OpenCV data types errors

Question

I am writing an application in C++ using OpenCV library. I am trying to do simple operations with matrices and on some data types operations output errors.

For example:

reduce(img, img, 0, CV_REDUCE_SUM, CV_8U); //does not work if img contains CV_8U
reduce(img, img, 0, CV_REDUCE_SUM, CV_64F); //does work if img contains CV_8U

And if try simple per element division:

Mat A = B / C; //generates data type assertion error if B and C are CV_32F

But does work if B and C are CV_64F.

I would really appreciate your help. Don't want to use CV_64F as it slows down my program significantly.

I've posted an answer for reduction operations. Division works perfectly with CV_32F types. Please provide a minimal reproducible example to see where is the problem. — Miki
– Miki, Commented Apr 27, 2016 at 12:14

Miki · Accepted Answer · 2016-04-27 12:13:42Z

1

Reduction operations needs a suitable destination type, otherwise they will fail in case they might produce inconsistent result due to saturation.

If you are summing (CV_REDUCE_SUM) CV_8U values, you need a destination type big enough to contain the sum, which is either: CV_32S, CV_32F or CV_64F.

You can check this, as well as other combinations in the source code.

answered Apr 27, 2016 at 12:13

Miki

42k13 gold badges131 silver badges219 bronze badges

Sign up to request clarification or add additional context in comments.

Collectives™ on Stack Overflow

OpenCV data types errors

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related