A method for detecting ecological restoration state change based on multi-temporal slope image registration

By using image layering and base reference map generation methods, the problem of local registration drift in slope ecological restoration status detection was solved, and more accurate detection of changes in ecological restoration status was achieved.

CN122391249APending Publication Date: 2026-07-14JIANGSU SHANSHUI ECOLOGY ENVIRONMENT CONSTR ENG CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
JIANGSU SHANSHUI ECOLOGY ENVIRONMENT CONSTR ENG CO LTD
Filing Date
2026-06-17
Publication Date
2026-07-14

AI Technical Summary

Technical Problem

Existing multi-temporal slope image registration methods are prone to local registration drift due to changes in vegetation cover and illumination in slope ecological restoration status detection, leading to misidentification of false changes in the change detection results.

Method used

By identifying base pixels, ecological cover pixels, and transient interference pixels in image layering, the most suitable base texture fragment is selected to generate a slope base reference map, which is then uniformly registered. The registration relationship of occluded areas is transmitted using adjacent registered units to reduce errors.

Benefits of technology

It improves the spatial consistency and verifiability of detecting changes in ecological restoration status in multi-temporal slope images, reduces errors caused by changes in vegetation cover and illumination, and improves the accuracy of detection results.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122391249A_ABST
    Figure CN122391249A_ABST
Patent Text Reader

Abstract

The application discloses a kind of ecological restoration state change detection methods based on multi-temporal slope image registration, it is related to image processing technical field, including including following steps: S1, obtains multi-temporal slope image and determines slope monitoring area: the image of same slope under different acquisition time phases is regarded as input, and the same slope monitoring area is determined from each time phase image, the slope monitoring area is divided into position corresponding slope unit, so that the slope unit in different acquisition time phases has consistent space number, this step outputs multi-temporal slope unit image sequence, S2, image layering is carried out to each time phase slope image, the present application extracts each slope unit most clear and exposed substrate texture fragment from multi-temporal slope image, generates uniform slope substrate reference map, and each time phase slope image is registered to the slope substrate reference map and then change detection is carried out.
Need to check novelty before this filing date? Find Prior Art