April 2020
Intermediate to advanced
438 pages
12h 2m
English
The cv2.dnn.blobFromImage() function was used to convert the image from the OpenCV format to Caffe blob format. First, the pixel values were normalized so that they were within the (0,1) range. Then the size of the resizing target image was provided. Finally, the mean value was subtracted. No color-channel swapping was required since both OpenCV and Caffe use the same color format of BGR.
The cv2.minMaxLoc() function was used to compute the local maxima of the confidence map.
Read now
Unlock full access