Abstract
Precise reconstruction of the morphological structure of the soybean canopy and acquisition of plant traits have great theoretical significance and practical value for soybean variety selection, scientific cultivation, and fine management. Since it is difficult to obtain all-around information on living plants with traditional single or binocular machine vision, this paper proposes a three-dimensional (3D) method of reconstructing the soybean canopy for calculation of phenotypic traits based on multivision. First, a multivision acquisition system based on the Kinect sensor was constructed to obtain all-around point cloud data of soybean in three viewpoints, with different fertility stages of soybean as the research object. Second, conditional filtering and K-nearest neighbor filtering (KNN) algorithms were used to preprocess the raw 3D point cloud. The point clouds were matched and fused by the random sample consensus (RANSAC) and iterative closest point (ICP) algorithms to accomplish the 3D reconstruction of the soybean canopy. Finally, the plant height, leafstalk angle and crown width of soybean were calculated based on the 3D reconstruction of soybean canopy. The experimental results showed that the average deviations of the method was 2.84 cm, 4.0866° and 0.0213 m, respectively. The determination coefficients between the calculated values and measured values were 0.984, 0.9195 and 0.9235. The average deviation of the RANSAC + ICP was 0.0323, which was 0.0214 lower thanthe value calculated by the ICP algorithm. The results enable the precise 3D reconstruction of living soybean plants and quantitative detection for phenotypic traits.
Highlights
Soybean is grown on a large area in the northern Heilongjiang Province of China, accounting for more than half of the province’s total area, but the overall yield level is not high because of environmental constraints, and soybean varieties in the northern Heilongjiang Province are of a single type, so the yield traits are not fully consistent with other regions [2]
The plant height deviations of the obtained samples were within the allowable range, which verified the validity of the 3D reconstruction method for soybean canopy
A 3D reconstruction algorithm of the soybean canopy based on multivision technology is proposed
Summary
Soybean is an important economic crop that occupies an important position in. Measurements based on 2D images are often less accurate because of interplant shading and are even unsuitable for measuring morphological parameters such as canopy width because of dimensional limitations [10]. Accurate acquisition of the 3D point cloud of the soybean canopy and calculation of phenotypic traits such as plant height, leafstalk angle and crown width are necessary for selecting and breeding excellent soybean varieties. The PMD camera can acquire both color and deep images It is relatively cheap and portable, with a high framerate, but the low resolution and sensitivity to external light limits its further application [12,13]. In response to the problems of the fact that the full-view 3D structural morphology of the soybean canopy cannot be obtained from single and dual views, the low efficiency of manual acquisition of plant phenotypes and the high price of high-end devices, three. On the basis of the reconstruction results, the phenotypic traits of the soybean canopy were calculated
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
