Soybean seed classification method based on hyperspectral image RGB reconstruction

• Main Authors: YANG XU, SONG SHAOZHONG, MA KEJIA
• Format: Patent
• Language: Chinese; English
• Published: 28.05.2024
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; PHYSICS

The invention belongs to the technical field of hyperspectral application, and particularly relates to a soybean seed classification method based on hyperspectral image RGB reconstruction, and the method specifically comprises the following steps: 1, obtaining a soybean seed image through a hyperspectral image collection device; the method comprises the following steps: acquiring a soybean seed image by using hyperspectral image acquisition equipment; carrying out black and white correction on the collected hyperspectral image to obtain a processed soybean seed image; obtaining a new data set according to a hyperspectral image synthesis principle; performing image enhancement and threshold segmentation on the original data set and the new data set, and well separating soybean seed samples from a background by adopting a morphological method; training through an SE-Resnet50 model in combination with a grey wolf optimization algorithm, and extracting a plurality of features of a test sample; and outputting a cl