摘要
图像在过度曝光采集的情况下会因为光圈的进光量过大导致图像的细节关键信息缺失,研究过度曝光图像的缺失信息修复算法,为图像的细节特征分析奠定基础。传统方法采用图像小波尺度分解方法进行图像缺失信息修复,随着光圈的增大,对细节信息的修复效果不好,提出一种基于改进的小波包分解的过度曝光图像缺失信息修复算法。构建了过度曝光图像的纹理信息特征传导模型,采用Harris角点检测算法实现对过度曝光图像的灰度特征匹配,采用小波包分解方法对图像缺失信息进行位置和尺度信息的重构,实现缺失信息修复。仿真结果表明,采用该算法进行过度曝光图像修复,图像的细节特征得到准确有效复原,提高图像的识别能力。
Image in excessive exposure the acquisition because of the aperture of the amount of lightcauses the image details the key information is missing, the excessive exposure the missing informationof the image inpainting algorithm, the details of image feature analysis lay the foundation. In thetraditional method, the image wavelet scale decomposition method is used to repair the image missinginformation. As the aperture increases, the repair effect of the detail information is not good. The textureinformation of the image is constructed by using the Harris corner detection algorithm. The gray levelfeatures of the image are matched. The location and scale information of the image is reconstructed bywavelet packet decomposition method.
出处
《科技通报》
北大核心
2016年第8期146-149,共4页
Bulletin of Science and Technology
关键词
图像
曝光
光圈
小波分析
修复
image
exposure
aperture
wavelet analysis
repair