A Histogram Modification Method for Improving Visual Inspection Accuracy

TL;DR: This paper makes an effort for optimization of remote control visual inspection by making the identical histogram of reference image and input image and as a result of the histogram equalization and specification algorithm, this system has been developed only using PC, CCD Camera and Visual C++ for universal workplace.

Abstract: Recently the remote controls of many types are used in the electrical component. The goal of this paper is the improvement of accuracy of remote control appearance visual inspection by using histogram equalization and specification method. The histogram is composed of horizontal axis of image intensity value and vertical axis of pixel number in image. And the histogram of reference image and input image are different because of minutely lighting distinction. The minutely lighting distinction is the main reason of visual inspection error in many cases. Also most of appearance visual inspection, appearance problems are detected by comparing of reference image and input image. Therefore we made an effort for optimization of remote control visual inspection by making the identical histogram of reference image and input image. As a result of the histogram equalization and specification algorithm, we could increase the exactness of remote control visual inspection and prevent inspection error from physical and spirit condition of human. Specifically, if the proposed algorithm is not used, the inspection error rate of 7.5[%] is showed. In case of proposed method, we could see the inspection error rate of 1.8[%]. Also this system has been developed only using PC, CCD Camera and Visual C++ for universal workplace.