红外技术, 2019, 41 (4): 317, 网络出版: 2019-05-10
基于微相机阵列成像系统的图像拼接研究
Image Stitching Based on a Microcamera Array Imaging System
摘要
采用同心球镜及微相机阵列成像系统设计是解决传统光学中大视场与高分辨率之间相互制约关系的有效手段之一。介绍了一套同心球镜微相机阵列的成像系统,通过拼接微相机阵列采集的图像,实现大视场高分辨率成像。在获取整个图像的过程中,实现多幅子交叉重叠的图像拼接是微相机成像系统关键技术之一。针对亿级像素多幅子图像拼接算法复杂、运算量大的问题,分析了 SURF、BRISK算法的优缺点,设计了一种 SURF与 BRISK相结合的图像拼接算法,在 SURF基础上引入 BRISK描述符,利用选取的像素点对生成二进制的描述符序列。经测试,该算法在保证成像质量的情况下,大幅减少了计算量,加快了计算速度,有一定的工程应用价值。
Abstract
An imaging system based on concentric spherical mirrors and microcamera arrays can be an effective method to simultaneously provide a large field of view and a high resolution. We introduce such an imaging system by stitching images from a microcamera array. An image stitching algorithm that converts multiple overlapped images into a panoramic image is essential for the system. Considering the large amount of calculation required for gigapixel image stitching, an algorithm is designed based on SURF and the BRISK descriptor. BRISK combined with SURF is used to generate the descriptor of the bit string to hasten the calculation. The algorithm can perform on the same level as other methods while requiring less time.
王一博, 刘云猛. 基于微相机阵列成像系统的图像拼接研究[J]. 红外技术, 2019, 41(4): 317. WANG Yibo, LIU Yunmeng. Image Stitching Based on a Microcamera Array Imaging System[J]. Infrared Technology, 2019, 41(4): 317.