Road underground hidden danger detection method and system based on radar image and artificial intelligence

Abstract

The invention provides a road underground hidden danger detection method and system based on radar images and artificial intelligence. The road underground hidden danger detection method comprises the steps of S1, screening the B-SCAN sample pictures containing underground hidden danger targets from a three-dimensional ground penetrating radar database; S2, performing target detection marking on each target of the sample picture, and performing data enhancement processing to form an underground hidden danger target detection data set; S3, respectively training the two R-CNN target detection neural networks to obtain two target detection models with cavity, void, pipeline and manhole object detection capabilities; S4, reading the to-be-detected B-SCAN pictures of each channel acquired by the three-dimensional ground penetrating radar, performing multi-GPU parallel target detection by using the two target detection models, and generating two groups of reasoning results; and S5, fusing and outputting the two groups of reasoning results through model integration. The method and the system are high in anti-interference performance, rich in target information and high in accuracy.

Description

technical field [0001] The invention relates to a hidden danger detection method, in particular to a road underground hidden danger detection method based on radar images and artificial intelligence, and also relates to a road underground hidden danger detection system using the road underground hidden danger detection method based on radar images and artificial intelligence. Background technique [0002] The huge road collapse safety accidents caused by underground hidden dangers of urban roads such as hollows or voids can cause damage to vehicles and property in the slightest, and cause casualties in severe cases, and the trend is increasing year by year. Due to the hidden and sudden characteristics of underground hidden dangers, the current competent business departments are in a passive situation struggling to deal with post-event emergency rescue. Therefore, new rapid and non-destructive large-scale census technologies are urgently needed to change passive to active, and...

AI Technical Summary

Problems solved by technology

[0006] The scope of use is limited, and it is only suitable for use on road sections with good geological conditions such as expressways. For typical scenarios such as municipal roads with the greatest hazards and demands, old residential areas, and key areas, because the roads are covered with manholes, pipelines, pipe galleries, Abnormal objects such as crushed stones and layer interfaces, and road interference objects such as various bridges and culverts, light poles around the road, electric wires, gantry frames, and motor vehicle guardrails, etc. There are often multiple abnormal objects in a segment that interfere with each other, resulting in virtual Too many alerts and too low a recall to be useful

[0007] Without target information, the detected underground hazards usually need to be further processed according to size, depth and category (site survey, 2D radar re-inspection, manhole cover until drilling verification, etc.), and the method based on image classification cannot give the target Relevant information such as position, depth, and size need to be analyzed manually by interpreters. In a large amount of repetitive work, not only may wrong labeling or missing labeling be possible, but also because a composite map corresponds to a larger cube (for example, 10m x 1.8 m x 2m size), will add positioning error, often miss the real target

[0008] It is difficult to continue to improve: the problems that arise are due to the limitation of the principle of image classification, and the effect of continuous improvement is very small. Only by innovating from the technical route according to the actual situation can it meet the needs of engineering practice

Images