基于对比度受限直方图均衡化和非锐化掩模的织物表面瑕疵图像增强算法  

Fabric surface defects image enhancement algorithm based on CLAHE and unsharp mask

作　　者：葛万凯 赵世海[1] 范雨佳 GE Wankai;ZHAO Shihai;FAN Yujia(School of Mechanical Engineering,Tiangong University,Tianjin 300387,China;School of Textile Science and Engineering,Tiangong University,Tianjin 300387,China)

机构地区：[1]天津工业大学机械工程学院,天津300387 [2]天津工业大学纺织科学与工程学院,天津300387

出　　处：《毛纺科技》2021年第12期68-74,共7页Wool Textile Journal

基　　金：天津市自然科学基金项目(18JCYBJC20200);中国纺织工业联合会应用基础研究项目(J202003)。

摘　　要：针对现有织物表面瑕疵图像增强算法仍存在无法兼顾改善对比度和增强瑕疵细节特征的问题,提出了一种融合对比度受限直方图均衡化(CLAHE)对比度增强和非锐化掩模细节增强的图像增强算法。首先,将采集到的白坯布瑕疵图像加权灰度处理;然后对灰度图像进行CLAHE处理,得到改善对比度的图像,对灰度图像进行基于均值滤波的非锐化掩模锐化处理,得到细节增强图像;最后将改善对比度后的图像作为输入图像,细节增强图像作为引导图像,进行引导滤波,实现图像增强。通过主观评价和客观评价指标对算法处理结果进行多组对比分析,其中融合算法的MG、PSNR和信息熵值均最大,并对算法处理后图像进行瑕疵识别,融合算法准确率最高,达到98%以上。结果表明,融合算法有效改善织物瑕疵图像对比度的同时也增强了瑕疵细节信息,有利于提高后续瑕疵识别和分类的准确性。In order to solve the problems that the existing fabric surface defect image enhancement algorithms are still unable to improve the contrast and enhance the detail features of fabric surface defects,an image enhancement algorithm which combines CLAHE contrast enhancement and non-sharpening mask detail enhancement was proposed.First of all,the grayscale defect image was weighted grayscale processing;then the grayscale image was processed by CLAHE to get the image with improved contrast,and the grayscale image was sharpened by non-sharpening mask based on mean filtering to get the detail enhancement image;finally,the improved contrast image was used as the input image,and the detail enhancement image was used as the guide image to guide filtering to achieve image enhancement.The processing results of the algorithm were compared and analyzed by subjective evaluation and objective evaluation indexes,among them,the MG,PSNR and information entropy values of the fusion algorithm were the largest,and the defect recognition of the image processed by the algorithm was carried out,the accuracy of the fusion algorithm was the highest,reaching more than 98%.The results show that the fusion algorithm not only improves the contrast of fabric defect images effectively,but also enhances the detail information of fabric defects,which is helpful to improve the accuracy of subsequent defect identification and classification.

关 键 词：瑕疵检测 对比度增强 细节增强 CLAHE 非锐化掩模 

分 类 号：TS106[轻工技术与工程—纺织工程] TP18[轻工技术与工程—纺织科学与工程]