Maik Angermann's profile - activity

Hi,

does anyone have a good tesseract example or tutorial. I dont get it how to use the paramemters? I tried:

`string test;

Ptr<text::OCRTesseract> ocr = text::OCRTesseract::create();

ocr->run(getSrcImg(), test); `

the API at http://docs.opencv.org/master/d7/ddc/... doesn't help me since I am not familiar with the functions and it's inputs.

Thanks, Are you refering to a linux or windows system? I tried to include libtif but it seems to be very old fashioned. I have read that libitf is included in windows systems since a while.

Kbarni, Well this seem to be a long road to the finish. Anyway, I managed to convert the the TIFFS in bmp but stil the problem is that OpenCV can't handle it because of the image size.

So how can I open huge images even if I dont use TIFF files.

Hi everone.

I build an OPENCV3.0 project in order to read big geo tiff satellite images. I was able to link gdal lib to OCV but still have problems opening tiffs. I am not sure if the there are 2 issues. One might be the coding of the tif and the other might be the file size. It seems that I can not open file sizes bigger than 300 Mbyte. My goal is it to open up geo tiffs (basically satelite images) up to 800 Mbyte. Either I get an out of memory failure or the program fails with exception. imgread_LOAD_GDAL will result in "Exception at 0x7633c42d, code: 0xe06d7363: C++ exception, flags=0x1 (execution cannot be continued) at e:\docs\opencv3\sources\modules\core\src\system.cpp:656"

I use QT creator with VS2010 compiler.

Does anyone have any experiences with that and might help me. Thanks

You can set a ROI (region of interest) to mark the area in the source image to be copied. Set the roi in the destination image and copy the image. Once you have set a roi, the certain part of the image will be treated as whole Image.

So your code could be something like: [code] *IplImage img1, img2; cvSetImageROI(img2, cvRect(0,320,640,320); cvSetImageROI(img1, cvRect(0,320,640,320); cvcopy(img1,img2); cvReleaseROI(img2); cvReleaseROI(img1);

[/code] You need to tell the source image the roi in order to copy the part you want to copy and you need to tell the destination a roi in order to copy the certain part of the image into the specific half of your image

Hope that helps

Maik

Hi there, probably an old question, but I didn't find a good solution yet.

I am using a GigE Camera streaming at 20 FPS. I do some line detection with Canny Edge and Hough lines. Besides this I want to store the Images from the camera. I also save some data like timestamp and GPS Position of the Camera which are also saved with the pictures.

The point is, if I save every Image, the framerate goes down to 8 FPS or even less. Using JPEGS seems to be the quickest possibility to save the Images. Is there another way the save Images as Video quickly.

Later on I want to replay the Images/ Video and stream their data to a program. Using the OpenCV Videowriter function only accepts non variable framerates.

So is there I possibility to write Images real quick with variable framerates. I could also use the timestamp for replay purposes.

Any other suggestions / experiences MAik

after struggling a lot I found out something. Using an 12bit Bayern Pattern Image will be too dark though. So if one converts the Bayern Pattern into a 16 Bit IplImage the colors will be displayed with a wrong value. What needs to be done after the cvconvert (Bayer2BGR) is to scale the color value (which is still based on the 12 Bit Image) to the 16 Bit Image. The IplImage now will be displayed with the correct Image colors

Cheers Maik

Hi folks, I'd like to build a filestream from an object that has several members. Generally speaking the object contains an IplImage, time and the GPS Coordinates of the Camera. The Image is captured by a Camera. So I have a big while loop with every step updating the Image an it's data. My goal now is to save not just the Image but also it's data into a file and later on to replay all this data. I was looking for keywords like serialization etc. but I am not quite sure if serialization of an Image is even possible. Has anyone a good advice or maybe made some experience in this area

Cheers Maik

Hi,

I developed some software to read out the images from a GigE Camera using its own SDK. Ramdomly I found the CV_CAP_GIGANETIX of the highgui. Has anybody experience how to use the CV_CAP_GIGANETIX for GigE Cameras directly in OpenCv

Kind Regards

Has anyone already successfully used GigE Vision Cams together with OpenCV. I use OpenCV for the Edge detection. Now we have a photonfocus GIGE MV1 D1312 Color Model. The cam is shipped with the ebus SDK of Pleora with several samples. Because the use of the bayer pattern I have to use the Plearo own Image Format PVImage, which is retrieved from the buffer, then I convert the specific Image into an IPLImage. Even though I use the colored BG8 PVImage the retrieved IplImage is a Greyscaled one. So I have to use the cvconvert function to get a 3 Channel BGR Image. It seems that all this conversions and copies from buffers in images reduce the Speed of the cameras Image aquisisation. Has anyone a better walkthrough getting the Images as IPLImage

Regards Maik

Hi,

I am facing kind of the same problem. I found a sample converting PV Image from ebus sdk (which uses several bayer Patterns) to IplImage. But if you use the Switch case structure like in the sample, a BG8 Color Image will be changed into a monochrome Image. If one uses nchannels=3 the program crashes. I Need to read out the buffer as PVImage the to convert it into an IplImage, after this I could use the cvConvert function. I know that the depth of a Bayer BG8 is 8 bit unsigned but I dont know how many channels to use for the IPLImage then. Setting channel to 1 seems the only possibility but also changes the colored Image into a Grey one. cheers Maik