SAR crop classification method based on multi-scale time-varying information extraction network
By building a multi-scale time-varying information extraction network, preprocessing and classification of timing SAR images, the problems of low utilization rate of crop timing information and large model parameters in the prior art are solved, and efficient and accurate crop classification and all-weather monitoring are achieved.
CN119649210AActive Publication Date: 2025-03-18深圳市规划和自然资源数据管理中心(深圳市空间地理信息中心) +1
3 Cites 0 Cited by
Patent Information
- Application Number
- CN202411609252.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-12
- Publication Date
- 2025-03-18
- Estimated Expiration
- 2044-11-12
AI Technical Summary
Technical Problem
When using deep learning methods to classify time-series SAR images in the prior art, there are problems such as low utilization of crop timing information, relying on a large number of two-dimensional structures, and large model parameters.
Method used
Using a method based on a multi-scale time-varying information extraction network, the initial network model that integrates multi-scale time-varying information module and multi-scale spatial information module is constructed to preprocess and classify time-varying SAR images to reduce dependence on other auxiliary data.
Benefits of technology
It realizes efficient and accurate classification of crops under low parameter conditions, and can monitor crop planting areas all day and all day.
✦ Generated by Eureka AI based on patent content.
Abstract
The invention relates to an SAR (Synthetic Aperture Radar) crop classification method based on a multi-scale time-varying information extraction network. The method comprises the following steps: acquiring a to-be-classified time sequence SAR image; preprocessing the to-be-classified time sequence SAR image, and performing registration according to a preprocessing result and preset label data to obtain to-be-classified slice data; and inputting the to-be-classified slice data to a pre-constructed classification network model to obtain a crop classification result of the to-be-classified time sequence SAR image, the pre-constructed classification network model being obtained by training an initial network model constructed by fusing a multi-scale time-varying information module and a multi-scale spatial information module. Therefore, the information in the time sequence SAR image is classified by constructing the multi-scale time-varying information network model, so that automatic, intelligent, efficient and accurate classification of crops is realized under the condition of low parameter quantity, and all-day and all-weather crop planting area monitoring is realized.
Need to check novelty before this filing date? Find Prior Art