HOG and LBPH feature fused detection method for copy of identity card

Abstract

The invention discloses an HOG and LBPH feature fused detection method for a copy of an identity card. The method comprises that a lot of identity card and non identity card pictures are selected as positive and negative samples of a training sample set, HOG and LBPH features are extracted from the training sample set, a SVM is trained to obtain first and second classifiers, the first classifier is used to carry out target detection on a test image, and the LBPH feature of a target detection result is obtained; and the second classifier makes determination according to the LBPH feature of thetarget detection result, and targets whose determination results are true are reserved. The HOG classifier is used for detection at first, the LBPH classifier is used to re-detect HOG detection results, the method is simple, rapid and efficient, and the detection accuracy is high.

Description

technical field [0001] The invention relates to the technical field of pattern recognition, in particular to an ID card copy detection method combining HOG and LBPH features. Background technique [0002] With the development of science, technology and economy, the workload of government enterprises and institutions is increasing, the scope of office activities is increasing, and the market-oriented requirements for office speed and accuracy are getting higher and higher. The identities available in the market However, there are relatively few certificate copy detection software. Most of the procedures for the detection of ID card copies are still completely manual, which causes a lot of waste of time, manpower and material resources, and human factors have a greater impact on the results. Most of the traditional ID card copy detection software only use a single feature for detection, and the detection accuracy is low. Both of these detection methods have obvious deficienc...

AI Technical Summary

Problems solved by technology

[0004] Aiming at the problem that the traditional manual detection of ID card copies or the detection of ID card copies based on only a single feature has low detection efficiency and detection accuracy, the present invention provides a method for detecting ID card copies that combines HOG and LBPH features , first perform HOG target detection, and then use LBPH and SVM technology to detect and classify again, which can accurately and quickly judge whether the operation of the copy of the ID card is correct, and improve the detection efficiency and detection accuracy

Images