Image anomaly detection method in combination with CNN migration learning and SVDD

Abstract

The invention discloses an image anomaly detection method in combination with CNN migration learning and SVDD. The method comprises the steps of manually capturing images around a to-be-detected imageobject according to video data, making a to-be-detected pillar number data set, expressing image data depth features by utilizing a CNN, fully extracting features of pillar number samples through pre-trained weight and parameter network models, and solving the problem of minority class data in unbalanced data; and constructing a positive sample feature set which needs to participate in training in a classifier, finally performing parameter optimization by utilizing an SVDD algorithm, grid search and the like, forming a normal domain of positive sample feature training, and realizing identification of a number state of a contact network through a boundary. The automated processing level is relatively high, so that the workload of operators can be greatly reduced; and the problem of pillarnumber anomaly of the contact network is discovered early, so that the inspection efficiency is improved.

Description

technical field [0001] The invention relates to the field of state detection of the numbering of high-speed railway catenary pillars, in particular to an image anomaly detection method combined with CNN migration learning and SVDD. Background technique [0002] As an important part of the train, the catenary system of high-speed railway has a vital impact on the safety of the train. The status detection of the catenary pillar number is an important work step before the identification of the catenary pillar number. The abnormal detection of the catenary pillar number can not only quickly determine the road safety information, but also provide great convenience for the management of high-speed railways. [0003] The traditional method of anomaly detection through image and pattern recognition is to determine and extract corresponding features through a fixed algorithm adapted to the detection object or manually, and then perform anomaly detection through an anomaly detection a...

AI Technical Summary

Problems solved by technology

[0004] The technical problem to be solved by the present invention is to provide an image anomaly detection method combining CNN (convolutional neural network) transfer learning and SVDD (Support Vector Data Description, support vector data description), to solve the problem of extracting images in existing anomaly detection methods. The feature information is inaccurate, and the technical problems of unbalanced data cannot be well solved, and the abnormal detection of high-speed train catenary images is effectively realized

Images