随着生产技术的快速发展，各大企业的生产效率不断提高，产品的检测成为了制约生产流程的一个重要环节。高效率的检测不仅能够保证产品的质量，提消费者对产品的认可度，而且能够节约生产成本，减少原料的浪费。本文提出一种基于FPGA的针对介质表面检测及处理系统。 本文研究了一种基于FPGA的介质表面检测及处理系统，通过对采集到的介质表面图像进行边缘检测来检测产品是否存在裂缝等瑕疵。首先利用CCD图像传感器，对待检测物品进行图像采集，并将所采集图像进行数据转换后传送给FPGA开发板，FPGA开发板将所采集到的图像传送到SDRAM中加以存储，并通过逻辑运算对采集图像进行边缘检测的处理，提取出边缘特征后的图像送到VGA口进行VGA图像显示。进行过边缘检测的图像可以轻易分辨产品的表面特征，进而分辨出产品是否合格。针对本系统，本文分别进行了图像采集模块、图像存储模块、图像处理模块和图像显示模块的设计，完成了介质表面检测及处理系统的设计，验证了系统的可行性。 本文边缘检测的算法以及各模块控制电路的设计是在Quartus II软件上采用Verilog HDL语言编写的，并且利用Modelsim仿真软件进行了仿真，仿真结果表明所设计的各逻辑电路模块的正确行，验证了本文方案的可行性。

With the development of production technology, the production efficiency of enterprises continuously improve, product quality testing has become a major factor that restricts the production process. High quality detection can not only raise consumer acceptance of products, but also can save production costs, reduce the waste of raw materials. The author proposes a surface detection and processing system based on FPGA for products. The author proposes a surface detection and processing system based on FPGA for products, through the image surface products on the acquisition of the edge detection to detect the existence of cracks and other defects of products. Firstly, using a CCD image sensor to obtain the image of products, then the CCD image sensor transfer the image to FPGA development board after converting the image format; the image will be stored in SDRAM by FPGA development board, on the other side, FPGA development board will get edge measuring for the image through logic operation, and after the image processed, FPGA development board will send it to the VGA display. The processed image can be easily distinguished whether the product is qualified by its surface features. In this system, the author designed four modules including image acquisition module, image storage module, image processing module and image display module. The feasibility of the system is justified. In this paper, the edge detection algorithm and the module control circuit is created by Verilog HDL language in Quartus II software, and simulated by Modelsim software, the results show that the logic circuit is feasible.