LCD缺陷检测系统的图像压缩感知方法

摘要

图像采集是液晶显示屏(LCD)缺陷检测系统的关键步骤,而高清屏图像的大数据量给检测系统的实时数据采集、传输和处理造成了很大压力.为解决这一问题,本文提出了一种先行压缩采样后列压缩采样的压缩感知方法,并给出了一种LCD缺陷图像的压缩感知电路实现方案,该电路主要由压缩采样、控制、图像转置存储等模块组成.这种方法充分考虑了图像整体的稀疏性,提高了压缩效率,且结构简单,易于硬件实现.在FPGA上的仿真实验验证了此方法的有效性.%Image acquisition is the key step in the LCD Mura detection system,but the large data of high-definition screen image may bring severe pressure on the detection system in real-time data acquisition,transmission and processing.To solve this problem,a compressive sensing method is proposed by applying the compressive sampling first in row and then in column.And an implementation of compressive sensing circuit for LCD Mura image is also presented,it consists of compressive sampling module,control unit and image transpose RAM,etc.This method takes a full consideration of the whole image sparsity,making it a more effective way of image compressive sampling.The structure of the circuit is simple and easy to implement.Experiments on FPGA show the effectiveness and efficiency of the proposed method.