Parcel scale crop classification method, device, equipment and medium

• Main Authors: CHENG XIAOHUI, ZHAO CHUANBAO, HU XIN, LIU YANG, HOU JINGXIN, HUANG SHUJUAN
• Format: Patent
• Language: Chinese; English
• Published: 16.07.2024
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; PHYSICS

The invention discloses a land parcel scale crop classification method, apparatus and device, and a medium. The method comprises the steps of obtaining a historical high-resolution remote sensing image and historical land parcel vector data, generating land parcel sample data, inputting a preset SAM model for fine tuning, and obtaining a fine tuning SAM model; obtaining a current high-resolution remote sensing image, and inputting the current high-resolution remote sensing image into the fine-tuning SAM model to obtain a plot segmentation result The method comprises the following steps: acquiring a current time sequence SAR image, filtering and denoising, respectively extracting gray features, texture features and polarization features, and cascading to obtain a first cascading feature; the land parcel segmentation result and the first cascade feature are nested, and mean value operation is carried out on each feature of each land parcel to obtain a second cascade feature; and inputting the second cascade fea