A method for predicting the anti-skid ability of airport asphalt pavement and a method for constructing a prediction system

By building an airport asphalt pavement anti-skid ability prediction system based on deep learning image recognition, the problem of rapid and accurate evaluation of the anti-skid performance of airport asphalt pavements in existing technologies has been solved, and real-time and accurate evaluation of the anti-skid performance has been achieved, ensuring the safety of airport operations.

CN114842345BActive Publication Date: 2025-09-09SHANDONG HI-SPEED ROAD & BRIDGE INT ENG CO LTD +1
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202210569735.0
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-05-24
Publication Date
2025-09-09
Estimated Expiration
2042-05-24

AI Technical Summary

Technical Problem

Existing technologies make it difficult to quickly and accurately evaluate the anti-skid performance of airport asphalt pavements, leading to traffic accidents.

Method used

An airport asphalt pavement anti-skid ability prediction system based on deep learning image recognition is adopted. By collecting airport asphalt pavement images under different weather conditions, preprocessing and segmenting them, and using the ApRdAntiSkidNet model for training and prediction, accurate classification of anti-skid performance is achieved.

Benefits of technology

It realizes real-time and accurate evaluation of the anti-skid performance of the airport's asphalt pavement, and can provide timely feedback to the unmanned driving system or relevant departments to guide vehicle speed adjustment or pavement adjustment to ensure the safety of airport operations.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114842345B_ABST
    Figure CN114842345B_ABST
Patent Text Reader

Abstract

The present invention proposes a method for predicting the anti-skid ability of airport asphalt pavement and a method for constructing a prediction system used in the method. First, a large number of photos of airport pavements under different weather conditions and the anti-skid levels of the pavements are collected, and the ApRdAntiSkidNet network constructed by the present invention is used for training to obtain an optimal image recognition model. After that, it is only necessary to collect photos of the airport pavement to use the ApRdAntiSkidNet image recognition model to predict the anti-skid ability of the pavement in real time, and to promptly feed back the anti-skid ability to the unmanned driving system or relevant departments, so as to make corresponding restrictions on the aircraft speed or adjust the pavement, thereby ensuring the operational safety of the airport. The ApRdAntiSkidNet network architecture uses a hierarchical construction method similar to that of convolutional neural networks. For images of airport asphalt pavements, downsampling at different scales and magnifications is adopted, and the texture and structural features of the asphalt pavement can be extracted at different scales, which is ultimately conducive to the accurate classification of anti-skid ability.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • An asphalt pavement antiskid performance detection method and system

    CN109671077A

  • Image processing method, image processing device and storage medium

    CN114359092A