Train part anomaly detection method and system, electronic equipment and storage medium

Abstract

The embodiment of the invention relates to the technical field of train anomaly detection, and discloses a train part anomaly detection method and system, electronic equipment and a storage medium. In the embodiment of the invention, the running image of the train to be detected can be firstly acquired, and the abnormal detection operation of the train parts is performed on the running image of the train to be detected through a preset abnormal detection model based on region division so as to obtain an abnormal detection result, the preset abnormal detection model is set to be that each target object in the train driving image is classified into different areas, wherein a hierarchical relationship exists between the areas. Obviously, according to the anomaly detection mode provided by the embodiment of the invention, the target object is subjected to attribution by taking the region as the unit, and then the anomaly detection operation of the train parts is performed by taking the region as the unit, so that the detection accuracy of anomaly detection is improved, and the technical problem that the anomaly detection accuracy is relatively poor is solved.

Description

technical field [0001] The invention relates to the technical field of train anomaly detection technology, in particular to a method, system, electronic equipment and storage medium for abnormal detection of train components. Background technique [0002] With the continuous increase of train mileage, the technology of train manufacturing and operation is also constantly improving. [0003] In order to improve the safety of train operation, periodic anomaly detection is generally performed on the operation status of the train. [0004] As far as the operation of abnormality detection is concerned, computer vision-based abnormality detection is an important method for abnormal identification of train parts and components. The identification of components has the characteristics of high efficiency and non-invasiveness. [0005] In terms of computer vision-based anomaly detection of train components, some anomaly detection methods have been proposed for certain specific types...

AI Technical Summary

Problems solved by technology

[0010] However, the above-mentioned anomaly detection methods still have the problem of poor accuracy in the final detection effect: the core of traditional image processing methods lies in template matching, which requires a large number of normal train images in the historical template library, and needs to cover Various scenarios to deal with diverse external environments and train situations

This is good for qualitative detection of missing and deformation of large parts, but the effect of accurate detection is not ideal, and it is easily affected by the environment, and the generalization is not strong

Images