舰船远程监控平台模糊视频高动态图像拼接研究

Research on high dynamic image mosaic of fuzzy video in ship remote monitioring platform

为提高船舶目标检测中图像拼接质量,本次针对传统拼接方法存在的问题:无法同时实现拼接图像画质以及结构相似度的双向目标,研究了一种新的舰船远程监控平台模糊视频高动态图像拼接方法。该方法主要分为三步骤,首先对待拼接的模糊视频高动态图像进行处理,包括图像去噪、图像分割、图像畸形矫正,然后在此基础上利用基于特征信息方法进行图像配准,主要包括图像特征提取、图像匹配等,最后采用直接平均算法实现图像融合,完成图像拼接。结果表明:与传统图像拼接方法相比,本方法操作下,图像峰值信噪比分别提高1.8 d B和2 d B,结构相似度分别提高10.8%和8.2%,证明了本方法的有效性和优越性。

In order to improve the quality of image mosaic in ship target detection,a new method of high dynamic image mosaic based on Fuzzy video of ship remote monitoring platform is studied in this paper,aiming at the problems of traditional mosaic methods:it is impossible to achieve the two-way target of image quality and structure similarity at the same time.The method consists of three steps.Firstly,the high dynamic image of the fuzzy video to be stitched is processed,including image denoising,image segmentation and image deformity correction.Then,image registration is carried out based on feature information method,including image feature extraction,image matching and so on.Finally,image fusion is realized by using direct averaging algorithm to complete image stitching.The results show that compared with the traditional image mosaic method,the PSNR of the image is increased by 1.8 dB and 2 dB respectively,and the structural similarity is increased by 10.8%and 8.2%respectively,which proves the effectiveness and superiority of this method.