Research on Object Depth Information Measurement System in Image

Abstract

With the rapid development of computer science and technology and the electronic industry, 3d reconstruction technology in computer vision has received extensive attention, among which the accuracy of 3D reconstruction mainly depends on the acquisition of scene depth information. However, ordinary camera can obtain two-dimensional monocular visual image when shooting the objective world scene. This situation leads to the machine cannot get the object distance, size and motion speed information from the image itself, which affects the decision system to make the correct decision. In order to solve the traditional depth information measurement method complex operation, high cost, measurement equipment occupies a large space and load. In this paper, based on Harris-Sift algorithm, a monocular vision technique is proposed to measure the absolute depth information of the object in the image using monocular vision. The comparison results show that the error rate between the actual distance and the measured distance is less than 3.5%. This shows that this method can accurately measure the absolute depth of objects in static and short distance, and is superior to other methods.