Research On Palmprint Recognition System Based On Line Features By Bi-Directional Matching

Biometrics, which refers to the identification of a person by certain biological features such as fingerprint, iris, face and palmprint or behavioral features.It can be used to reinforce or even replace the traditional identification tools. It is becoming a potential technology in this era.Palmprint identification is a new branch of biometric technology. Since palmprint has the uniqueness and stability as fingerprint, the personal authentication can be determined completely by the line ,geometry and texture features. The identification system based on palmprint technology is friendly and ecommerce. Palmprint identification technology will become a new research and application field for personal authentication .It is also the important supplement of the identification technology based on the fingerprint.However, how to develop an automated recognition system for identification and verification of a large collection of palmprint image data with accuracy and reliability remains a challenging task. in general, the design of an automated biometric recognition system involves image data collection, image preprocessing, feature extraction, feature matching.This paper concentrates on the research of the key technologies and algorithms in palmprint identification system and we developed a palmprint recognition system which based on the palmprint line features. In detail, this paper concentrates on the research of the following key technologies and algorithms: 1.Palmprint preprocessing, alignment, and segmentation:Generally, noises always exist in palmprint images. Besides, different palms have different sizes, and because of the translational and rotational placements of hands, the collected palmprint images from the same palms are different from time to time. Therefore, it is essential to process the palmprint images including alignment, normalization and segmentation before feature extraction.In this paper, a new algorithm for palmprint image alignment and segmentations is been used. We select the largest inscribed circle which is in the palm as the ROI. First, we designed algorithm to segment the hand image from the background according to the characteristic that the background of the captured image is single black, and obtained binary image directly with the algorithm. Then we use the Susan operator to detect the corner. At last we select the largest inscribed circle as the ROI. The experiments show that this method has low computational complexity, high accuracy.With this method, we can get enough information from palmprint; in another aspect, as the sizes are different between two persons, the radius of ROI can be selected as a feature for palmprint matching.2.Palm-line extracting and construction of feature vector:Feature extraction is to quantify the nature of images, It is the key problem to the pattern recognition. Palm line feature is one of the most important feature in the hand. Besides the line feature express simple and the space of line feature is small which can fast retrieval and matching.This paper analyzed palmprint image and use the canny edge detection operator to get the palm line . Then, improved Hilditch algorithm is used. An approach for getting rid of twig which brought forward by this dissertation is applied; finally, palm-line image of single pixel is obtained. By this operation, most of principle line can be detected correctly. In order to express palm line accurately ,the endpoint of the palm line is used to represent a line. The feature vector of palmprint is obtained by combining the radius of ROI and principle line feature.3. Design a matching algorithm:Classification is the key step to speed up the retrieval process and improve the recognition rate. In this paper, a two-stage method of geometrical feature matching and line feature matching is proposed. First we classified the palmprint to several group by the number of the palm line. In each group, According to the radius of ROI, we can extract the classes that have similar feature with the sample image. Then we start accurate matching using the line feature.4. An Online Palmprint Recognition System:A palmprint recognition system is developed with functions of ROI alignment, features extracting, user enrollment and user recognition. The performance of the system is tested on our database. The experimental result indicate that the system achieves high recognition rate, proved that the method is effective and available.