BeS's profile - activity

overview network karma followed questions activity

2013-09-19 11:57:04 -0600	commented question	Background removal with changing light You can look in the direction of adaptive background subtraction method. For example can see my implementation of the algorithm VIBE: https://github.com/BelBES/VIBE, this algorithm is quite fast and stable enough to minor noise.
2013-06-28 11:31:07 -0600	answered a question	Feature Detection Interface vs Specific Classes Can you write initModule_nonfree () before calling siftDetector = cv :: FeatureDetector :: create ("SIFT")? Probably cv :: FeatureDetector :: create ("SIFT") returns you to the null pointer. This occurs because the algorithm SIFT and SURF are outside module features2d.
2013-06-27 13:25:10 -0600	answered a question	Detect Chessboard on Images/Scans from Books and Newspapers Hi! You can use the Calibration module for detecting the chessboard in the image. For this function, there findChessboardCorners(...) upd: for your image, you can adapt the example of opencv / squares. Here is a sample output of the adapted code: Best regards, BeS
2013-06-26 12:56:14 -0600	commented question	How to detecting multiple objects Hi! There are many different methods for detecting objects based on Feature Matching, Contours Matching, Machine Learning etc. Selection of a particular method and adapting it to work with multiple instances of the class depends on the application and the type of objects. Please describe the problem in more detail. If there are examples of images of scenes with objects, add them to the description. Best regards, BeS
2013-06-21 08:10:40 -0600	commented question	Problem for compiling OpenCV with OpenNI2 I am not sure whether OpenCV is OpenNI2 compatible. You may want to use OpenNI1.
2013-06-21 03:17:22 -0600	received badge	● Editor (source)
2013-06-21 03:14:59 -0600	answered a question	How to work on a non rectangular region for feature matching using SURF or SIFT Hi! Yes, you can use a mask to select the desired region in SURF/SIFT. That is, to create an image of the same size as the original and mark it non-zero values for the pixels that need isoplzovat SIFT / SURF and zero all other pixels. SIFT SURF Best regards, BeS
2013-06-20 15:47:58 -0600	received badge	● Supporter (source)
2013-06-20 15:43:13 -0600	commented question	Best method for detecting multiple white ovals on black background You could show the example images?
2013-06-20 07:24:18 -0600	received badge	● Nice Answer (source)
2013-06-19 16:59:23 -0600	answered a question	Count Objects (Red Caps in Coca Cola case) Hi! I used the Hough transform to find circles. Then, remove the circle having a lot of overlap with other circles. Got a baseline for a solution to your problem. Probably you should try this approach. I used Hough with the following parameters: `circles = cv2.HoughCircles(img, cv2.HOUGH_GRADIENT, 1.35, 5, np.array([]), 108, 33, 1, 30)` Best regards, BeS
2013-06-18 03:44:29 -0600	received badge	● Nice Answer (source)
2013-06-18 02:12:58 -0600	received badge	● Teacher (source)
2013-06-17 14:43:43 -0600	answered a question	Background subtraction Hello! You can try to use the difference between two adjacent frames or sufficiently fast adaptive algorithm VIBE(i apologize for the sloppy code, it was not possible to refactor) Best regards, BeS