Image matching via hybrid methods of computational intelligance basement

By the Development of technology, intelligent systems, human imitative inventions of machines have gained velocity. Throughout increase in virtual life today, the World has become a small screen fits. With this advanced security has become one of the indispensable elements. The security of the system developed with different algorithms and alternative formulation investigated. Only for his own personality, not changes and works with principles of the recognition of the distinctive features of physiological biometrics, vulnerabilities the companies resources that threaten, descriptors, such as forgetting the password used to Access the card loss etc. factors increased the demand for the use of biometric systems.In this thesis, for the efficiency of working systems based on image processing, understanding and application of ANN, Object Recognition Systems, the methods used in Template Matching, Theoretical foundations are discussed and picture frames with face detection algorithms have been developed and template matching algorithms were validated.In this study, MATLAB 7.10.0 (R2010a) program was used and Graphical User Interface (GUI) was designed. MATLAB code has been developed for the connection of the data base. Created a pre-processing step the various functions performed by Gabor wavelets for filtering, for the classification, Artificial Neural Networks (ANN) was used. To extract the feature vectors Gabor wavelets, for classification artificial neural networks. are used. In the desired area to see the detail of the designated areas, and to analyze the correlation equations of the same phase to be able to detect whether a new search algorithm by using the search and verification was carried out with the target area. Use the Graphical User Interface (GUI) to obtain the image. Network trained using the detected face is scanned and displayed within a frame. The template is the same as that shown in the image sought to be confirmed in phase correlation equations is done.Keywords: Artificial Neural Networks, Graphical User Interface, Gabor wavelet, Pattern Matching, Phase Correlation, Object Recognition.