In my application, I have a small piece of code which will merge two images.
The merging of images are with the guidance of a Mask layer.
The platform is opencv on EMGU for C#.
The algorithm I have used is very simple and hence the memory requirement is astronomical.
I am looking for an algorithm implemented, which take very less memory for this merging operation.
The existing algorithm looks like this.
* 3 inputs. (1) Layer1 RBG 24-bit image (2) Layer2 RBG 24-bit image (3) MaskLayer Gray 8-bit image
* convert all the three inputs to float data type image
* divide the mask layer by 256 (0.00392 in float)
* multiply the layer1 and layer2 with mask layer
* add the resulted layer1 and layer2 to form the final result, the merged image.
The above algorithm is unacceptable because of this huge memory requirement since the operations are in float data-type.
Please bid here, I will share the existing code.
If you don't have experience in image processing & opencv, kindly don't bid.