This forum is disabled, please visit https://forum.opencv.org
 | 1 | initial version |
All images are converted to gray first because ORB detector and descriptor work with grayscale intensity to detect the features and compare with the neighbouring pixels.
Reducing number of features will speed up the detection but not necessarily improve the accuracy.
You have detections for an image of a black screen because the sensitivity of your detector is high. You need to increase the detector's threshold to enhance the detection by reducing the noise.
