Unlock instant, AI-driven research and patent intelligence for your innovation.

A Face Recognition Method Based on Hidden Markov Model

A Hidden Markov and face recognition technology, which is applied in the field of statistical method face recognition, can solve the problems of large number of sequence calculations, slow recognition speed, low efficiency, etc., and achieve the effect of reducing the amount of calculation

Inactive Publication Date: 2016-02-10
HOHAI UNIV
View PDF2 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

To recognize a face, a sequence must be used, resulting in too many sequence calculations, and the recognition speed is slow and inefficient

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • A Face Recognition Method Based on Hidden Markov Model
  • A Face Recognition Method Based on Hidden Markov Model
  • A Face Recognition Method Based on Hidden Markov Model

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0043] Below in conjunction with accompanying drawing, the present invention is described in further detail:

[0044]The structure of the one-dimensional human face hidden Markov model used in the present invention, such as figure 1 As shown, the hair, forehead, eyes, nose, and mouth regions are taken as hidden states, and each state can only be transferred to itself and the next state, and the initial state is the hair state.

[0045] The EHMM structure that the present invention uses, as figure 2 As shown in , a one-dimensional state is used as a super state, and the super state is internally expanded into a sub-HMM in the horizontal direction. The state of each sub-HMM can be transferred to itself or the next sub-state, and the last state of the sub-HMM can be transferred to the next super-state in the super-HMM. The starting state is the first horizontal substate in the hair state.

[0046] The present invention improves the face recognition process based on the Hidden...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention relates to a face recognition method based on a hidden markov model. The method specifically includes the following steps: A face hidden markov model is established. Samples of the sample images are taken and the feature vector is obtained, then the initial parameters for the models are determined. The hidden markov model is trained repeatedly and iteratively so that each face can create a hidden markov model with different parameters. The sample of the ready image recognition is taken and the feature vector is got. The feature vector of the ready image recognition is segmented, and the maximum similarity on all hidden markov models is progressively calculated, and the hidden markov modes with a minimum similarity are ruled out. At last, the recognition results are got. The implicit state which is determined by the structure of human face hidden markov model (HMM), always starts from the state of the head and only transfers to the next state and the speciality of viterbi algorithm of dynamic programming, the observation vector is segmented, and the face models with little possibilities are eliminated through the middle results while calculating the similarity. In this way, the effect of reducing calculated amount is achieved.

Description

technical field [0001] The invention relates to an observation sequence segmentation method, which is used for Hidden Markov (Hidden Markov Model, HMM) face recognition, and belongs to the field of statistical method face recognition. Background technique [0002] As an important research direction of image recognition, face recognition technology is a research hotspot in current biometric recognition technology. Compared with the recognition of other biometric features, face images can be more intuitive and convenient to verify people's identities, and are the most common patterns in human vision. Therefore, face recognition identification methods have great potential in the fields of business, security, and law. It has broad application prospects and has become one of the most successful technologies in computer vision, image analysis and understanding. [0003] Because face recognition is affected by environmental factors such as illumination, posture, angle, etc., its r...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G06K9/00G06K9/62
Inventor 刘惠义王志超周斌秦川
Owner HOHAI UNIV