A method for generating a CT three-dimensional structure of rice stems based on a GAN network

By using a GAN-based method to generate 3D structures of rice stalks using CT (Computer-Assisted Animation) technology, and employing a unified collaborative stereo complementarity module and a differential perception stereo complementarity module for feature information fusion and differential complementarity calculation, the problem of insufficient fine structure recovery and insufficient extraction of detailed features under sparse viewpoint conditions in the generation of 3D structures of rice stalks using CT technology is solved, and high-fidelity and stable 3D structure generation is achieved.

CN122244292APending Publication Date: 2026-06-19ANHUI AGRICULTURAL UNIVERSITY
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
ANHUI AGRICULTURAL UNIVERSITY
Filing Date
2026-01-30
Publication Date
2026-06-19

Smart Images

  • Figure CN122244292A_ABST
    Figure CN122244292A_ABST
Patent Text Reader

Abstract

This invention discloses a method for generating three-dimensional structures of rice stems using CT based on GAN networks, belonging to the field of agricultural data processing technology. The method includes the following steps: S1: collecting and processing rice stem data; S2: constructing a rice stem CT dataset based on the processed rice stem data, including multiple low-resolution three-dimensional structures, top and bottom cross-sectional CT slices, and high-resolution three-dimensional structures; S3: constructing a rice stem CT three-dimensional structure generation model, and training and testing the model based on the constructed rice stem CT dataset; S4: generating rice stem CT three-dimensional structures based on the tested model. This invention effectively solves the problem that traditional rice stem CT reconstruction methods struggle to accurately recover the fine internal structure of the stem under sparse viewpoint conditions, thus better meeting real-world application needs.
Need to check novelty before this filing date? Find Prior Art