Showing papers by "Masoud Nosrati published in 2007"

Using 2D wavelet and principal component analysis for personal identification based On 2D ear structure

Abstract: Ear structure as a new class of biometrics can be used in many applications such as security systems. Ear structure is physiologically unique and stable, so ear recognition can be a good choice for a biometric security system. Even though, using ear biometric is not customary but it can be used with other biometrics like face or fingerprint simultaneously to increase the reliability of a biometric security system. In this paper, regarding existence of important information in edges and high frequency points in ear structure, we apply 2D wavelet to the normalized image. By decomposing the image into three images (horizontal, vertical and diagonal) using wavelet, we find three independent features in three directions. We combine these decomposed images to reach the feature matrix. This allows considering the changes in the ear images from three basic directions simultaneously. We apply PCA on feature matrix to feature dimension reduction and classification. Our experience in using this approach for different images demonstrates the accuracy of 90.5% and 95.05% recognition rate for two sets of databases.

Online Farsi handwritten words recognition using a combination of 3 cascaded RBF neural networks

Abstract: In this paper, we propose a method for online Farsi handwritten words recognition. At first, words are broken to their sub-words. Each sub-word is made of some strokes. The sign of the sub-word is found from the positions and shapes of its sub-strokes. After that, we classify sub-words according to their signs. Some online features are extracted from the main-stroke after the preprocessing stage. Preprocessing contains operations such as dehooking, smoothing, normalization and boundary size equalization. A combination of 3 cascaded RBF neural networks are learned and used in hierarchical recognition system. The first RBF net divides sub-words into classes, while the second one subdivides each class into sub-classes. The third RBF network recognizes sub-words in each sub-class. In this paper, we use a 1000-sub-word database of the most frequently used Farsi words. The performance of the system in the first and the second RBF classifiers is 99.7% and 98.9% respectively. The rate of correct performance of the third RBF net is 82.46% making the total recognition rate of the system on the database 81.3%.

A Novel Approach for Fingerprint Singular Points Detection Using 2D-Wavelet

Abstract: The success of many methods in fingerprint identification strongly depends on the accurate detection of singular points on the fingerprints. In this paper we propose a new approach for detecting the singular points. The method is based on measuring the maximal disturbance for direction of ridges in fingerprint images. We do not use the directional field in neighborhood of points directly so our method is comparatively fast. For this purpose we used 2D wavelet to detect high frequency components in three directions: horizontal, vertical and diagonal. The results of our experiments on different fingerprint databases confirm the ability of approach for rotation invariant, fast and accurate detection of singular points.