融合边缘信息的对比度增强算法  

Contrast Enhancement Algorithm for Edge Information Fusion

作　　者：郭钰璐 董丽丽[1] 许文海[1] GUO Yulu;DONG Lili;XU Wenhai(College of Information Science and Technology,Dalian Maritime University,Dalian 116026,China)

机构地区：[1]大连海事大学信息科学技术学院

出　　处：《红外技术》2019年第7期616-622,共7页Infrared Technology

基　　金：中国自然科学基金项目(61701069);中央高校基本科研业务费专项资金资助(3132019200)

摘　　要：针对红外图像中对比度低,细节不清晰,视觉效果模糊等问题,提出一种结合边缘信息的对比度增强算法。首先,使用引导滤波将原始红外图像分解为基础图像和细节图像,并且通过使用对比度限制的直方图均衡来处理基础图像,提高图像对比度,克服“过度增强”现象;利用Gamma变换处理细节图像,增强细节信息;再将处理后的两幅图像融合成图ImageSD;然后为了有效地改善ImageSD的亮度不均匀的现象,对原始图像进行自适应直方图均衡和拉普拉斯锐化滤波;最后,将两个图像进行线性加权并融合以重建出最终的红外图像。结果表明,该方法可以更好地提升原始图像的对比度,丰富细节信息。Aiming at the problems of low contrast,unclear details,and blurred visual effects in infrared images,this paper proposes a contrast enhancement algorithm that combines edge information.First,the boot filter is used to decompose the original infrared image into base and detail images,followed by the application of histogram equalization processing of the base image with limited contrast and improvement of the image contrast to resolve the "over-enhancement" phenomenon.Next,the detail image is processed using the Gamma transform;details are enhanced,and the images are merged into a picture ImageSD;subsequently,to effectively improve the brightness unevenness of ImageSD,adaptive histogram equalization and Laplacian sharpening filtering are performed on the original image.Finally,the two images are linearly weighted and fused to reconstruct the final infrared image.Results show that the proposed method can better enhance the contrast of the source image and enrich detailed information.The objective evaluation index is superior than the existing methods,and the fusion result has a better subjective visual effect.

关 键 词：图像增强 引导滤波 限制对比度的直方图均衡化 锐化滤波 融合边缘信息 

分 类 号：TN219[电子电信—物理电子学]