Research On Pose-Free Stitching And Color Consistency Correction For Aerial Images

Nowadays,satellite and aerial remote sensing based imaging techniques have very important applications in quickly acquiring ground information,monitoring territorial resources,etc.In many practical situations,the observing range of a single image is too limited to cover the whole investigating areas,so stitching multiple images into a wide-view seamless mosaic image is a necessary technique to support applications based on overall survey and global analysis.Generally,image stitching is composed of four routine steps:geometric alignment,color correction,seamline selection and boundary fusion.As the most vital modules in low-flight aerial image stitching,global consistent alignment and color consistency correction are the researching majority in this paper.Global consistent alignment.we propose a generic framework for globally consistent alignment of images captured from approximately planar scenes via topology analysis,capable of resisting the perspective distortion meanwhile preserving the local alignment accuracy.Firstly,to estimate the topological relations of images efficiently,we search for a main chain connecting all images over a fast built similarity table of image pairs(mainly for the unordered image sequence),along which the potential overlapping pairs are incrementally detected according to the gradually recovered geometric positions and orientations.Secondly,all the sequential images are organized as a spanning tree through applying a graph algorithm on the topological graph,so as to find the optimal reference image which minimizes the total number of error propagation.Thirdly,the global alignment under topology analysis is performed in the strategy that images are initially aligned by groups via the affine model,followed by the homography refinement under the anti-perspective constraint,which manages to keep the optimal balance between aligning precision and global consistency.Color consistency correction.we propose an effective color correction method which is feasible to optimize the color consistency across images and guarantee the imaging quality of individual image meanwhile.Our method first apply specific algorithms to detect coherent-content regions in inter-image overlaps where reliable color correspondences are extracted.Then,we parameterize the color remapping curve as transform model,and express the constraints of color consistency,contrast and gradient in an uniform energy function.It can be formulated as a convex quadratic programming problem which provides the global optimal solution efficiently.To evaluate our global consistent alignment approach,two groups of low-flight aerial images covering large areas are used as the testing datasets.The experimental results show that our proposed approach can make a superior aligning accuracy over commercial software termed PTGui and meanwhile keep the global consistency well.Besides,to evaluate the generosity and validity of our color correction algorithm,we experiment on both ground panorama dataset and remote sensing dataset including multi-temporary satellite images and color edited aerial images.Comparing with existing methods,our color correction algorithm demonstrates outstanding performance in both qualitative and quantitative evaluation.