摘要
图像采集过程中由于光照角度的影响产生阴影,对图像阴影的有效去除能提高图像的暗原色修复能力,改善成像质量。传统的阴影去除方法采用盒子滤波算法,在图像出现光照色差的情况下,阴影去除的效果差。提出一种基于纹理损失最小约束的跟踪的图像阴影去除算法。对图像纹理分块结构进行重构,获取图像阴影的暗原色特征,对图像进行降噪处理,构建纹理损失最小约束函数,以此为约束参量图像阴影跟踪自适应均衡补偿,实现阴影有效去除。仿真结果表明,该算法进行图像阴影去除的效果较好,峰值信噪比较传统方法有所提高,展示了较好的图像处理性能。
As the box filtering algorithm adopted in the traditional shadow elimination method has poor shadow elimination effect when the illumination color aberration appears in the image, an image shadow tracking and elimination algorithm based on texture loss least constraint is proposed, with which the texture block structure of the image is reconstructed to obtain the dark primary color characteristics of the image shadow, and then the image is processed with noise reduction. The texture loss least constraint function is constructed, which is taken as a constraint parameter to perform the image shadow tracking adaptive equalization compensation and eliminate the shadow effectively. The simulation results show that the algorithm has good effect of image shadow elimination, the peak signal-to-noise ratio (SNR) is better than that of the traditional method, and has the superior image processing performance.
出处
《现代电子技术》
北大核心
2016年第24期104-108,共5页
Modern Electronics Technique
基金
国家自然科学基金(61065004)
内蒙古自治区自然科学基金面上项目(2014MS0616)
关键词
纹理损失
最小约束
图像阴影去除
峰值信噪比
texture loss
least constraint
image shadow elimination
peak signal-to-noise ratio