Study on 3-dimensional stereoscopic image model in intelligent coal mine

Abstract

This paper briefly describes the image analysis technology needed in the construction of intelligent coal mine, mainly for the reconstruction process of 3-dimensional (3D) image model in coal mine. If all the pixel points are concerned to calculate in whole picture, it should ask for higher computing processing equipment and large amount of computation. Therefore, this paper proposes a method which is firstly carried out target detection or image segmentation to get object to be recognized on the image, then taken image registration of target area according to the detected or segmented pixels, and the dense point matrix is diffused and fused according to the sparse point matrix, and then the 3D model is generated. In the 3D model, the length, depth and angle are easier to get accurate results, which plays a very important role in the subsequent scene analysis. In image 3D reconstruction, the region of registration has been greatly reduced due to image recognition or segmentation, and the number of registration points found by the improved registration using swarm intelligence algorithm is 6 times than algorithm without improvement, and the depth data obtained in the subsequent 3D reconstruction is basically consistent with that in the actual scene. In the future production monitoring, it is the judgement that can meet the corresponding business logic and production requirements.

Keywords

• Coal mine
• Target detection
• Image segmentation
• Image registration
• 3D model establishment