A novel recursive sub-image histogram equalization (RSIHE) is developed to overcome the drawbacks of generic histogram equalization (HE) for gray scale images. Compared to some of the conventional HE methods, such as bi-histogram equalization and recursive mean-separate histogram equalization, the proposed RSIHE method yields better image compensation. The scanning electron microscope images are used as test images to evaluate the efficiency of the developed algorithm. The algorithm is implemented in software with a frame grabber card, forming the front-end video capture element.