Utilizing artificial intelligence to collect pavement surface condition data

Abstract EN

Nowadays, data collection mechanisms are developed from ancient methods, including visual surveying to modern methods, which rely on digital imaging devices and laser scanning.

This is expensive, delay of the pavement maintenance management and consuming of time.

In this paper, a robust method proposed for automotive detection of linear distresses from real life movies of Iraqi highways.

In the suggested method, common types of cracks, potholes and raveling are distinguished and collected automatically using high speed imaging techniques.

The data extracted using the suggested method can be used for selecting priority of management maintenance of country roads and making decisions for treatment or rehabilitation.