Stumpage depth map generation method based on short video image

Abstract

The invention discloses a stumpage depth map generation method based on a short video image, and the method comprises the steps: firstly taking a to-be-processed stumpage as the center, and slightly moving a photographing device to photograph a small-motion short video; selecting a first frame of the small motion short video as a reference standing tree image, and performing front and back background annotation on the standing tree image by using a Graph Cut standing tree segmentation algorithm to obtain a standing tree segmentation image; extracting sub-pixel angular points to obtain featurepoints; after error points are eliminated by tracking the matched feature points, obtaining parameters of the shooting device by utilizing bundle adjustment, and performing distortion correction on asmall motion short video image sequence; carrying out the dense stereo matching by using a plane scanning method to obtain an initialized standing tree depth map; and then, according to the similaritybetween the image blocks, performing joint filtering on the standing tree depth map, removing abnormal pixel values, and filling holes to obtain a final standing tree depth map. According to the method, the flexibility and convenience of standing tree scene depth information acquisition are improved, and the expenditure of data acquisition and storage can be effectively saved.

Description

technical field [0001] The invention relates to the technical field of standing tree image processing, in particular to a method for generating a standing tree depth map based on a short video image. Background technique [0002] At present, obtaining the distance of each point in the scene relative to the shooting device is one of the important tasks of computer vision system. Widely used in agriculture and forestry measurement. The standing tree depth map refers to the image containing the standing tree object and depth information. The special structure of the standing tree and the complex growth environment make the acquisition method of the depth map limited in the standing tree scene. [0003] In the prior art, the depth camera equipment required to shoot images of standing trees is expensive, subject to great interference from ambient light, and the post-processing cycle is long; this method of obtaining depth clues from a single target image to generate a depth map,...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to provide a method for generating a standing tree depth map based on a short video image, which improves the flexibility and convenience of obtaining depth information of a standing tree scene, can effectively save the cost of data collection and storage, and solves the problem of the current depth map. Difficulty in obtaining prior information of standing tree images by generative algorithms

Images