In this paper, a novel method for palmprint recognition, called Fisherpalms, is proposed. In this method, each pixel of a palmprint image is considered as a coordinate in a high-dimensional image space. A linear projection based on Fisher s linear discriminant is used to project palmprints from this high-dimensional original palmprint space to a significantly lower dimensional feature space (Fisherpalm space), in which the palmprints from the different palms can be discriminated much more efficiently. The relationship between the recognition accuracy and the resolution of the palmprint image is also investigated. The experimental results show that, in the proposed method, the palmprint images with resolution 32-32 are optimal for medium security biometric systems while those with resolution 64 64 are optimal for high security biometric systems. High accuracies (>99%) have been obtained by the proposed method and the speed of this method (responding time≤0.4s) is rapid enough for real-time palmprint recognition.

The palm print is a new and emerging biometric feature for personal recognition. The stable line features or "palm lines, "which are comprised of principal lines and wrinkles, can be used to clearly describe a palm print and can be extracted in low-resolution images. This paper presents a novel approach to palm line extraction and matching for use in personal authentication. To extract palm lines, a set of directional line detectors is devised, and then these detectors are used to extract these lines in different directions. To avoid losing the details of the palm line structure, these irregular lines are represented using their chain code. To match palm lines, a matching score is defined between two palm prints according to the points of their palm lines. The experimental results show that the proposed approach can effectively discriminate between palm prints even when the palm prints are dirty. The storage and speed of theproposed approach can satisfy the requirements of a real-timebiometric system.

This paper presents a novel approach of palmprint authentication based on the fusion of the phase and orientation information. This approach is an improvement of a previous palmprint recognition method-FusionCode method [1]. In the proposed approach, the phase information (Fusion Code) of a palmprint is extracted by using four 2-D Gabor filters with different orientations, and at the same time, the orientation information (called OrientationCode) of the palmprint is also extracted. The FusionCode and the OrientationCode are fused to make a new feature, called the Palmprint Phase Orientation Code (PPOC). At the matching stage, a modified Hamming distance is defined to measure the similarity of two PPOCs. This approach is tested on a palmprint database containing 7605 samples and the experimental results show that the PPOC approach greatly improves the performance of the FusionCode method.

According to the fact that the basic features of a palmprint, including principal lines, wrinkles and ridges, have different resolutions, in this paper we analyze palmprints using a multi-resolution method and define a novel palmprint feature, which called wavelet energy feature (WEF), based on the wavelet transform. WEF can reflect the wavelet energy distribution of the principal lines, wrinkles and ridges in different directions at different resolutions (scales), thus it can efficiently characterize palmprints. This paper also analyses the discriminabilities of each level WEF and, according to these discriminabilities, chooses a suitable weight for each level to compute the weighted city block distance for recognition. The experimental results show that the order of the discriminabilities of each level WEF, from strong to weak, is the 4th, 3rd, 5th, 2nd and 1st level. It also shows that WEF is robust to some extent in rotation and translation of the images. Accuracies of 99.24% and 99.45% bave been obtained in palmprint verification and palmprint identification, respectively. These results demonstrate the power of the proposed approach.

作为一种较新的生物特征，掌纹可用来进行人的身份识别。在用于身份识别的诸多特征中，掌纹线，包括主线和皱褶，是最重要的特征之一。本文为掌纹识别提出一种有效的掌纹线特征的表示和匹配方法。该方法定义了一个矢量来表示一个掌纹上的线特征，该矢量称为线特征矢量（line feature vector，简称LFV）。线特征矢量是用掌纹线上

Palmprint is a novel biometric method to identify a person. Generally, there are two types of features in palmprint, i. e. structural features and statistical features. Structural features, such as lines, can characterize a palm exactly, but are difficult to be extracted and represented. Contrarily, statistical features can be extracted and represented easily, but are unable to reflect the structural information of a palmprint. The fact that the principal features of both Chinese character and palmprint are lines motivates us to try some methods of Chinese character recognition to identify palmprint. In this paper, we use the idea of an efficient Chinese character recognition method, directional element feature (DEF), to define a novel palmprint feature, named fuzzy directional element energy feature (FDEEF) which is a statistical feature containing some line structural information about palmprints. It can be extracted and represented easily and, at the same time, has a strong ability to distinguish palms. Two other lowdimensional features: global fuzzy directional element energy feature (GFDEEF) and block edge energy feature (BEEF) are also derived from FDEEF in this paper. The experimental results demonstrate the power of this method.

Palm-lines, including the principal lines and wrinkles, can describe a palmprint clearly. This paper presents a novel approach of line feature extraction for palmprint recognition called the directional line energy feature (DLEF). The directional lines in different directions are first extracted using a set of directional line detectors. Then each directional line magnitude image is divided into several overlapped small grids and the magnitudes of the line points in these grids are used to compute the DLEF. A template-matching method based on Euclidean distance is adopted to measure the similarity of two DLEFs. Best results have been obtained when DLEFs with 6 different directions were employed. Accuracies of 97.92% and 97.5% are obtained by using the proposed approach in one-against-one matching and one-against-320 matching, respectively.

This paper presents a novel approach for palmprint recognition based on the valley features. This approach uses the bothat operation to extract the valleys from a very low-resolution palm image in different directions to form the valley feature, and then define a matching score to measure the similarity of the valley features. The experimental results shows that the proposed approach can effectively discriminate palmprints and can obtain about 98% accuracy in palmprint verification.

Palmprint is a new biomentric method to recoginze a person. The features in a palmprint include principal lines, wrinkles and ridges, etc. Line strcture feature, which includes princlpal lines and wrinkes, is one of the most popular methods in palmprint recognition. However, the line structure feature does not contain the thickness and width information of principal lines and wrinkles, which are very important to discrimiate paimprints. Ridges are very important to discrimiante paimprints. Ridges are not included in the structure feature either. So these methods cannot distinguish different palmprints with similar line structure. Furthermore, the line extraction is a difficult task. the fact that princlpal lines, wrinkles and ridges liave different resolutions monvates us to analyza the palmprint using multi-resolution analysts inethod. A novel palmprint feature, named wavelet energy features, Is defined employing waveiet, which is a powrful tool of multi-resolution analysis, in this paper. WEF can reffect the wavelet energy distribution of the principnl lines, weinktes and ridges in several different waveier decomposition level (scaie), so its ability to discriminate palms is very strong. Easiness to compute is another virtue of WEF. The very high recognition rats obtained in experiments shows the effect of the proposed method.

In this paper, a novel method for palmprint recognition is proposecL In this method, each pixel of a palmprint image is considered as a coordinate in a high-dimensional image space. A linear projection based on Fisher's linear discriminant (FLD) is used to project palmprints from this high-dimensional space to a significantly lower dimensional feature spuce, in which the ratio of the determinant of the between-class seatter to that of the within-class satter is maximized. High accuracies of 99% and 99.2% have been obtained in one-to-one matching and one-to-300 matching test respectively and the speed of this method is rapid enough for real-time palmprint recognition.