Freight wagon blocking key missing fault identification method based on support vector machine

Abstract

The invention discloses a freight wagon blocking key missing fault identification method based on a support vector machine, and belongs to the field of freight wagon running fault testing. The freight wagon blocking key missing fault identification method includes a training process of the support vector machine and a process of online blocking key missing identification of the support vector machine. With the freight wagon blocking key missing fault identification method based on the support vector machine, when a train passes through, images of a running freight wagon body, freight wagon type and freight wagon information can be acquired; whether the images include the blocking key component or not can be confirmed through image naming; coarse positioning of a blocking key can be completed through positioning shaft ends so as to further complete fin positioning of the blocking key;HOG (Histogram of Oriented Gradients) features of blocking key images are extracted, and blocking key identification classifiers are inserted to complete blocking key missing fault identification. With the freight wagon blocking key missing fault identification method, training samples can be conveniently added, and corresponding classifiers can be trained based on different detection stations. The freight wagon blocking key missing fault identification method can be run automatically without mannual interference, running parameter can be controlled through configuration files; identification speed is fast, and automatically identified results can be stored in a friendly mode so as to facilitate verification and review of the identified results by train examination staff.

Description

technical field [0001] The invention belongs to the field of railway freight car running fault detection, and relates to a fault identification method for a railway freight car gear loss failure, in particular to a support vector machine-based fault identification method for a railway freight car gear loss loss fault. Background technique [0002] The trackside image detection system for railway wagon faults is a system that integrates high-speed digital image acquisition, large-capacity image real-time processing technology, precise positioning technology, network technology and automatic control technology, referred to as TFDS. [0003] At present, the TFDS system mainly performs fault identification on the collected truck body images by manual viewing, which has high labor costs and high labor intensity, and the effect of fault identification is restricted by subjective factors such as the mental state of the inspector and the inspection experience. The application of aut...

AI Technical Summary

Problems solved by technology

[0004] Image recognition technology has been widely and successfully applied in many fields such as face recognition, intelligent transportation, medical image processing, industrial product inspection, etc., but its application in the field of automatic recognition of truck fault images is relatively limited, and it is still in the preliminary research stage.

Images