Online Defects Detection System For Print Products Based On VC++

A variety of defects will turn out on the surface of the printing products inevitably, which include color distortion, ink splash pollution, black spots, fuzzy text, wrinkling, abrasions and dislocation, etc. During the processing of the high-speed printing, the quality of the printing must be available by the operator. Some measures should be taken to ensure the success of production progress and improve the quality of printing products.This dissertation carries out a detailed research and testing on image processing algorithm of the on-line defect detection .Firstly, a gray proessing method of true color image which generate color palette is proposed after comparing two different methods, according to the needs of high efficency and speed.Secondly, this paper present a fast algorithm based on connected components labeling using c language, the algorithm implement encoding through a successive scan,then proceed a label and merge the conflict objects by analyzing the connectivity in the above row. Meanwhile, the objects description parameters are acquired such as area, center of figure. The algorithm use the method of run-lists and dynamic array, perform just single scan, needn't to build equivalences table and to unite equivalent labels ,in this way, resolve label redundancies in conventional algorithms. Experiments show that it could correctly detect defect region. This algorithm can count the trait of printing effectively.Finally, a distributed detection system scheme based on Server/Client mode and an image data transmission method based on TCP/IP protocol are proposed according the analyses to the technical requirement and application features of print detection, and which solved the problem of image processing of board printing and image data transmission effectively. Furthermore, an interface of the software system has been developed on the basis of above.