Seeking for the optimal seam-line is an important part in remote sensing image mosaic. Most algorithms for finding the optimal seam-line only emphasize the minimum difference between pixels rather than the integrity of objects. Therefore, the authors put forward a new algorithm for finding the optimal seam-line based on minimum gradient point in local area in this paper. This method adopts the gradient to indicate the change of gray scale and then finds the pixel of the minimum gradient in local area progressively according to the principle that the change of gray scale is milder when the gradient is smaller. Meanwhile, an improved difference method to solve the band effect caused by the hard correction method is proposed in this paper. The experimental results show that the optimal seam-line determined in this paper can avoid the great gray scale variation area and the objects can be reserved unbroken. It avoids the band effect well and the seam-line is removed smoothly, the mosaic image also has a good sense of sight. In addition, the proposed method is very simple, effective, and easy to realize.
程红, 郑悦, 孙文邦. 基于局部区域梯度最小点的遥感图像镶嵌算法[J]. 国土资源遥感, 2014, 26(1): 31-36.
CHENG Hong, ZHENG Yue, SUN Wenbang. Mosaic algorithm for remote sensing images based on minimum gradient point in local area. REMOTE SENSING FOR LAND & RESOURCES, 2014, 26(1): 31-36.
[1] Davis J.Mosaics of scenes with moving objects[C]//IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Santa Barnara:IEEE, 1998:354-360.[2] 温红艳, 周建中.基于灰色理论的遥感图像最佳镶嵌线检测[J].计算机工程与应用, 2009, 45(15):31-33. Wen H Y, Zhou J Z.Optimal seam line detection algorithm of remote sensing image mosaic based on grey system theory[J].Computer Engineering and Applications, 2009, 45(15):31-33.[3] 方亚玲, 焦伟利.利用对称动态轮廓模型自动检测最优镶嵌线[J].科学技术与工程, 2007, 7(14):3451-3456 Fang Y L, Jiao W L.Detecting the optimal seam line automatically in image mosaic with twin snakes model[J].Science Technology and Engineering, 2007, 7(14):3451-3456.[4] Wen H Y, Zhou J Z.An improved algorithm for image mosaic[C]//International Symposium on Information Science and Engineering.Shanghai:ISISE, 2008:497-500.[5] Chon J, Kim H, Lin C S.Seam-line determination for image mosaicking:A technique minimizing the maximum local mismatch and the global cost[J].ISPRS Journal of Photogrammetry and Remote Sensing, 2010, 65(1):86-92.[6] 王军, 朱宝山, 朱述龙, 等.基于相关系数的遥感图像拼接线检测算法[J].测绘与空间地理信息, 2011, 34(3):151-153. Wang J, Zhu B S, Zhu S L, et al.A stitching line detection algorithm of remote sensing image based on correlation coefficient[J].Geomatics and Spatial Information Technology, 2011, 34(3):151-153.[7] 方贤勇.图像拼接技术研究[D].杭州:浙江大学, 2005. Fang X Y.Studies on image mosaic[D].Hangzhou:Zhejiang University, 2005.[8] Pan J, Wang M, Li D R, et al.Automatic generation of seamline network using area Voronoi diagrams with overlap[J].IEEE Transactions on Geoscience and Remote Sensing, 2009, 47(6):1737-1744.[9] 范永弘, 万惠琼, 靳建立, 等.基于DSM的遥感图像拼接线自动生成技术[J].测绘科学技术学报, 2011, 28(1):33-36. Fan Y H, Wan H Q, Jin J L, et al.Remote sensing image mosaic line auto-extraction based on DSM[J].Journal of Geomatics Science and Technology, 2011, 28(1):33-36.[10] 武新, 张焕龙, 舒云星.基于视觉感知的镶嵌图像质量评价方法[J].计算机工程, 2008, 34(18):220-222. Wu X, Zhang H L, Shu Y X.Mosaic image quality evaluation method based on visual perception[J].Computer Engineering, 2008, 34(18):220-222.