不均匀光照下的合作目标图像分割方法  

Cooperative target image segmentation method under uneven illumination

作　　者：黄亮[1,2,3,4] 郝颖明 HUANG Liang;HAO Yingming(Key Laboratory of Opto-Electronic Information Processing,Chinese Academy of Sciences,Shenyang Liaoning 110016,China;Shenyang Institute of Automation,Chinese Academy of Sciences,Shenyang Liaoning 110016,China;Institute for Robotics and Intelligent Manufacturing,Chinese Academy of Sciences,Shenyang Liaoning 110169,China;University of Chinese Academy of Sciences,Beijing 100049,China)

机构地区：[1]中国科学院光电信息处理重点实验室,沈阳110016 [2]中国科学院沈阳自动化研究所,沈阳110016 [3]中国科学院机器人与智能制造创新研究院,沈阳110169 [4]中国科学院大学,北京100049

出　　处：《计算机应用》2024年第S01期229-234,共6页journal of Computer Applications

摘　　要：针对复杂光照条件下目标测量中,合作目标图像很难被快速、准确分割这一难题,将模糊积分引入图像分割领域,改进传统的局部自适应阈值算法。首先,采用面积求和表(SAT)算法求出合作目标图像的积分图像;其次,根据模糊测度和所选择的离散模糊积分公式计算模糊积分图像;最后,根据模糊积分图像计算每个像素点的阈值,对原图像进行二值化。实验结果表明,改进后的基于模糊积分的局部自适应阈值算法分割的平均结构相似性(SSIM)指标相较于改进前普遍提升了10%~30%,分割效率大幅提升,最终成功地分割合作目标图像。Aiming at the difficult problem that cooperative target images are difficult to be segmented quickly and accurately in target measurement under complex lighting conditions,fuzzy integral was introduced into the field of image segmentation to improve traditional local adaptive threshold algorithm.Firstly,Summed Area Table(SAT)algorithm was used to calculate the integral images of cooperative target images.Then,the fuzzy measure and the selected discrete fuzzy integral formula were utilized to calculate fuzzy integral images.Finally,the pixel-wise thresholds were calculated according to the fuzzy integral image for the binarization of the original image.Experimental results show that the improved local adaptive threshold algorithm based on fuzzy integral has a general improvement of 10%-30%in average SSIM(Structural SIMilarity)index of segmentation results,significantly improves the segmentation efficiency,and successfully segments cooperative target images.

关 键 词：合作目标图像 阈值分割 局部自适应阈值分割 模糊测度 模糊积分 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]