基于红外和可见光图像融合的夜间环境图像增强  

Research on night environment image enhancement algorithm based on infrared and visible light image fusion

作　　者：邱永生 孙国兵 王毓武 隋胜春 QIU Yong-sheng;SUN Guo-bing;WANG Yu-wu;SUI Sheng-chun(College of Electronic Engineering,Heilongjiang University,Harbin 150080,China)

机构地区：[1]黑龙江大学电子工程学院,哈尔滨150080

出　　处：《哈尔滨商业大学学报（自然科学版）》2021年第4期422-427,449,共7页Journal of Harbin University of Commerce：Natural Sciences Edition

基　　金：黑龙江大学研究生创新科研项目(No YJSCX2020-171HLJU)。

摘　　要：图像增强技术可以有效提高图像的整体或局部特性增加图像有用信息,改善图像的视觉效果.夜间图像增强是图像处理领域中的关键技术,使用图像增强可以提高夜间环境图像的信息量.对于单一传感器采集的夜间可见光图像,传统算法通过提高图像的饱和度和对比度进行图像增强,处理后的夜间图像易出现边缘锐化或图像伪影的现象.相比于单一传感器的夜间图像增强,红外和可见光图像融合技术可以大幅度的提高图像信息有效率.利用红外和可见光图像的成像特点将两者进行融合,有效地提高夜间复杂环境下图像的信息量,从而达到夜间图像增强的效果.实验证明使用红外和可见光图像融合的夜间环境图像增强方法优于传统算法,有效抑制边缘锐化、图像伪影和图像过度增强的问题,同时提高了夜间环境下图像的主客观质量.Image enhancement technology can effectively improve the overall or partial characteristics of high-resolution images,increase useful information of images,and improve the visual effects of images.Night image enhancement is a key technology in the field of image processing.Using image enhancement can increase the amount of information in the night environment image.For visible light images collected at night by a single sensor,traditional image enhancement improves the saturation and contrast of the image,and the processed night-time image will appear edge sharpening or image artifacts.Compared with the night image enhancement of a single sensor,infrared and visible light image fusion technology can greatly improve the utilization of image information.The imaging characteristics of infrared and visible light images are fuse the to effectively improve the image quality at night in complex environments.so as to achieve the effect of night image enhancement.Experiments proved that the night environment image enhancement method using infrared and visible light image fusion was better than traditional algorithms,effectively edge sharpening,image artifacts and image over-enhancement problems,while improving the subjective and objective quality of images in the night environment.

关 键 词：多源图像融合 夜间图像增强 红外和可见光图像 图像伪影 

分 类 号：TP751[自动化与计算机技术—检测技术与自动化装置]