Plant point cloud segmentation method and system based on two-dimensional-three-dimensional integration and storage medium

Abstract

The invention provides a plant point cloud segmentation method based on two-dimensional and three-dimensional integration. Firstly, multi-angle images of a plant are shot, then a three-dimensional point cloud model of the plant is reconstructed through a motion recovery structure (SFM) algorithm, and meanwhile the mapping relation between each image and the three-dimensional point cloud model is obtained. Preprocessing operations such as denoising are carried out on the reconstructed point cloud model, and a segmentation result based on the three-dimensional point cloud model is trained by using a Point Net network model. And meanwhile, based on the mapping relationship between each image and the three-dimensional point cloud model, a segmentation result is acquired based on the two-dimensional image of the plant through a MaskRCNN network. Finally, linear weighted integration is performed on the two-dimensional and three-dimensional results to obtain a final plant point cloud segmentation result. The invention also provides a system for realizing the plant point cloud segmentation method based on the two-dimensional and three-dimensional integration and a storage medium for enabling the processor to execute the plant point cloud segmentation method based on the two-dimensional and three-dimensional integration.

Description

technical field [0001] The present invention relates to the field of three-dimensional point cloud segmentation of plants, and more specifically, to a plant point cloud segmentation method, system and storage medium based on two-dimensional-three-dimensional integration. Background technique [0002] At present, the traditional methods of 3D point cloud model segmentation of plants include the segmentation method based on region growth and the segmentation method of deep learning. The segmentation method based on region growth mainly utilizes the geometric characteristics of stems and leaves of plants, and the stem is first segmented out through the region growth algorithm. , and then start from the intersection of stems and leaves, and then segment the point cloud of each leaf through region growth; this type of method has high requirements for the type of plants, and it is difficult to find an accurate threshold to segment the plants, so it needs to be done manually many ti...

AI Technical Summary

Problems solved by technology

[0005] The present invention aims to solve at least to a certain extent the technical defects in the point cloud segmentation described in the above-mentioned prior art that the segmentation accuracy is not high, the outline of the plant is clearly expressed, and the problem of occlusion between the leaves of the plant cannot be taken into account at the same time.

Images