红外技术, 2018, 40 (4): 382, 网络出版: 2018-06-09   

结合最佳缝合线和改进渐入渐出法的图像拼接算法

Image-stitching Algorithm by Combining the Optimal Seam and an Improved Gradual Fusion Method
作者单位
1 重庆理工大学电气与电子工程学院,重庆 400054
2 重庆理工大学计算机科学与工程学院,重庆 400054
摘要
针对图像拼接时因曝光差异较大或运动物体存在导致融合图像产生拼接缝、重影等问题,提出了一种基于最佳缝合线和改进的渐入渐出融合法相结合的图像拼接算法。首先,在两幅待拼接图像的重叠区域,根据动态规划思想获得最佳缝合线,然后将图像沿其进行拼接,最后由改进的渐入渐出法的融合规则来确定融合图像重叠区域像素点的灰度值,即将重叠区域均分成左、中、右3 部分,分别求左部分在待拼接左图像、右部分在待拼接右图像的像素点的灰度值与两幅图像中该像素点加权平均后的灰度值的差值,通过比较其差值的绝对值与设定阈值的大小来确定融合像素点的灰度值,中间部分融合像素点的灰度值则由其加权平均值确定,从而完成图像融合。实验结果表明,该方法能使图像重叠区域过渡更平滑自然,消除重影现象且没有明显拼接痕迹。
Abstract
Aiming at the problem of the seam or ghost due to different exposure conditions or moving objects during image stitching, an image-stitching algorithm based on the optimal seam and an improved gradual fusion method is proposed. First, in the overlapping areas of two stitching images, the optimal seam is obtained according to the idea of dynamic programming; then the images are stitched along the optimal seam. Finally, the pixel gray values of the fusion image are determined by the fusion rules of the improved gradual fusion method, namely: The overlapping areas are divided into three equal parts from left to right, and the difference between the pixel gray values of the left stitching image in the left part of the overlapping areas and right stitching image in the right part of the overlapping areas and the pixel gray weighted average of the two images are calculated. Then the fusion pixel gray values of the left and right parts are determined by comparing the absolute value of the difference with the set threshold; the fusion pixel gray value of the middle part is the weighted average. Thus, the image fusion is completed. Experimental results show that the method can make the image overlapping areas smoother and more natural and eliminate the phenomenon of ghost and stitching traces.

罗永涛, 王艳, 张红民. 结合最佳缝合线和改进渐入渐出法的图像拼接算法[J]. 红外技术, 2018, 40(4): 382. LUO Yongtao, WANG Yan, ZHANG Hongmin. Image-stitching Algorithm by Combining the Optimal Seam and an Improved Gradual Fusion Method[J]. Infrared Technology, 2018, 40(4): 382.

本文已被 3 篇论文引用
被引统计数据来源于中国光学期刊网
引用该论文: TXT   |   EndNote

相关论文

加载中...

关于本站 Cookie 的使用提示

中国光学期刊网使用基于 cookie 的技术来更好地为您提供各项服务,点击此处了解我们的隐私策略。 如您需继续使用本网站,请您授权我们使用本地 cookie 来保存部分信息。
全站搜索
您最值得信赖的光电行业旗舰网络服务平台!