpha already defined in LIBCMTD.lib(_ctype.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _isdigit already defined in LIBCMTD.lib(_ctype.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _setlocale already defined in LIBCMTD.lib(setlocal.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: __gmtime64 already defined in LIBCMTD.lib(gmtime64.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: __localtime64 already defined in LIBCMTD.lib(loctim64.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: __wfopen already defined in LIBCMTD.lib(wfopen.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _fclose already defined in LIBCMTD.lib(fclose.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _fread already defined in LIBCMTD.lib(fread.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _fwrite already defined in LIBCMTD.lib(fwrite.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: __close already defined in LIBCMTD.lib(close.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: __read already defined in LIBCMTD.lib(read.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: __write already defined in LIBCMTD.lib(write.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: __commit already defined in LIBCMTD.lib(commit.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: __lseeki64 already defined in LIBCMTD.lib(lseeki64.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _wcstombs already defined in LIBCMTD.lib(wcstombs.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _fprintf already defined in LIBCMTD.lib(fprintf.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: __CrtSetDbgFlag already defined in LIBCMTD.lib(dbgheap.obj)
1>MSVCRTD.lib(MSVCR90D.dll) : error LNK2005: _atoi already defined in LIBCMTD.lib(atox.obj)
1>MSVCRTD.lib(ti_inst.obj) : error LNK2005: "private: __thiscall type_info::type_info(class type_info const &)" (??0type_info@@AAE@ABV0@@Z) already defined in LIBCMTD.lib(typinfo.obj)
1>MSVCRTD.lib(ti_inst.obj) : error LNK2005: "private: class type_info & __thiscall type_info::operator=(class type_info const &)" (??4type_info@@AAEAAV0@ABV0@@Z) already defined in LIBCMTD.lib(typinfo.obj)
1>LINK : warning LNK4098: defaultlib 'MSVCRTD' conflicts with use of other libs; use /NODEFAULTLIB:library
1>LINK : warning LNK4098: defaultlib 'LIBCMT' conflicts with use of other libs; use /NODEFAULTLIB:library
1>wxsqlite3.obj : error LNK2001: unresolved external symbol "public: void __thiscall wxStringData::Free(void)" (?Free@wxStringData@@QAEXXZ)
1>SEView.obj : error LNK2001: unresolved external symbol "public: void __thiscall wxStringData::Free(void)" (?Free@wxStringData@@QAEXXZ)
1>TransformTool.obj : error LNK2001: unresolved external symbol "public: void __thiscall wxStringData::Free(void)" (?Free@wxStringData@@QAEXXZ)
1>VideoLoader.obj : error LNK2001: unresolved external symbol "public: void __thiscall wxStringData::Free(void)" (?Free@wxStringData@@QAEXXZ)
1>wxSampleConfig.obj : error LNK2019: unresolved external symbol "public: void __thiscall wxStringData::Free(void)" (?Free@wxStringData@@QAEXXZ) referenced in function "public: virtual void * __thiscall Neurotec::wxNBadImageFormatException::`scalar deleting destructor'(unsigned int)" (??_GwxNBadImageFormatException@Neurotec@@UAEPAXI@Z)
1>RangeValidator.obj : error LNK2001: unresolved external symbol "public: void __thiscall wxStringData::Free(void)" (?Free@wxStringData@@QAEXXZ)
1>RecognizeTask.obj : error LNK2001: unresolved external symbol "public: void __thiscall wxStringData::Free(void)" (?Free@wxStringData@@QAEXXZ)
1>SESample.obj : error LNK2001: unresolved external symbol "public: void __thiscall wxStringData::Free(void)" (?Free@wxStringData@@QAEXXZ)
1>SESampleFrm.obj : error LNK2001: unresolved external symbol "public: void __thiscall wxStringData::Free(void)" (?Free@wxStringData@@QAEXXZ)
1>LearnTask.obj : error LNK2001: unresolved external symbol "public: void __thiscall wxStringData::Free(void)" (?Free@wxStringData@@QAEXXZ)
1>ObjectSelection.obj : error LNK2001: unresolved external symbol "public: void __thiscall wxStringData::Free(void)" (?Free@wxStringData@@QAEXXZ)
1>OptionsDlg.obj : error LNK2001: unresolved external symbol "public: void __thiscall wxStringData::Free(void)" (?Free@wxStringData@@QAEXXZ)
1>Precompiled.obj : error LNK2001: unresolved external symbol "public: void __thiscall wxStringData::Free(void)" (?Free@wxStringData@@QAEXXZ)
1>FrameCapturerThread.obj : error LNK2001: unresolved external symbol "public: void __thiscall wxStringData::Free(void)" (?Free@wxStringData@@QAEXXZ)
1>ImageDirectoryLoader.obj : error LNK2001: unresolved external symbol "public: void __thiscall wxStringData::Free(void)" (?Free@wxStringData@@QAEXXZ)
1>ImageFileLoader.obj : error LNK2001: unresolved external symbol "public: void __thiscall wxStringData::Free(void)" (?Free@wxStringData@@QAEXXZ)
1>ImageSelections.obj : error LNK2001: unresolved external symbol "public: void __thiscall wxStringData::Free(void)" (?Free@wxStringData@@QAEXXZ)
1>CameraConfigurationDialog.obj : error LNK2001: unresolved external symbol "public: void __thiscall wxStringData::Free(void)" (?Free@wxStringData@@QAEXXZ)
1>Database.obj : error LNK2001: unresolved external symbol "public: void __thiscall wxStringData::Free(void)" (?Free@wxStringData@@QAEXXZ)
1>DrawSelectionTool.obj : error LNK2001: unresolved external symbol "public: void __thiscall wxStringData::Free(void)" (?Free@wxStringData@@QAEXXZ)
1>FrameCapturer.obj : error LNK2001: unresolved external symbol "public: void __thiscall wxStringData::Free(void)" (?Free@wxStringData@@QAEXXZ)
1>..\..\..\..\Bin\Win32_x86\SentiSightSampleWX.exe : fatal error LNK1120: 1 unresolved externals
1>Build log was saved at "file://c:\Program Files\Neurotechnology\SentiSight 3.1 SDK Trial\Samples\SentiSight\CPP\SentiSightSampleWX\Obj\Win32\Debug\BuildLog.htm"
1>SentiSightSampleWX - 94 error(s), 2 warning(s)
========== Build: 0 succeeded, 1 failed, 0 up-to-date, 0 skipped ==========

Thanks all.

Hi,

please use instructions below to recompile wxWidgets.

1. Open file C:\wxWidgets-2.8.12\include\wx\msw\setup.h(579) and set wxUSE_GRAPHICS_CONTEXT value to 1, i.e. #define wxUSE_GRAPHICS_CONTEXT 1
2. Open VS command prompt.
3. Go to C:\wxWidgets-2.8.12\build\msw
4. nmake /A /f makefile.vc UNICODE=1 USE_GDIPLUS=1 RUNTIME_LIBS=static CPPFLAGS=/MTd BUILD=debug
5. nmake /A /f makefile.vc UNICODE=1 USE_GDIPLUS=1 RUNTIME_LIBS=static CPPFLAGS=/MT BUILD=release
6. Open VS 64-bit command prompt.
7. Go to C:\wxWidgets-2.8.12\build\msw
8. nmake /A /f makefile.vc UNICODE=1 USE_GDIPLUS=1 TARGET_CPU=amd64 RUNTIME_LIBS=static CPPFLAGS=/MTd BUILD=debug
9. nmake /A /f makefile.vc UNICODE=1 USE_GDIPLUS=1 TARGET_CPU=amd64 RUNTIME_LIBS=static CPPFLAGS=/MT BUILD=release

Finally, Visual Studio include and library paths have to be setup.

Include:

C:\wxWidgets-2.8.12\include
C:\wxWidgets-2.8.12\include\msvc

Lib (Win32):

C:\wxWidgets-2.8.12\lib\vc_lib

Lib (x64):

C:\wxWidgets-2.8.12\lib\vc_amd64_lib\
Best regards,

Martynas
Software Consultant
Neurotechnology
Martynas
Neurotechnology
 
Posts: 2195
Joined: Sep 19, 2008 10:02
Location: Vilnius, Lithuania

Re: SentiSight SDK questions

Postby benjimitsu » Dec 10, 2012 02:53

Thank Martynas,
I figure it out now.
I have another question. Does Sentisight have function to support the measurement of the size of object? For example, I know Sentisight can recognize a nut, but how can I measure its length?
I saw in the advertise video when sentisight detects a calculator. It seems like the software can measure the distance or size of the calculator or something.
Thanks all,
:)
benjimitsu
 
Posts: 3
Joined: Dec 05, 2012 02:05

Re: SentiSight SDK questions

Postby Martynas » Dec 10, 2012 11:46

benjimitsu wrote:Thank Martynas,
I figure it out now.
I have another question. Does Sentisight have function to support the measurement of the size of object? For example, I know Sentisight can recognize a nut, but how can I measure its length?
I saw in the advertise video when sentisight detects a calculator. It seems like the software can measure the distance or size of the calculator or something.
Thanks all,
:)

Hi,

SentiSight does not have such functionality, but you can implement it by yourself. First of all you will need to define the size of the learned object and save it (i.e. in the database). Once the SentiSight engine recognize your learned object it is returning the shape. so you can calculate the size of returned shape and compare it with your saved size.

Note, that the camera should be positioned at the same exact distance from the object during both learning and recognition.
Best regards,

Martynas
Software Consultant
Neurotechnology
Martynas
Neurotechnology
 
Posts: 2195
Joined: Sep 19, 2008 10:02
Location: Vilnius, Lithuania

How to rename of learn model in database [SentiSight]

Postby patiljagdish » Dec 14, 2012 03:45

Hi all,
I want to change the name of learn model present in sentisight database, rather than learning it again.

Admin: merged to "SentiSight SDK questions"
patiljagdish
 
Posts: 15
Joined: Oct 22, 2012 10:33

Re: How to rename of learn model in database [SentiSight]

Postby Martynas » Dec 14, 2012 08:29

patiljagdish wrote:Hi all,
I want to change the name of learn model present in sentisight database, rather than learning it again.

Hello,

the model by itself does not contain the name. The id (name) is specified when the model is added to SentiSight engine. So if you have changed the model name in your database, then add it to the SentiSight engine with the new name.
If you need to change the name of the model in the SentiSight engine, when the model is already added, then you will need to remove this model and add it again with the new name.
Best regards,

Martynas
Software Consultant
Neurotechnology
Martynas
Neurotechnology
 
Posts: 2195
Joined: Sep 19, 2008 10:02
Location: Vilnius, Lithuania

Re: SentiSight SDK questions

Postby theshadowmaker » Dec 16, 2012 08:39

A quick question: where does provided SentiSight sample (particularly C# one) save models created with it?? (whatever.bin). I've found the .db with the database, but can't find the .bin alone. I need them because I'll use learn provided in samples for creating models, and I'll add them (by code) in my app for recognition. I've tried searching in Windows 7 for name.bin and no luck :(

Thx!! :)
theshadowmaker
 
Posts: 1
Joined: Dec 16, 2012 08:27

Re: SentiSight SDK questions

Postby Martynas » Dec 17, 2012 08:27

theshadowmaker wrote:A quick question: where does provided SentiSight sample (particularly C# one) save models created with it?? (whatever.bin). I've found the .db with the database, but can't find the .bin alone. I need them because I'll use learn provided in samples for creating models, and I'll add them (by code) in my app for recognition. I've tried searching in Windows 7 for name.bin and no luck :(

Thx!! :)

Hello,

the SentiSightSample does not store models as separate files. It uses the SQLite database to keep learned models and their ids. If you need to get learned models from this database, then you need implement the functionality in this sample to retrieve the models from it.
Also you can use tutorial "Learning" to learn your models and save them as files.
Best regards,

Martynas
Software Consultant
Neurotechnology
Martynas
Neurotechnology
 
Posts: 2195
Joined: Sep 19, 2008 10:02
Location: Vilnius, Lithuania

Any function or method to display the model recognise? [Se]

Postby patiljagdish » Dec 27, 2012 05:25

Hello,
I want to display the the name of model in console or write in some file which is being recognize.
Is there any function to do so ??

Admin: merged to "SentiSight SDK questions"
patiljagdish
 
Posts: 15
Joined: Oct 22, 2012 10:33

Re: Any function or method to display the model recognise ?

Postby vaidasz » Dec 28, 2012 15:35

patiljagdish wrote:Hello,
I want to display the the name of model in console or write in some file which is being recognize.
Is there any function to do so ??


Hello,

There is no special function for the name. But here is how you could get the name.
When you add model to the engine
Code: Select all
engine.Recognition.AddModel(model, (uint)1);

You specify modelId, which is
"A specified model identifier. Later it will be referred in recognition functions as returned value of best recognized model id. Its value may be numeric, literal or any other type. It must be unique for all models added to recognition
module."
You should store this id somwhere with it's name (for example, in database).
Please check SentiSightSampleCS sample how it is implemented - it uses database as described above.
Vaidas Z
Neurotechnology
vaidasz
Neurotechnology
 
Posts: 559
Joined: Apr 30, 2010 06:57
Location: Vilnius, Lithuania

Re: SentiSight SDK questions

Postby Sergey » Jun 19, 2013 08:24

SentiSight 3.2 for object recognition on PC platform is now available. The new version includes these updates:
  • Faster object recognition when using blob-based algorithm.
  • Color usage mode for accuracy improvement when using blob-based algorithm.
  • Image quality check to improve learning and recognition accuracy by accepting only good quality images or video frames.
SentiSight Embedded 1.1 for Android platform also includes the mentioned improvements.
Sergey
Site Admin
 
Posts: 103
Joined: Oct 30, 2007 11:39

sentisight sdk resolution and reflective object

Postby chani » Jun 05, 2014 07:13

Hello, I have to question regarding to the sentisight sdk.
1. right now I use the trial version, and try to manipulate the c# sample code. I noticed that the resolution is very low. also the app doesnt allow me to upload too big images.
is there a way to change this? is that different in the full version?

2. when I test the accuracy rate of recognition, I find that when I try to recognize a reflective image, the accuracy rate is much lower than in matte objects. before I try to add code to neutralize the brilliance, I wonder if there is already any code in the sentisight sdk that applies it.

thank you very much!
Chani

Admin: merged to "SentiSight SDK questions"
chani
 
Posts: 2
Joined: Jun 05, 2014 06:59

Re: sentisight sdk resolution and reflective object

Postby vaidasz » Jun 06, 2014 13:54

Hello,

chani wrote:Hello, I have to question regarding to the sentisight sdk.
1. right now I use the trial version, and try to manipulate the c# sample code. I noticed that the resolution is very low. also the app doesnt allow me to upload too big images.
is there a way to change this? is that different in the full version?

Please explain what do you mean by saying "resolution is very low". Also explain what do you mean by saying "too big images".
The functionality of trial and non-trial version of the SDK are identical. The only difference is the licensing service.

chani wrote:2. when I test the accuracy rate of recognition, I find that when I try to recognize a reflective image, the accuracy rate is much lower than in matte objects. before I try to add code to neutralize the brilliance, I wonder if there is already any code in the sentisight sdk that applies it.

thank you very much!
Chani

Yes, you are correct. The reflective objects recognition is worse than matte objects.
There is no any functionality to preprocess the images.
Vaidas Z
Neurotechnology
vaidasz
Neurotechnology
 
Posts: 559
Joined: Apr 30, 2010 06:57
Location: Vilnius, Lithuania

Re: sentisight sdk resolution and reflective object

Postby chani » Jun 08, 2014 11:56

thank you for your answer!

in "too big images" I mean that if I try yo upload to the app an image heavier that 3 Mb I receive an error message says "32 bit image can"t process image this large". also I never manage to apload video files unless I have created them using the "learn/recognize from camera" option in the app.

I think I was wrong about the resolution - I have noticed that the images that are created when I use the app are 640X480. but now I relize that it depends on the camera definitions...

However, I have some more questions:

1. I made my own changes in the sample code and i want to execute it to another computer, but the .exe file never works. when I click on it I receive a message that the program has stopped, although when I debug it runs nicely. what could be the problem?

2. is there a maximum number for the scoring? I want to normalize it into "similarity percentage" but I didn"d find any documentation about the scoring method.

3. is there a way to manipulate the frames rate? my camera is supposed to capture 30fps, but when I use the app to capture a shot of video I receive about 100 frames for a 30 seconds shot.

thanks again!
chani
 
Posts: 2
Joined: Jun 05, 2014 06:59

Re: sentisight sdk resolution and reflective object

Postby vaidasz » Jun 09, 2014 16:36

chani wrote:thank you for your answer!

in "too big images" I mean that if I try yo upload to the app an image heavier that 3 Mb I receive an error message says "32 bit image can"t process image this large". also I never manage to apload video files unless I have created them using the "learn/recognize from camera" option in the app.

Send an e-mail with issue description to vaidasz [eta] neurotechnology.com. Send couple of such images as well.
Also provide steps to reproduce the issue.

chani wrote:I think I was wrong about the resolution - I have noticed that the images that are created when I use the app are 640X480. but now I relize that it depends on the camera definitions...

Yes, you are correct. You could choose resolution that camera returns.

chani wrote:1. I made my own changes in the sample code and i want to execute it to another computer, but the .exe file never works. when I click on it I receive a message that the program has stopped, although when I debug it runs nicely. what could be the problem?

It seems you are missing libraries. Are you sure you have all necessary libraries in your SDK start-up path or in system environment variables.
As you may know, our .Net dlls depends on native dlls. For example, Neurotec.dll depends on NCore.dll. And .Net dlls could not be loaded from system environment variables - they either have to be at same directory as your application start-up path is or in GAC.
My recommendation would be to have all (.Net and native) dlls at your application start-up path.
Also make sure you are using correct dlls - if it is 32bits OS, then dll for 32bits OS; if it is 64bits OS, then dll for 64bits OS.

chani wrote:2. is there a maximum number for the scoring? I want to normalize it into "similarity percentage" but I didn"d find any documentation about the scoring method.

No, there is not maximum. The more features are extracted in both enrolled model and the one sent for matching, the more score there could be scored. That means, it is not possible to convert the score to similarity percentage.
The scores in range [1-10] could be treated as:
1. Score 700.000 would be 10;
2. Score 500.000 is about 8;
3. Score 150.000 – 200.000 is about 5;
4. 50.000 is bad and it is something like 1-2.

chani wrote:3. is there a way to manipulate the frames rate? my camera is supposed to capture 30fps, but when I use the app to capture a shot of video I receive about 100 frames for a 30 seconds shot.

thanks again!

It depends on the codec you choose. Try choosing different one.
Vaidas Z
Neurotechnology
vaidasz
Neurotechnology
 
Posts: 559
Joined: Apr 30, 2010 06:57
Location: Vilnius, Lithuania

Detect document type using SentiSight

Postby kirill » Jul 01, 2014 09:28

Hi.
I need to extract data from ID image. First I need to detect what type of document user provided: UK driving license, California driving license and so on. So I"m going to use SentiSight to detect a type of ID.
What is the best approach to learn such type of images: using image mask or using polygon?

ID examples:
http://www.businesscarmanager.co.uk/wp- ... sample.jpg
http://roadwarrior.blogs.pressdemocrat. ... cense3.jpg

Admin: merged to "SentiSight SDK questions"
kirill
 
Posts: 2
Joined: Jul 01, 2014 08:36

Detect ID type using SentiSight

Postby kirill » Jul 01, 2014 11:17

Hi,
I need to detect ID type(UK driving license, California driving license and so on) by image provided by user. What is the best approach to learn SentiSight in this case: using image mask or using polygon?

ID examples:
http://www.businesscarmanager.co.uk/wp- ... sample.jpg
http://roadwarrior.blogs.pressdemocrat. ... cense3.jpg

Thanks in advance.
kirill
 
Posts: 2
Joined: Jul 01, 2014 08:36

Re: Detect document type using SentiSight

Postby Martynas » Jul 02, 2014 09:48

kirill wrote:Hi,
I need to detect ID type(UK driving license, California driving license and so on) by image provided by user. What is the best approach to learn SentiSight in this case: using image mask or using polygon?

ID examples:
http://www.businesscarmanager.co.uk/wp- ... sample.jpg
http://roadwarrior.blogs.pressdemocrat. ... cense3.jpg

Thanks in advance.

Hello,

to be able to say if such documents can be recognized by using a SentiSight, we need to get as many sample images as you can provide. So please send them to support@neurotehcnology.com.
Best regards,

Martynas
Software Consultant
Neurotechnology
Martynas
Neurotechnology
 
Posts: 2195
Joined: Sep 19, 2008 10:02
Location: Vilnius, Lithuania

Re: SentiSight SDK questions

Postby minijo » Aug 08, 2014 16:57

we have recently purchased SentiSight sdk and will run it in Linux. it would be apprecaited if some could advise me of the FULL list of Linux programmes required to opearte the system .
Thanks
minijo
 
Posts: 1
Joined: Jul 07, 2014 17:15

Re: SentiSight SDK questions

Postby Martynas » Aug 11, 2014 07:37

minijo wrote:we have recently purchased SentiSight sdk and will run it in Linux. it would be apprecaited if some could advise me of the FULL list of Linux programmes required to opearte the system .
Thanks

Hi,

the requirements are provided on our website.
Best regards,

Martynas
Software Consultant
Neurotechnology
Martynas
Neurotechnology
 
Posts: 2195
Joined: Sep 19, 2008 10:02
Location: Vilnius, Lithuania

Re: SentiSight SDK questions

Postby Sergey » Mar 04, 2015 11:46

SentiSight 3.3 for object recognition on PC platform is now available. The new version includes SentiSight libraries update to match the Neurotec Biometric 5.1 libraries.

SentiSight Embedded 1.2 for Android platform also includes the mentioned improvements.
Sergey
Site Admin
 
Posts: 103
Joined: Oct 30, 2007 11:39

NPixelFormat

Postby D99000706 » Aug 26, 2015 17:24

The function NImageCreateFromDataEx takes as first argument a type NPixelFormat_ with the underscore.
I see in the documentation there are useful types like:

NPixelFormat f = NPF_RGB_8U;

but these are of type NPixelFormat without the underscore.
How can I convert this useful value NPF_RGB_8U to the proper type?
D99000706
 
Posts: 5
Joined: Aug 26, 2015 12:51

Re: SentiSight SDK questions

Postby Martynas » Aug 27, 2015 07:25

D99000706 wrote:The function NImageCreateFromDataEx takes as first argument a type NPixelFormat_ with the underscore.
I see in the documentation there are useful types like:

NPixelFormat f = NPF_RGB_8U;

but these are of type NPixelFormat without the underscore.
How can I convert this useful value NPF_RGB_8U to the proper type?

Hi,

use the NPixelFormat without the underscore.
Best regards,

Martynas
Software Consultant
Neurotechnology
Martynas
Neurotechnology
 
Posts: 2195
Joined: Sep 19, 2008 10:02
Location: Vilnius, Lithuania

Re: SentiSight SDK questions

Postby D99000706 » Aug 27, 2015 12:04

Martynas wrote:Hi,

use the NPixelFormat without the underscore.

This code causes errors:

NPixelFormat pixFormat = NPF_GRAYSCALE_8U; // this causes error

NResult result = NImageCreateFromDataEx(pixFormat, image.size().width, image.size().height, 0, image.step1(), image.data, image.size().area(), 0, &hnimage);


1 error C2664: "Neurotec::NResult Neurotec::Images::NImageCreateFromDataEx(Neurotec::Images::NPixelFormat_,Neurotec::NUInt,Neurotec::NUInt,Neurotec::NSizeType,Neurotec::NSizeType,const void *,Neurotec::NSizeType,Neurotec::NUInt,Neurotec::Images::HNImage *)" : cannot convert argument 1 from "Neurotec::Images::NPixelFormat" to "Neurotec::Images::NPixelFormat_"

2 IntelliSense: no suitable conversion function from "Neurotec::Images::NPixelFormat" to "Neurotec::Images::NPixelFormat_" exists

Any suggestions??
D99000706
 
Posts: 5
Joined: Aug 26, 2015 12:51

Re: SentiSight SDK questions

Postby Martynas » Aug 28, 2015 07:42

D99000706 wrote:This code causes errors:

NPixelFormat pixFormat = NPF_GRAYSCALE_8U; // this causes error

NResult result = NImageCreateFromDataEx(pixFormat, image.size().width, image.size().height, 0, image.step1(), image.data, image.size().area(), 0, &hnimage);


1 error C2664: "Neurotec::NResult Neurotec::Images::NImageCreateFromDataEx(Neurotec::Images::NPixelFormat_,Neurotec::NUInt,Neurotec::NUInt,Neurotec::NSizeType,Neurotec::NSizeType,const void *,Neurotec::NSizeType,Neurotec::NUInt,Neurotec::Images::HNImage *)" : cannot convert argument 1 from "Neurotec::Images::NPixelFormat" to "Neurotec::Images::NPixelFormat_"

2 IntelliSense: no suitable conversion function from "Neurotec::Images::NPixelFormat" to "Neurotec::Images::NPixelFormat_" exists

Any suggestions??

Hi,

if you are using a C++, then cast the correct format


Code: Select all
NPixelFormat pixFormat = NPF_GRAYSCALE_8U;
NResult result = NImageCreateFromDataEx((NPixelFormat_)pixFormat, image.size().width, image.size().height, 0, image.step1(), image.data, image.size().area(), 0, &hnimage);
Best regards,

Martynas
Software Consultant
Neurotechnology
Martynas
Neurotechnology
 
Posts: 2195
Joined: Sep 19, 2008 10:02
Location: Vilnius, Lithuania

Re: SentiSight SDK questions

Postby D99000706 » Aug 28, 2015 13:45

Martynas wrote:Hi,

if you are using a C++, then cast the correct format


Code: Select all
NPixelFormat pixFormat = NPF_GRAYSCALE_8U;
NResult result = NImageCreateFromDataEx((NPixelFormat_)pixFormat, image.size().width, image.size().height, 0, image.step1(), image.data, image.size().area(), 0, &hnimage);

We are using C++ and tried the casting as you suggest:

NPixelFormat pixFormat = NPF_RGB_8U; // can't use this?????
//NUInt pixFormat = 0x00303003; // rgb 8U value from .hpp file
HNImage hnimage;

NResult result = NImageCreateFromDataEx(
(NPixelFormat_)pixFormat,
.....)

and it still gives errors:

error C2440: 'type cast' : cannot convert from 'Neurotec::Images::NPixelFormat' to 'Neurotec::Images::NPixelFormat_'

If we use the NUint that is commented out it works okay.
D99000706
 
Posts: 5
Joined: Aug 26, 2015 12:51

Re: SentiSight SDK questions

Postby Martynas » Sep 02, 2015 12:01

D99000706 wrote:We are using C++ and tried the casting as you suggest:

NPixelFormat pixFormat = NPF_RGB_8U; // can't use this?????
//NUInt pixFormat = 0x00303003; // rgb 8U value from .hpp file
HNImage hnimage;

NResult result = NImageCreateFromDataEx(
(NPixelFormat_)pixFormat,
.....)

and it still gives errors:

error C2440: 'type cast' : cannot convert from 'Neurotec::Images::NPixelFormat' to 'Neurotec::Images::NPixelFormat_'

If we use the NUint that is commented out it works okay.

Hi,

please create a small test application and send it to support@neurotechnology.com with the issue description.
Best regards,

Martynas
Software Consultant
Neurotechnology
Martynas
Neurotechnology
 
Posts: 2195
Joined: Sep 19, 2008 10:02
Location: Vilnius, Lithuania

SentiSight Zonal Recognition

Postby ssimpson » Oct 27, 2015 17:51

Is it possible to have the SentiSight recognition engine only look for objects in a particular region / zone of an image? Or, will the image need to be cropped in a separate process?
ssimpson
 
Posts: 3
Joined: Oct 27, 2015 17:46

Re: SentiSight SDK questions

Postby Martynas » Oct 28, 2015 08:02

ssimpson wrote:Is it possible to have the SentiSight recognition engine only look for objects in a particular region / zone of an image? Or, will the image need to be cropped in a separate process?

Hello,

no, there is no possibility to define that the object would be recognized only in specified region. You need to crop the image before recognition is done.
Best regards,

Martynas
Software Consultant
Neurotechnology
Martynas
Neurotechnology
 
Posts: 2195
Joined: Sep 19, 2008 10:02
Location: Vilnius, Lithuania

LearnMode Question

Postby ssimpson » Oct 30, 2015 13:01

The documentation for the demo app says:
Code: Select all
Learning/Recognition
1. Learning mode
LowProfile
    Fastest and has smallest template size, suitable in most situations.
HighProfile
    Additional not rotation invariant information added, improves recognition quality for not rotated objects (has no impact on
    rotated object recognition). About 5%-10% slower and has about 20%-60% bigger template size than for low profile learning mode.
HighProfileEx
    Additional information is added, improves recognition for all types of objects. About 50%-100% slower and has about
    5%-10% bigger template size compared with low profile.


However, the javadoc and C API documentation within the SDK says:
Code: Select all
Enum Constant Summary
HIGH_PROFILE
          Fastest and has the smallest template size, suitable in most situations.
HIGH_PROFILE_EX
          Adds additional information, improves recognition for all types of objects.
LOW_PROFILE
          Adds additional not rotation invariant information, improves recognition quality for not rotated objects (has no impact on rotated object recognition).


Which is correct?
ssimpson
 
Posts: 3
Joined: Oct 27, 2015 17:46

Java LearnMode Runtime Exception

Postby ssimpson » Nov 02, 2015 19:46

When trying to change the default learn mode to SELrnMode.HIGH_PROFILE_EX, I receive the following exception at run-time:

Code: Select all
     [java] com.neurotec.lang.NIllegalArgumentException: Value pValue points to is not in property standard values listProblem: Value pValue points to is not in property standard values list
     [java]    at com.neurotec.lang.RuntimeErrorCreator.create(RuntimeErrorCreator.java:22)
     [java]    at com.neurotec.lang.ChainedErrorCreator.create(ChainedErrorCreator.java:22)
     [java]    at com.neurotec.lang.NError.get(NError.java:174)
     [java]    at com.neurotec.lang.NError.get(NError.java:191)
     [java]    at com.neurotec.lang.NError.getLast(NError.java:205)
     [java]    at com.neurotec.lang.NResult.checkUnchecked(NResult.java:245)
     [java]    at com.neurotec.lang.NResult.raiseError(NResult.java:333)
     [java]    at com.neurotec.lang.NResult.check(NResult.java:234)
     [java]    at com.neurotec.lang.NObject.setProperty(NObject.java:923)
     [java]    at com.neurotec.sentisight.SEEngine$SELearning.setMode(SEEngine.java:851)


Also, when calling getMode() to determine the default configuration, I receive the following exception at run-time:

Code: Select all
     [java] java.lang.RuntimeException: java.lang.reflect.InvocationTargetException
     [java]    at com.neurotec.lang.NValue.toClassInternal(NValue.java:1029)
     [java]    at com.neurotec.lang.NValue.toClass(NValue.java:1193)
     [java]    at com.neurotec.lang.NValue.toClass(NValue.java:1210)
     [java]    at com.neurotec.lang.NObject.getProperty(NObject.java:836)
     [java]    at com.neurotec.sentisight.SEEngine$SELearning.getMode(SEEngine.java:843)
     [java] Problem: java.lang.reflect.InvocationTargetException   at <snip>.Learning.main(Unknown Source)
     [java]
     [java] Caused by: java.lang.reflect.InvocationTargetException
     [java]    at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
     [java]    at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:57)
     [java]    at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
     [java]    at java.lang.reflect.Method.invoke(Method.java:606)
     [java]    at com.neurotec.lang.NValue.toClassInternal(NValue.java:1022)
     [java]    ... 5 more
     [java] Caused by: java.lang.IllegalArgumentException: value is an unknown NEnum value
     [java]    at com.neurotec.lang.NTypes.getEnum(NTypes.java:908)
     [java]    at com.neurotec.sentisight.SELrnMode.get(SELrnMode.java:90)
     [java]    ... 10 more


The getMode() method seems to work, if the SELrnMode.LOW_PROFILE or SELrnMode.HIGH_PROFILE mode is actually set via the API.
ssimpson
 
Posts: 3
Joined: Oct 27, 2015 17:46

Re: LearnMode Question

Postby Martynas » Nov 03, 2015 13:36

ssimpson wrote:The documentation for the demo app says:
Code: Select all
Learning/Recognition
1. Learning mode
LowProfile
    Fastest and has smallest template size, suitable in most situations.
HighProfile
    Additional not rotation invariant information added, improves recognition quality for not rotated objects (has no impact on
    rotated object recognition). About 5%-10% slower and has about 20%-60% bigger template size than for low profile learning mode.
HighProfileEx
    Additional information is added, improves recognition for all types of objects. About 50%-100% slower and has about
    5%-10% bigger template size compared with low profile.


However, the javadoc and C API documentation within the SDK says:
Code: Select all
Enum Constant Summary
HIGH_PROFILE
          Fastest and has the smallest template size, suitable in most situations.
HIGH_PROFILE_EX
          Adds additional information, improves recognition for all types of objects.
LOW_PROFILE
          Adds additional not rotation invariant information, improves recognition quality for not rotated objects (has no impact on rotated object recognition).


Which is correct?


The correct one is the upper one. By the way the recent version of the SDK has merged LowProfile and HighProfile. So use the HighProfile instead of the LowProfile.
Best regards,

Martynas
Software Consultant
Neurotechnology
Martynas
Neurotechnology
 
Posts: 2195
Joined: Sep 19, 2008 10:02
Location: Vilnius, Lithuania

Re: Java LearnMode Runtime Exception

Postby Martynas » Nov 03, 2015 13:37

ssimpson wrote:When trying to change the default learn mode to SELrnMode.HIGH_PROFILE_EX, I receive the following exception at run-time:

Code: Select all
     [java] com.neurotec.lang.NIllegalArgumentException: Value pValue points to is not in property standard values listProblem: Value pValue points to is not in property standard values list
     [java]    at com.neurotec.lang.RuntimeErrorCreator.create(RuntimeErrorCreator.java:22)
     [java]    at com.neurotec.lang.ChainedErrorCreator.create(ChainedErrorCreator.java:22)
     [java]    at com.neurotec.lang.NError.get(NError.java:174)
     [java]    at com.neurotec.lang.NError.get(NError.java:191)
     [java]    at com.neurotec.lang.NError.getLast(NError.java:205)
     [java]    at com.neurotec.lang.NResult.checkUnchecked(NResult.java:245)
     [java]    at com.neurotec.lang.NResult.raiseError(NResult.java:333)
     [java]    at com.neurotec.lang.NResult.check(NResult.java:234)
     [java]    at com.neurotec.lang.NObject.setProperty(NObject.java:923)
     [java]    at com.neurotec.sentisight.SEEngine$SELearning.setMode(SEEngine.java:851)


Also, when calling getMode() to determine the default configuration, I receive the following exception at run-time:

Code: Select all
     [java] java.lang.RuntimeException: java.lang.reflect.InvocationTargetException
     [java]    at com.neurotec.lang.NValue.toClassInternal(NValue.java:1029)
     [java]    at com.neurotec.lang.NValue.toClass(NValue.java:1193)
     [java]    at com.neurotec.lang.NValue.toClass(NValue.java:1210)
     [java]    at com.neurotec.lang.NObject.getProperty(NObject.java:836)
     [java]    at com.neurotec.sentisight.SEEngine$SELearning.getMode(SEEngine.java:843)
     [java] Problem: java.lang.reflect.InvocationTargetException   at <snip>.Learning.main(Unknown Source)
     [java]
     [java] Caused by: java.lang.reflect.InvocationTargetException
     [java]    at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
     [java]    at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:57)
     [java]    at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
     [java]    at java.lang.reflect.Method.invoke(Method.java:606)
     [java]    at com.neurotec.lang.NValue.toClassInternal(NValue.java:1022)
     [java]    ... 5 more
     [java] Caused by: java.lang.IllegalArgumentException: value is an unknown NEnum value
     [java]    at com.neurotec.lang.NTypes.getEnum(NTypes.java:908)
     [java]    at com.neurotec.sentisight.SELrnMode.get(SELrnMode.java:90)
     [java]    ... 10 more


The getMode() method seems to work, if the SELrnMode.LOW_PROFILE or SELrnMode.HIGH_PROFILE mode is actually set via the API.


We will check this on our side.
Best regards,

Martynas
Software Consultant
Neurotechnology
Martynas
Neurotechnology
 
Posts: 2195
Joined: Sep 19, 2008 10:02
Location: Vilnius, Lithuania

Shape feature type in SentiSight Algorithm Demo

Postby Irina chrome » Nov 11, 2015 08:19

Hi, everyone!

I downloaded a SentiSight Algorithm Demo to try this algorithm.
In shape feature type I cannot save a model.
Data Source: image file.
I loaded an image, selected object bounds and tried to save a model.
Then I saw a message:"Model is empty.Please add at least one shape to model."

What did I do wrong?

Admin: merged to "SentiSight SDK questions"
Irina chrome
 
Posts: 1
Joined: Nov 11, 2015 07:52

Re: Shape feature type in SentiSight Algorithm Demo

Postby Martynas » Nov 11, 2015 13:23

Irina chrome wrote:Hi, everyone!

I downloaded a SentiSight Algorithm Demo to try this algorithm.
In shape feature type I cannot save a model.
Data Source: image file.
I loaded an image, selected object bounds and tried to save a model.
Then I saw a message:"Model is empty.Please add at least one shape to model."

What did I do wrong?

Hello,

this is a bug in the demo application. I have reported it to our developers. I will inform you about the fix as soon as it will be available.
Best regards,

Martynas
Software Consultant
Neurotechnology
Martynas
Neurotechnology
 
Posts: 2195
Joined: Sep 19, 2008 10:02
Location: Vilnius, Lithuania

Re: SentiSight SDK questions

Postby Martynas » Jul 04, 2016 10:13

millahjovich wrote:i had been a task to do a project by my supervisor to monitor and track motion object using SentiSight SDK


Hello,

SentiSight SDK is for devlopers who want to use computer vision-based object recognition in their application. It is not suitable for motion detection, but rather suitable to find if the image contains the learned object. SentiVeillance SDK has mode for motion detection.

My supervisor just give me the SentiSight folder and tell me to work on it. But my problem is after i search the folder i can"t open the application to sentisight. So i want to know what is the application name to open the SentiSight application? thanks a million. :D


SDK provide libraries, which allows to use the SDKs functionality in your application. Please take a look at sample/tutorials, to find how to use functionality of the SDK.
The SDK includes already built samples, i.e. SentiSightSampleWX.exe, which can be run to see how the SentiSight functionality can be used. Other samples/tutorials are provided as source codes in "samples", "tutorials" folders of the SDK.
Best regards,

Martynas
Software Consultant
Neurotechnology
Martynas
Neurotechnology
 
Posts: 2195
Joined: Sep 19, 2008 10:02
Location: Vilnius, Lithuania

model save question

Postby D99000706 » Sep 22, 2016 14:47

We are using the SentiSight 3.3 SDK to train and recognize images.
We thought it was possible to separate the training and recognition into 2 programs by saving the model to disk. However, we cannot find how to do this in the documentation.
We see a function in the docs called SEModel.Load(System.Byte[]) but cannot find an associated save or write function. Is this possible? Can you guide us on how to do this?

Thanks.
D99000706
 
Posts: 5
Joined: Aug 26, 2015 12:51

save a model to disk

Postby D99000706 » Sep 22, 2016 14:49

We are using the SentiSight 3.3 SDK to train and recognize images.
We thought it was possible to separate the training and recognition into 2 programs by saving the model to disk. However, we cannot find how to do this in the documentation.
We see a function in the docs called SEModel.Load(System.Byte[]) but cannot find an associated save or write function. Is this possible? Can you guide us on how to do this?

Thanks.
D99000706
 
Posts: 5
Joined: Aug 26, 2015 12:51

Re: SentiSight SDK questions

Postby Martynas » Sep 22, 2016 14:55

Hi,

SEModel.Save().ToArray() will return the byte array, which you can save as file by using standard programming tools. By the way model saving as a file is shown in tutorial "Learning".
Best regards,

Martynas
Software Consultant
Neurotechnology
Martynas
Neurotechnology
 
Posts: 2195
Joined: Sep 19, 2008 10:02
Location: Vilnius, Lithuania

SentiSight SDK questions

Postby rebootMet » Aug 08, 2017 13:18

Thanks for the info on the SDK, I somehow missed that while looking around the site. Iualso nderstand the the DriverName a little more now.

The application seemed to crash when close was called every time I used DI670NT.dll when DI672NT.dll was in the INI file. Any ideas why this might happen?

It has been a few years since I worked much will with DLLs, but from what I remember the DLL gets a separate address space for each process that loads it so why do you need multiple copies of the same DLL?

Thanks
Mark
rebootMet
 
Posts: 1
Joined: May 15, 2017 13:09
Location: Россия

PreviousNext

Return to Public

Who is online

Users browsing this forum: No registered users and 2 guests