Pipe gallery video intelligent attribute detection method and system based on cloud processing

Abstract

The invention discloses a pipe gallery video intelligent attribute detection method and system based on cloud processing. The method comprises the following steps: marking various attribute learning sample sets of a pipe gallery; constructing a deep neural network, and training the neural network by using the data of the sample set; using the trained network model to carry out attribute detection on a newly obtained to-be-identified image; using a confidence threshold to screen attribute detection results, and eliminating fuzzy target interference; using a support as a background landmark object for pipeline positioning, wherein a positioning segmentation line can be adaptively adjusted along with the movement of a camera; and performing time domain factor limitation, reducing camera jitter interference, and weakening the influence of error detection results. The pipe gallery video can be effectively identified and analyzed, pipe gallery attribute detection can be rapidly carried out, and the efficiency of daily operation, maintenance and management of the pipe gallery is improved.

Description

technical field [0001] The invention belongs to the technical field of computer digital image processing and pattern recognition, relates to the field of intelligent analysis of pipe gallery video, and in particular to a method and system for detecting intelligent attribute of pipe gallery video based on cloud processing. Background technique [0002] The comprehensive pipe gallery (also known as "common ditch" or "common pipeline") is an underground urban pipeline comprehensive corridor, that is, a tunnel space is built underground in the city, and various engineering pipelines such as power, communication, gas, heating, water supply and drainage are integrated It is an important infrastructure and "lifeline" to ensure the operation of the city. The construction of the pipe gallery is of great significance. In the daily maintenance of the pipe gallery, it mainly relies on manual inspection or robot inspection to observe and monitor video analysis and judgment. When the con...

AI Technical Summary

Problems solved by technology

[0003] In the traditional implementation process, the management of the attributes of the pipelines in the utility corridor mainly relies on the manual judgment of the workers during the inspection process, and the statistics of the number of optical cables in the utility corridor are reported to the data center for unified management. However, this method exists The following problems: 1. Low efficiency. In order to check the number of cables in a pipe gallery, workers need to walk through the entire pipe gallery and record at every fork in the pipe gallery; 2. The accuracy rate is low. The number will not change suddenly, but it does not rule out that a certain pipeline is broken or stolen. According to the traditional method of only counting at each fork, this situation cannot be identified, which will cause great trouble to operation and maintenance.

Images