Performance evaluation of CPU-GPU and CPU-only algorithms for detecting defective tablets through morphological imaging techniques

Pharmaceutical industries which are intended for the packaging of different tablets in a strip of blister need to make sure that the tablets are free from defects before letting them go into the packing box. The purpose of this project is to speed-up the system process via implementing the image processing algorithm on GPU. Morphological and mathematical operations have been implemented on both GPU and CPU. We found out 44.94x speed improvement in CPU-GPU algorithm as compare to CPU-only version. We then further optimized the GPU algorithm to reduce global memory traffic and noticed 9.76x additional improvement in speed of processing the image. The complete GPU-CPU algorithm runs 54.9x faster than a CPU-only algorithm.