The Hand Shape Recognition Method Research Based On Hidden Markov Models (HMM)

Nowadays,people attach more importance to technology secure with the rapid growth of information technology.Owing to these traditional Identity Recognition methods have obvious disadvantages.Biometric technology causes an unprecedented attention of the world because of its unique advantage.Biometrics is the most secure and convenient way to satisfy the requirements for identify digitalization and virtualization in the coming network society,which combines biology technology and information technology to exploit physical features in human body or behavioral features to identify a person.The Hand Shape Recognition as one mainstream technology is becoming another important biometric identity method with its simple acquisition devices,fast certification after the fingerprinting biometric identification method.At present in foreign counties,in the commodity market share of biometric identification,the Hand Shape Recognition is basically in the same level with fingerprint identification.But literatures and samples at home and abroad on Hand Shape Recognition are very few.The Hand Authentication Technology is still in the research and development stage.In this paper,I mainly used cannon G9 digital camera to collect hand images and utilized the image processing,pattern recognition technologies and research methods comprehensively to achieve the image preprocessing,feature extraction,matching decision-making process and so on.I designed and completed a hand shape authentication system successfully and obtained good experimental results.In this paper,the important tasks and research results are as follows:1.This paper introduced hand-shaped sample collection methods particularly.In the image segmentation part,according to the characteristics of threshold segmentation,I put forward a image segmentation based on maximum between-cluster variance.First of all,we select a threshold,all the pixels divided into two categories,gray-scale images obtained overall mean,again in accordance with the pattem recognition theory, we can be obtained for these two types of between-cluster variance.To between-class variance as a measure of export of different types of threshold separation performance measurement criteria,Maximization between-class variance of the process is automatically determines the threshold process.2.The focus of testing uses a Methods of SUSAN operator.With other comer detection operator compared,SUSAN operator for edge detection and comer of the build-up does not require the calculation of differential,Moreover non-linear response characteristics of SUSAN operator helps to reduce the noise,another characterized of SUSAN operator is that on the edge of the response will be as smooth or fuzzy edge enhanced.3.In this paper,we have adopted a HMM-based authentication algorithm of the hand. We use the two characteristic parameters:Contour point to the centroid of the radial distance & Contour points in the direction of a curvature.At this time we have structures based on the characteristics of HMM classifier,after we carried out certified hand.We adopted the use of data samples from laboratory.In this paper,we proposed a systematic method of performance tests to demonstrate the feasibility and effectiveness of the algorithm.4.We designed a self-identification system.It realized the image preprocessing,feature extraction and matching the function of certification.In the image acquisition stage,we use a digital camera as the acquisition of equipment.Images in the collection,we only require hands open as far as possible.This system is relatively simple,more humane,but also more easily accepted by users.We have created a small hand-shaped image library.We opponents of fractal image preprocessing,feature extraction and recognition algorithms have been effective verification by the small hand-shaped image library,and we have a specific test results.The experimental results show that our algorithm used has a good robustness,feasibility and accuracy.