Full-growth-period tomato 3D phenotype dynamic acquisition and analysis method and system

Abstract

The invention discloses a full-growth-period tomato 3D phenotype dynamic acquisition and analysis method and system. The method comprises the following steps: obtaining an environment difference distribution diagram based on environment monitoring and determining a target single plant; acquiring a multi-view image sequence of the full-growth-period based on image acquisition; obtaining a local scene point cloud based on three-dimensional reconstruction; obtaining a complete point cloud and a split point cloud through point cloud preprocessing and manual segmentation; carrying out organ segmentation through point search based on an adjacency relation to realize extraction of each internode and each leaf; performing automatic extraction and calculation of phenotypic parameters on the basis of organ segmentation; aiming at the point clouds of the same single plant in different periods, achieving phenotype parameter registration by utilizing a real topological structure; and performing data analysis based on the environment difference and the phenotype registration result. According to the method, the environment difference is combined with the full-growth-period tomato 3D phenotype, threshold setting or manual labeling is not needed in the algorithm aspect, and the method and system can adapt to tomato phenotype parameter extraction and registration in different periods.

Description

technical field [0001] The invention belongs to the field of agricultural automation, and in particular relates to automatic measurement of tomato phenotype parameters, in particular to a method and system for dynamically acquiring and analyzing tomato 3D phenotypes in the whole growth period. Background technique [0002] Tomato is one of the most valuable vegetable crops in the world. As one of the three major tomato growing regions in the world, my country has a pivotal position in the international market. High yield and high quality are the important goals of tomato cultivation. However, the acquisition of tomato phenotypic data still mainly relies on the traditional phenotypic analysis method, which is time-consuming, labor-intensive, cumbersome, highly subjective and easy to cause damage to tomato, which seriously hinders the development of tomato cultivation technology. [0003] In the prior art, it is proposed to use the laser scanner to obtain the three-dimensiona...

AI Technical Summary

Problems solved by technology

However, the acquisition of tomato phenotypic data still mainly relies on the traditional phenotypic analysis method, which is time-consuming, labor-intensive, cumbersome, highly subjective and easy to cause damage to tomato, which seriously hinders the development of tomato cultivation technology

[0003] In the prior art, it is proposed to use the laser scanner to obtain the three-dimensional information of the plant and the canopy surface to estimate the leaf area, leaf area index and plant height, but the hardware equipment is expensive and the data is redundant; the point cloud data of the plant is obtained by three-dimensional reconstruction Combined with the region growing algorithm, deep learning, etc. to complete the extraction of individual plants or main stems, however, different thresholds and manual labeling are required for different individual plants, and accurate extraction cannot be achieved for the characteristics of the curved main stem of tomato

Images