Research On Presswork's Real-Time Defects Detection Based On Linear CCD Camera

Presswork has become an important part of human life. With the continuous development of people's living standards, more and more presswork appears in the market, which meets the needs of customers with high quality and diversity. At the same time, some problems come out in the presswork field along the development. That is how to detect the defects of the presswork quickly and objectively. The traditional methods which detect the defects of the presswork with human eyes can not complete the detecting job well because of the subjective aspects. While with the aid of computer science, an automatic detecting method which based on machine vision becomes available. In foreign countries some factories have made out this kind of product that is good in quality. In our country it is still underway.Based on image technology, some basic methods of image quality detection have been analyzed, and a scheme of real-time presswork quality detection equipment is built. This paper presents the paging detection method for common dot defects, methods for line and offset defects are also discussed.An experiment system of presswork quality real-time detection was developed, its software is developed with VC6.0, which can realize real-time detection with the use of digital image processing. Working principle, framework and working flow chart of the system has been illustrated and the paper explains the selecting principle and the parameters setting of the imaging equipments, such as line CCD camera and image grabber. The system can capture the image of the object that is moving at a high speed, and it can deal with the image message at real time.The experiment is carried out with the presswork image in the real-time detecting system. The experiment results demonstrate that the system can realize real-time detection of the presswork image, and they prove the methods proposed in the paper precisely detect the basic defects in the presswork image in motion, and that the method can be used to detect the presswork image at real-time.