A forest pest forecasting method based on image processing
By employing a dynamic routing mechanism guided by Shearlet transformation and direction guidance, the problem of distinguishing disease and damage structures in forest remote sensing images has been solved, enabling efficient prediction of disease severity and spatial distribution, which is applicable to forestry management.
CN122156972APending Publication Date: 2026-06-05SICHUAN AGRI UNIV
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- SICHUAN AGRI UNIV
- Filing Date
- 2026-03-04
- Publication Date
- 2026-06-05
Smart Images
Figure CN122156972A_ABST
Abstract
The present application relates to the technical field of image processing, and more particularly to a forest pest prediction method based on image processing, comprising the following steps: step 1: obtaining remote sensing images of a forest area to be detected collected at continuous multiple time points, splicing the three images to form a Shearlet feature map, and taking a main edge direction map as a direction guide map; step 2: inputting the Shearlet feature map into a pre-trained pest capsule network, combining pest capsule sets at each time point with time interval coding vectors to form a pest capsule time sequence; step 3: projecting the pest capsule time sequence into an input embedding sequence; and combining a routing distribution matrix and a pest routing matrix to generate a pest distribution prediction map. The present application simultaneously has pest grade prediction and spatial distribution prediction capabilities, and can provide comprehensive pest warning information for forestry management personnel.
Need to check novelty before this filing date? Find Prior Art