The Study On Human Cranio-maxillo-facial Multi-feature Information Extraction Technology Based On RBF Network

The biometric feature recognition technology is a secure personal authenticationtechnology. Now the fingerprint recognition and iris recognition have also been applied inactual. Biometric identification need to meet the universality, unique, invariant, can becollected and easily accepting. But in real life, it’s difficult to find a biological object tomeet the entire above all of the conditions at the same time. Each biometric technology hassome defect. In the future, the wide range of applications of biometric technology must beseveral comprehensive application of biometric identification technology. This article setthe cranio-maxillo-facial as the main object and extract the cranio-maxillo-facial multiplecharacteristics as the basis of different individuals. And for this purpose, the articles studythe key extraction technology and algorithm-depth in the identification process.The article extraction and analysis cranio-maxillo-facial shape characteristics, texturecharacteristics and physiological characteristics. The article use the same moment as theprimary means to express cranio-maxillo-facial shape feature and use wavelet packetdecomposition texture features. At the same time, the Delaire cephalometric analysisprovide the basis for cranio-maxillo-facial physiological characteristics extraction, and theexperiment vivificates that three different features extracts algorithm in thecranio-maxillo-facial identification process have played a good role.This paper study of the fusion algorithm based on multiple characteristics of RBFneural network in detail, and trying to establish a cranio-maxillo-facial feature recognitionmodel. The paper focuses on the key issues encountered in the training process of RBFneural network algorithm analysis using matlab software fusion algorithm for thesimulation of diverse characteristics. Through simulation analysis, verify the feasibility ofthe RBF neural network feature fusion, and prove that in some extent this algorithm canreduce the dimensionality of the original data, and improve the operating efficiency of thealgorithm. The fusion features can express the cranio-maxillo-facial overall characteristicsmore complete, and the recognition result is also better than the individual features. Theexperiments show that RBF neural network-based multi-feature recognition technology isreliable and effective.