基于图像配准的建筑夜景照明故障检测方法  

Fault Detection Method of Building Night Scene Lighting Based on Image Registration

作　　者：陈静[1] 宋雪桦[1] 孙妍 蒋思玮 王昌达[1] 李文鹏 王聚 CHEN Jing;SONG Xuehua;SUN Yan;JIANG Siwei;WANG Changda;LI Wenpeng;WANG Ju(School of Computer Science and Communication Engineering,Jiangsu University,Zhenjiang 212013;Shanghai Luoman Lighting Technologies Inc.,Shanghai 200000)

机构地区：[1]江苏大学计算机科学与通信工程学院,镇江212013 [2]上海罗曼照明科技股份有限公司,上海200000

出　　处：《计算机与数字工程》2022年第11期2538-2543,共6页Computer & Digital Engineering

基　　金：国家自然科学基金项目(编号:61902156);江苏省自然科学基金项目(编号:BK20180860);镇江市重大专项成果转化(编号:ZD2019004)资助。

摘　　要：为了提高建筑夜景照明故障检测效率,论文提出了一种基于图像配准的建筑夜景灯光照明故障检测方法。首先采用改进的限制对比度自适应直方图均衡方法进行图像预处理,然后利用SIFT算法进行特征提取,根据欧氏距离和RANSAC算法进行特征点的配准,最后根据观察区域提取的像素特征判断是否有故障。实验结果表明,论文灯光故障检测方法可以准确检测建筑夜景灯光故障,具有较好的应用价值。In order to improve the efficiency of building night scene lighting fault detection,this paper proposes a method of building night scene lighting fault detection based on image registration.Firstly,the improved Contrast Limited Adaptive Histogram Equalization method with limited contrast is used to preprocess the image,then the SIFT algorithm is used to extract the features,and feature points are registered according to Euclidean distance and RANSAC algorithm.Finally,the fault is judged according to the features of pixels extracted from the observation area.The results show that the fault detection method in this paper has a good application value.

关 键 词：夜景照明 故障检测 图像增强 SIFT算法 图像配准 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术] TU113.6[自动化与计算机技术—计算机科学与技术]