Study Of Binocular Stereovision Technology In Underwater Robot

The premise of underwater robot navigation location technology is the satisfying comprehension of the underwater environment. Therefore, it has significant theory significance and project application significance to apply the environment sensation sensor to the underwater robot for understanding environment and the realization of navigation location. This article carries out the related technology research closely revolving the application of binocular stereovision system in the underwater robot location.Binocular stereovision technology can be defined as the process of abstraction, identification, matching and 3D reconstruction of two images of the identical objective in external environment condition of the carrier. These images are gathered under different visual angles. A complete binocular stereovision system generally includes: digital image gather, camera calibration, image preprocessing and feature extraction, stereo matching and 3D reconstruction as well. This article mainly studies three important parts: the camera calibration, the image preprocessing and segmentation, the stereo matching and 3D reconstruction:1. The camera calibration: We summarize and classify various methods of camera calibration of three-dimensional vision, mainly introduce the linearity calibration method based on 3D calibration template, the two-stage calibration method based on RAC and Zhengyou Zhang's flexible new technique for camera calibration. Finally we carry on the emulation test of camera calibration using Zhengyou Zhang's flexible new technique and provide the result of the test.2. The image preprocessing and segmentation: The image preprocessing and segmentation is the foundation of the stereo matching and 3D reconstruction. This article mainly studies the median filter of image, the edge characteristic based image segmentation technique and the area based image segmentation technique. Based on these we present the improved Otsu threshold segmentation algorithm.3. The stereo matching and 3D reconstruction: We mainly research the stereo matching method based on the area correlation. We through inspect the...