A structure icing / snowing synchronous classification monitoring system and method based on energy sensing and computer vision
By using a monitoring system that combines energy sensing multiplexing and computer vision with temperature difference and wind power generation, synchronous classification and intelligent early warning of icing/snow accumulation are achieved. This solves the problems of high cost, high power consumption and inaccurate monitoring of existing monitoring equipment, and is suitable for all-weather monitoring in frigid and snowy regions.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- SOUTHEAST UNIV
- Filing Date
- 2026-04-14
- Publication Date
- 2026-07-10
AI Technical Summary
Existing monitoring equipment suffers from high cost, high power consumption, single function, easy failure, and inaccurate monitoring in frigid and snowy regions, making it difficult to achieve unattended operation around the clock and simultaneous classification of icing/snow accumulation.
A monitoring system based on energy sensing reuse and computer vision is adopted, which combines temperature difference and wind power generation. Through composite energy acquisition and feature reverse perception, deep learning image feature fusion is used to achieve icing/snow accumulation classification, and load monitoring and data fusion are combined to achieve intelligent early warning.
It achieves low-cost, low-power, all-weather monitoring, improves monitoring accuracy, is suitable for distributed building clusters, has self-powered capabilities, and features intelligent early warning functions.
Smart Images

Figure QLYQS_1