| In recent years, techniques of surface defect detect for steel strips advance toward a fast,precise,intelligentize stage. At present, the international on-line inspection's study focus on techniques of computer-vision for the surface defect of steel strips .The techniques of computer-vision need to do fleet operation for a great deal of image data being collected. The ability of Real-time processing large amounts of data has become the key point of realization of surface defects detection system.At present, because the development of DSP technology, the advantage of DSP in real-time image Acquire and process is being prominent .The DSP-based image processing system can be applicable in many Kinds of practical fields. This dissertation is based on the analysis of demand for the surface defect detect of steel strips in the high-speed Steel strip production line.An real-time detection-system based on DSP6000 for the surface defects in steel strips is described by the aims of the project and the system parameters. This paper contains the overall design, hardware design and the corresponding development of the software modules.in hardware, the design of system platform is carried out, consist of design of TMS320C6205 core-process module, it can effectively process massive real-time data in the high-speed Steel strip production line . In addition, the design of video capture circuit can face linear or area CCD signal acquisition. System has good stable, reliable and can be used widely and deeply in the further applications. In software, the code is written, what include initalizatiom test of SDRAM and CPLD module,I~2C configure for AD and dirver of PCI., do the basic work for the system in further practical application.At last, this paper analysis an defect real-time detection algorithm that can be run on DSP.The main modules work normally after several tests. The experimental results show that the detection system reached the basic requirements. |
PDF Full Text Request