A specified color block/cube detection with the camera. The codes are implemented with OpenCV 2.4.13 in visual studio 2013.
This is the visual part for the Dobot robot to grab the block/cube in specified color and detect the black slot (where the block should be put on).
- preprocess the image get from the camera (thresholding, smoothing, and etc.)
- detect the contours and find the minimum bounding boxes.
- restrick the shape, area and perimeter for better detection result.
- select the most suitable block for grab.