改进的红外CCD自动调光算法的研究  

Research on improved infrared CCD auto-dimming algorithm

作　　者：杨友良 刘爱旭 马翠红 

机构地区：[1]华北理工大学电气工程学院

出　　处：《激光杂志》2018年第3期110-113,共4页Laser Journal

基　　金：国家自然科学基金项目(No.61171058)

摘　　要：讨论了目前的CCD相机成像时调光算法的原理,详细地分析了它们存在的共同局限性,针对目前的红外CCD相机在采集高温物体图像时易出现过亮或过暗的缺陷,提出一种基于图像灰度均值和基于“中间灰度”的标准相结合的改进的自动调光算法。这种算法建立在温度、图像灰度值和积分时间的相互作用关系上,可自动获得高温物体曝光时需要的积分时间,为红外CCD相机获取高温物体清晰图像提供了一种可行途径。实验部分选择两个典型钢水温度值分别为1523.2℃和1573.2℃作为自动调光算法的理论验证,并且通过比较原始图像和最佳图像的均方误差和峰值信噪比。实验结果表明,该自动调光算法可以获得稳定清晰的高温钢水图像,图像的峰值信噪比提高了10d B。This paper discussed the principle of traditional CCD camera and analyzed their common limitations. For the defects such as over brightness or darkness of current infrared CCD camera in the acquisition of high tempera- ture object image. Therefore, this paper proposes an improved automatic adjustment optical algorithm based on the combination of image gray mean value and the ＂middle gray scale＂ standard. This algorithm is based on the interaction among temperature, image gray value and integration time, and it can automatically obtain the integration time required for high temperature exposure. It provides a feasible way for infrared CCD camera to obtain clear image of high temper- ature object. In the experiment,two typical temperature values of molten steel are 1523.2℃ and 1573.2℃ respective- ly, and the mean square error and peak signal to noise ratio of the original image and the best image are compared. The experimental results show that the automatic dimming algorithm can obtain stable and clear high temperature molten steel image,and the peak signal to noise ratio of image is improved by 10 dB.

关 键 词：红外CCD相机 自动调光 灰度均值 中间灰度 

分 类 号：TN249[电子电信—物理电子学]