基于衬砌表面图像识别的隧道冻结程度预测方法
By constructing a reflection change sequence and analyzing texture details, the water film stage and the actual freezing growth stage are separated, and a freezing level progression curve is generated. This solves the problem of misjudging thin water films and achieves accurate prediction and stable control of the freezing degree.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- CHONGQING UNIV
- Filing Date
- 2026-04-14
- Publication Date
- 2026-07-17
AI Technical Summary
In existing technologies, tunnel freezing degree prediction based on lining surface image recognition is prone to misjudging as ice layers due to the specular reflection of thin water films, causing heating devices to start prematurely, affecting temperature control rhythm and increasing energy consumption.
By constructing a reflection change sequence, extracting the brightness growth rate and reflection direction fluctuation amplitude, stripping short-term fluctuation segments, and combining the texture detail decay rate, the water film stage and the real freezing growth stage are separated, generating a freezing level progression curve and triggering control actions with a set duration.
It improves the accuracy of freeze detection, avoids misjudgments caused by specular reflection, ensures stable operation of temperature control, and reduces unnecessary energy consumption.
Smart Images
Figure CN122049819B_ABST
Abstract
Citation Information
Patent Citations
Apparatus and method for detecting road surface properties
CN103134777A
Sludge treatment automatic monitoring method based on computer vision
CN121392745A