基于FPGA的纸病预处理系统

Paper Defect Pretreatment System Based on FPGA

关键词:  纸病检测  现场可编程门阵列  硬件加速  图像预处理  高速相机

Key Words:paper defect detection  field programmable gate array (FPGA)  hardware acceleration  image preprocessing  high-speed camera

摘要:针对当前计算机串行数据处理方式难以适应在线纸病检测系统数据实时处理量大的瓶颈问题，采用“FPGA+计算机”的纸病检测硬件结构,利用FPGA(现场可编程门阵列)的并行特性实现图像处理算法的硬件加速,对CCD高速摄像机拍摄的纸病图像进行预处理,初步确定可能出现纸病区域的位置和大小,并将这些区域的图像信息通过以太网发送到计算机做纸病的进一步辨识和分类。试验结果表明，这种结构可以极大地减轻计算机的运算量,提高纸病检测系统的精确度和效率。

Abstract:In view of the bottleneck problem that the current computer serial data processing mode is hard to adapt to process the large amount of data in the online paper defect detection system, a paper defect detection hardware structure of “FPGA & computer” is applied. FPGA(Field Programmable Gate Array) can be used to achieve the hardware acceleration of image processing algorithm. First, paper image which was taken by high-speed CCD camera is preprocessed by FPGA. Secondly, the position and size of paper defects are preliminary determined. And then, these above information is sent to a computer through the Ethernet to further identification and classification. The results showed that this structure can greatly reduce computation volume, and improve the accuracy and efficiency of paper defect detection system.