Abstract: In order to preprocess the monitoring image, the image preprocessing was studied. Grayscale and binarization algorithm are highly complex, difficult to achieve and cost large computation. The weighted average method was redesigned, of which the floating-point computation was simplified to fixed-point for the benefit of achieving FPGA. An improved function based on Otsu algorithm was proposed and it was proved to be equivalent to Otsu algorithm. Moreover, the preprocessing algorithm was easy to be realized via hardware. After the Verilog programming, the algorithm was simulated on Matlab and ModelSim. The result demonstrates that the new algorithm is effective and achievable. Therefore, the design can be used as a preprocessing module of a real-time image processing system. As a result, it will save lots of time for the computations after preprocessing.

Key words: Field Programmable Gate Array (FPGA), image preprocessing, binarization, Otsu algorithm, iteration

摘要： 为实现对现场监控图像的实时预处理，针对灰度化和二值化算法复杂度高、计算量大和不易于硬件实现等问题，对图像预处理方法进行了研究。通过重新设计加权平均灰度算法的权重，使灰度算法由浮点运算简化为定点运算，易于现场可编程门阵列（FPGA）实现；提出一个基于Otsu算法的改进评价函数，证明了其与最大类间方差法、最小类内方差法的等价性。进而提出一种适于硬件实现的图像预处理算法，并通过Verilog编程以及在Matlab和ModelSim上联合仿真，证明了整个预处理功能的可实现性，最终设计出一种可用于实时图像处理系统前端的基于硬件实现的高速图像预处理模块。

关键词: 现场可编程门阵列, 图像预处理, 二值化, Otsu算法, 循环迭代