Real-time detection method and system for abnormal state of airport pavement

Abstract

The invention provides a real-time detection method for an abnormal state of an airfield pavement. The real-time detection method comprises the following steps: optimizing a trained YOLOv5 target detection model and a DDRNet semantic segmentation model based on a preset reasoning optimizer; the optimized YOLOv5 and DDRNet, a classical image processing algorithm and an image acceleration decoder are deployed on an image detection device of a movable body; controlling the movable inspection body to run according to a preset running path, and controlling the image acquisition device to acquire an airport pavement image; the image acceleration decoder acquires the acquired airport pavement image in real time and decodes the acquired airport pavement image; and detecting abnormal states in the decoded image by using YOLOv5, DDRNet and a classical image processing algorithm, wherein the abnormal states comprise pavement apparent diseases, pavement foreign matters and pavement marker line abnormal wear. The invention further provides a system for detecting the abnormal state of the airport pavement in real time. The abnormal state of the airport pavement can be found in real time.

Description

technical field [0001] The invention relates to the technical field of airport pavement detection, in particular to a real-time detection method and system for an abnormal state of an airport pavement. Background technique [0002] The airport pavement is an important facility of the airport. Under the influence of human factors, aircraft take-off and landing, and various climatic environments, incomplete marking lines, diseases and foreign objects on the airport pavement will affect the safe operation of the airport. [0003] At present, the airport mainly adopts manual visual inspection to inspect the pavement of the airport. The inspection process consists of one inspector driving the car and another staff member observing the surface condition of the pavement. However, this traditional manual inspection method has problems. Large amount, low efficiency, poor precision, easy to miss detection, strong subjectivity and other shortcomings. [0004] With the rapid growth of ...

AI Technical Summary

Problems solved by technology

[0003] At present, the airport mainly adopts manual visual inspection to inspect the pavement of the airport. The inspection process consists of one inspector driving the car and another staff member observing the surface condition of the pavement. However, this traditional manual inspection method has problems Disadvantages such as large quantity, low efficiency, poor precision, easy missed detection, strong subjectivity, etc.

[0004] With the rapid growth of air traffic volume, automatic inspection systems are used instead of manual inspections. Among them, radar detection technology and image processing technology are the current mainstream automatic inspection technologies, but the price of laser radar is relatively expensive, and it is difficult for airport pavement markings. The detection effect of the line is poor, and the detection accuracy is difficult to meet the detection requirements of small-sized cracks; in terms of image processing, the target detection algorithm and the semantic segmentation algorithm are constantly improving in accuracy, but the target detection algorithm cannot extract the contour information of the crack, so it cannot judge The severity of cracks; the semantic segmentation algorithm can extract the outline information of cracks, but the real-time detection is poor. In terms of classic image processing, the image processing process is artificially involved, but this processing method has poor environmental adaptability and cannot directly realize the airport pavement. Abnormal state detection

Images