机电工程技术2024，Vol.53Issue(2)：212-216,5.DOI:10.3969/j.issn.1009-9492.2024.02.046
基于亮度均衡化和仿射变换的交通标志图像修复研究
Research on Traffic Sign Image Restoration Based on Luminance Equalization and Affine Transformation
摘要
Abstract
Autonomous driving technology is the current research hotspot of artificial intelligence,in which the correct recognition of traffic signs is an important judgment basis for machines,but traffic sign images in natural scenes have a series of problems such as color deviation,shape deformation and occlusion,resulting in inaccurate recognition.For this problem,a method of brightness equalization and affine change is proposed,first using HSV color space,extracting the corresponding color,and then histogram equalization method for luminance information to solve the problem of color deviation.Subsequently,affine transformation is used to correct the shape of the triangle and circle images respectively,and to remove the background.The experimental part uses HOG+SVM detection method with low machine resource consumption for detection,and the results show that compared with traditional methods,this method has an improvement of about 20%in recognition rate.This proves that targeted restoration of traffic sign images has significant implications for improving recognition rates,as it consumes less machine resources and provides new ideas for current traffic sign recognition methods such as deep learning that consume more machine resources.关键词
HSV颜色空间/直方图均衡化/仿射变换Key words
HSV color space/histogram equalization/affine transformation分类
信息技术与安全科学引用本文复制引用
廖干洲,曾霞..基于亮度均衡化和仿射变换的交通标志图像修复研究[J].机电工程技术,2024,53(2):212-216,5.基金项目
2020年广东高校省级重点平台和重大科研项目(2020KTSCX197) （2020KTSCX197）
广州应用科技学院 2022年度校级教学质量与教学改革工程项目(2022ZG008) （2022ZG008）
