Towards Fast Edge Detection Approach for Industrial Products

Edge detection is a critical component of many elements of image processing applications and computer visualization, such as image segmentation, detection of imperfections in industrial products, medical image processing, and object identification. The edge detection technique seeks to accelerate image analysis by limiting the quantity of information processed. The object’s dimensions are a major component in manufacturing, engineering and other crucial fields, therefore the less time this takes, the more efficient the whole process becomes. This study details the use of unique image processing technology to advance the automated quantifying of an industrial object’s dimensions. Firstly, an image improvement approach founded on fuzzy entropy was devised to clarify the original deteriorated image. Secondly, the edges of the object were detected via the method of quick fuzzy edge detection. Finally, using Freeman Chain Codes (FCC), the essential corner locations were determined. Thus, the edges and corners of any given object image are accurately detected. This strategy is uncomplicated to implement, can efficiently identify size, shape, and any defects, and helps to determine acceptance or rejection of the final product. The results of the corresponding experiment indicate that the proposed method has a much lower computational cost in general compared to the related works.