Long-distance pipeline inspection method based on YOLOv3 pruning network and deep learning defogging model

Abstract

The invention belongs to an image processing technology based on deep learning, and particularly relates to a long-distance pipeline inspection method based on a YOLOv3 pruning network and a deep learning defogging model. The method comprises the following steps: 1, constructing and training an AOD-Net defogging network model; 2, designing a YOLOv3 backbone network and a loss function; 3, performing image data acquisition and training on the target area in an unmanned aerial vehicle inspection mode; 4, performing compression and accelerated calculation on the YOLOv3 model through a scaling factor gamma pruning method based on a BN layer; 5, deploying the AOD-Net and YOLOv3 joint model to an embedded module of the unmanned aerial vehicle for target task detection; and 6, returning the inspection task detection result of the long-distance pipeline of the unmanned aerial vehicle to the background system in real time. The system is used for being deployed on an unmanned aerial vehicle embedded module to perform long-distance pipeline inspection work, and the labor cost is greatly reduced while high detection precision, good real-time performance and high efficiency are guaranteed.

Description

technical field [0001] The invention belongs to an image processing technology based on deep learning, in particular to a long-distance pipeline inspection method based on a YOLOv3 pruning network and a deep learning dehazing model. Background technique [0002] Pipeline transportation, as another major transportation mode in my country after road, railway, sea and air transportation, plays an important role in national economic construction. Long-distance pipelines inevitably cross complex natural and geographical environments during the transmission process. Due to the impact of climate and natural disasters, pipelines are often damaged to varying degrees and pipeline leakage accidents occur. In severe cases, they even threaten the safety of people's lives and property. Therefore, fast and efficient fault detection of long-distance pipelines is one of the important procedures of pipeline inspection work. Judging from the research and application of various units in recent...

AI Technical Summary

Problems solved by technology

[0003] (1) Weather and geographic location cause poor image quality

[0004] Smog and bad weather will lead to a serious decline in image quality, weaken image feature extraction, and have a negative impact on the accuracy of subsequent UAV aerial photography target detection. The quality of the outdoor image is low, which eventually leads to low efficiency of task target discrimination and greatly reduced recognition accuracy.

[0005] (2) High labor cost and low efficiency

This lack of intelligent auxiliary means of detection is not only time-consuming, laborious and inefficient, but also manual detection may cause missed detection and false detection as the energy of the staff decreases.

[0007] (3) Poor real-time detection

[0008] The analysis of a large amount of image data collected by drone inspections is mostly offline processing in the background, and online preliminary diagnosis cannot be performed. The real-time performance is poor, which leads to protracted troubleshooting time and cannot effectively ensure the safe transportation of long-distance pipelines.

[0009] (4) Deployment of deep learning models on mobile devices is limited

Images