Tunnel security detection method and system based on inspection robot

Abstract

The invention discloses a tunnel security detection method and system based on an inspection robot. The method comprises the steps of obtaining visible light image information in a tunnel in real time, obtaining a contour image of a flame based on the image information, extracting a fractal dimension of the contour image, comparing the fractal dimension with a preset threshold, and judging whether a fire happens or not; and measuring the reflectance of the surface of a detection point and the distance value between the inspection robot and the ground, calculating the water level height of the corresponding detection point according to the distance value, comparing the reflectance and the water level height with corresponding set thresholds respectively, and determining whether to give a water level alarm or not. Whether accumulated water exists in the tunnel or not is measured through a laser scanner and an ultrasonic ranging method, direct contact between detection equipment and water is avoided, and better reliability is achieved; and besides, the detection means based on the laser principle and the ultrasonic principle is utilized, the misjudgment risk caused by a single detection means is reduced, and the ponding detection accuracy is improved.

Description

technical field [0001] The invention relates to the technical field of power system tunnel safety detection, in particular to a tunnel safety detection method and system based on an inspection robot. Background technique [0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art. [0003] As the main carrier of electric power transportation, urban cables are becoming more and more important. Therefore, the daily inspection and maintenance of cable tunnels are becoming more and more important. [0004] Due to the complex environment in the cable tunnel, there are safety hazards such as fire, water accumulation, and harmful gases. Using robot inspections instead of manual inspections has greatly improved the efficiency of inspections in the tunnel and reduced the workload of manual inspections. [0005] At present, tunnel security detection mainly includes personnel intrusion detec...

AI Technical Summary

Problems solved by technology

However, there are two shortcomings. One is that when there is water accumulation, the device itself must be partially or even completely immersed in water. Some old equipment has reduced waterproof performance and is easy to be damaged. On the other hand, since the detection point is only deployed in places prone to water accumulation Area, the detection effect is limited, it is impossible to estimate the area of ​​water accumulation on the ground, the scope of influence, and how long it will take to discharge the accumulated water, and it is impossible to have a visual perception of the accumulation of water, and it is necessary to manually confirm the water situation on the spot

[0008] (3) The damage to the cable sheath is mainly based on the vibrating optical fiber to determine whether the cable is vibrating, but it is impossible to determine whether the cable has been damaged and to confirm the damage.

[0009] (4) The existing security detection system currently does not conduct comprehensive detection of explosion-proof gas, but only monitors a single gas. Under a single condition, it is impossible to make comprehensive predictions and judgments on the danger level of the gas in the tunnel, and it is impossible to mark the danger range.

[0010] (5) The current flame recognition method based on the deep learning method mainly extracts the color of the flame as a feature through the color model. Objects with similar colors are poorly distinguishable and cannot accurately identify the flame.

Images