论文部分内容阅读
在大型古生物化石数字化过程中,为了充分展示化石的细节信息,往往需要拍摄大量的图像。为了实现大型古生物化石数字化数据的完整性,需要对这些大量的图像进行精密的图像拼接处理。基于这种应用需求的前提下,本文在自主研发的Mosaic of Image Program(MIP)图像拼接系统的基础上,对高精度的相机检校、畸变检校及改正和拼接缝的保真处理等方面进行研究,形成系统的古生物化石彩色合成影像数字化流程。在宜州化石馆的实际处理中,完成了杨氏锦州龙、蜥脚类恐龙、孔子鸟等大型古生物化石的数字化,几何失真小于0.36mm(畸变矫正精度优于1像元,拼接精度优于2像元,像片分辨率0.12mm)。同时采用基于SIFT的自动辐射归一化处理算法对拼接影像进行辐射均衡处理,矫正拼接影像辐射亮度的不均衡。
In the process of digitization of large paleontology fossils, in order to fully display the fossil details, often need to shoot a large number of images. In order to achieve the integrity of the digital data of large paleontology fossils, a large number of images need to be spliced in detail. Based on the premise of this application requirement, based on the Mosaic of Image Program (MIP) image mosaic system independently developed, this paper analyzes the high-precision camera calibration, distortion calibration and correction and the fidelity processing of splicing Research, the formation of a system of paleontology fossil color digital image synthesis process. In the actual treatment of Yizhou fossil Museum, Yang completed the Jinzhou dragon, sauropod dinosaurs, Confucius birds and other large paleontology fossils, the geometric distortion is less than 0.36mm (distortion correction accuracy better than 1 pixel, the stitching accuracy is better than 2 pixels, photo resolution of 0.12mm). At the same time, the SIFT-based automatic radiation normalization algorithm is used to radiate the stitched image to correct the unbalance of the radiance of the stitched image.