Hello the community,
Actually, I have to manipulate an image of 5120 x 3840. I need to divide by 5 the image, save it and then write 0 for each pixel.
I just use :
Image = Image * 0.2;
...
Image = Image * 0 ;
I don´t how it codes. Is it more efficient to use foreach ?
Thank you,
you could use accumulate(), and only multiply once:
Mat dst(image.size(), CV_32S, Scalar(0));
accumulate(image, dst);
accumulate(image, dst);
accumulate(image, dst);
accumulate(image, dst);
accumulate(image, dst);
dst *= 0.2;
// later:
dst.setTo(0);
It should be assumed that an operator is always faster than a loop, as OpenCV can use its own, optimized code. Similarly, an assignment is always faster than multiplication and anything multiplied by 0 is 0, so one should use simple zeroing
Asked: 2020-02-24 02:52:20 -0600
Seen: 266 times
Last updated: Feb 24 '20
just curious, why do you need to multiply with 0 ?
what is the context of your code ? what is it for ?
I get 5 images from a Gige camera (as fast as possible). Then, I sum those images (8 bit in grayscale) in a 32 bites image grayscale, I apply a median filter, and then I divide by 5. At the end of the process, I need to multiply the 32 bit image by 0, and then sum 5 new images. Am I clear ?
may be you can use a IIr filter