CUVI version 0.6 is under development. One of its new features is the histogram equalization of images.

Histogram Equalization is an image processing technique for contrast enhancement of low contrast images. Our implementation is 3 to 6 times faster than the one offered by OpenCV 2.2, with same output. The speed depends on image size as well as image intensity values. Currently, only 8 bit greyscale images are supported, and we hope to add support for other image types in future releases of CUVI.

The CUVI function for histogram equalization is:

CuviStatus histEq(CuviImage* src, CuviImage* dst)

It couldn’t get simpler. The input (src) is an 8 bit greyscale image and output (dst) is the image with corrected contrast.

Limitations:

  • Limited to 8 bit grayscale images.
  • Works only on devices with Compute Capability 1.2 or above.

Sample Image: