Palmprint Images Research Articles

Towards a robust palmprint representation for person identification

Biometrics, which refers to automatic identification of individuals based on their physiological and/or behavioural characteristics, is a widely studied field. Among the physiological biometric modalities, those based on palm have received the most attention due to its steady and unique features, which are rich in information with a low resolution. Although there is several palm capture devices, however, no of them is apt to provide the full features of the same palm. By using different capture devices, the palm features can be represented with different formats such as: greyscale images, near-infrared images, colour images, multispectral images and 3D shapes. In this context, we present in this paper a study that permits to propose robust palmprint representation for a reliable person identification system. Thus, a comparative study of the used palmprint image representations in the practice is performed using several databases each one contains 400 users.

MobileNet Convolutional Neural Networks and Support Vector Machines for Palmprint Recognition

Due to the ease of handling with low resolution images, low cost hardware and high recognition accuracy, palmprint recognition has been attracting in the recent years. Many different features extracted of the palmprint image have been employed to perform the recognition tasks. Most of the recognition efforts consequently has been put on the designing and obtaining a relevant set of effective hand-crafted features which is considered to be the drawback of the traditional image based biometric recognition systems. To overcome the aforementioned problem, in this paper we explore the applicability of MobileNet V2 deep convolutional neural networks on palmprint recognition by fine-tuning a pretrained MobileNet neural networks. We also explore the performance of dropout support vector machines (SVM) by training them on the deep features of the similar pretrained networks. The experiments are carried out using datasets provided by Hong Kong Polytechnic University of Science and Technology. The dataset consists of 6000 grayscale images of 128x128 pixels from 500 different palms. It is demonstrated that the proposed schemes exhibit state-of-the-art performance on the datasets. The second scheme, MobileNet V2 based features with SVM classifier, is able to achieve best average testing and validation accuracy rate of 100% outperforming the best previous reported results.

Complete Binary Representation for 3-D Palmprint Recognition

3-D palmprint recognition has attracted a wide range of attentions due to its potential for civilian applications. Designing an effective representation is a key problem for the 3-D palmprint recognition. In this paper, we propose a complete binary representation (CBR) for the 3-D palmprint multiple-dimensional feature representation and recognition. First, we propose a multiple orientation binary representation (MOBR) to extract 2-D gray-level features of 3-D palmprint. Then, we propose a novel and effective compact surface type (CST) to characterize 3-D palmprint surface feature, and design a CST binary representation (CSTBR) to capture the surface consistency within a local patch. Finally, we develop a CBR method by integrating the MOBR and CSTBR, which can effectively represent both 2-D orientation-level and 3-D surface-level features of 3-D palmprint images. We conduct extensive intercomparison experiments to demonstrate the effectiveness of the proposed CBR method on a benchmark 3-D palmprint database. Also, we carry out multiple intracomparisons to validate the superiority of the MOBR- and CST-based representations.

Multi-orientation and multi-scale features discriminant learning for palmprint recognition

Palmprint contains stable and effective features, especially the orientation and scale features of palm lines, and has now become an important identity recognition technique for surveillance and safety applications. Existing palmprint recognition methods using texture features can be roughly divided into two categories: coding-based and local descriptor based methods. As compared with the latter category, the former one can make full use of the palmprint specific features and acquire fast matching speed. However, most existing coding-based methods are based on the competitive coding scheme, in which the scale features of palmprint cannot be well exploited. In this work, we propose a discriminant orientation and scale features learning (DOSFL) for palmprint recognition. By introducing the idea of discriminant analysis into palmprint coding, DOSFL can extract the orientation and scale features with more favorable discriminability. Then, DOSFL utilizes four code bits to represent both the orientation and scale features of palmprint, and employs the Hamming distance for code matching. To make better use of the orientation and scale information contained in palmprint samples, we further propose a multi-orientation and multi-scale features discriminant learning (MOSDL) approach for palmprint recognition, which can fuse different orientation and scale feature information effectively in the discriminant learning process. Experimental results on two publicly available palmprint databases, including the HK PolyU database and UST palmprint image database, demonstrate that our proposed approach can achieve better recognition results than the compared methods.

Textural and Geometrical Features Based Approach for Identification of Individuals Using Palmprint and Hand Shape Images from Multiple Multimodal Datasets

Abstract Identification and classification of biometrics are important research areas in the field of image processing and pattern recognition. Biometrics are the measurement and statistical analysis of physiological and behavioral characteristics of humans. A wide variety of biometric modalities are available, with unimodal biometrics suffering from several factors. The proposed research is novel because it uses a single image of a hand in order to extract a variety of unique characteristics, like hand shape and the palmprint associated with individual hands. Moreover, it obtains higher accuracy with minimum effort. We have chosen the multimodal biometrics, i.e., palmprint and hand shape, from three datasets, i.e., PolyU Palmprint Database, GPDS Hand Database, and the Bosphorus Hand Database, for a total of 1,072 images. There are 302 textural features found in the palmprint images, and 12 geometrical features are extracted from the hand images. Classification models include Naïve Bayes, Support Vector Machine (SVM), K-Nearest Neighbor (IBk), Decision Tree, Random Tree, Random Forest, and Bagging. The train and test method is used to evaluate the performance of different classifiers. It is observed that Naïve Bayes, SVM, IBk, and Random Tree models result in classification accuracy of 99.44 % with palmprint images using the 302 textural features over the combined dataset. After feature reduction, similar accuracy is achieved with the top ten, and even with the top five, features. For geometrical features, an accuracy of 99.81 % is achieved with the hand images using Naïve Bayes, SVM, IBk, and Random Tree.

Precision direction and compact surface type representation for 3D palmprint identification

Compared with its 2D counterpart, a 3D palmprint image contains not only the 3D structure-based but also the 2D texture-based features of the palmprint. In this paper, we propose a precision direction code and compact surface type (PDCST) method for 3D palmprint representation and identification. Specifically, we propose the precision direction code (PDC) to depict the 2D texture-based features by exploiting not only the visible but also the potential direction features of the palmprint. Moreover, we use a simple yet efficient compact surface type (CST) to represent the 3D structure-based features of the palmprint. We combine the PDC and CST forming the PDCST descriptor to represent the multiple level and multiple dimensional features of 3D palmprint images. The two-phase sparse representation scheme is used to perform PDCST-based feature identification. Extensive inter-comparative and intra-comparative experimental results on three widely used palmprint databases clearly demonstrate the effectiveness of the proposed method.

Palmprint recognition based on discriminant multiscale representation

Nowadays, important scientific works are oriented to develop optimal palmprint representations, especially used in the palmprint recognition field. These representations must be discriminant, robust, compact, and easy-to-implement in order to ensure the best performance in terms of accuracy, computation cost, and storage requirement. In this way, our paper presents a multiscale analysis framework for efficient palmprint representation called discriminant Gabor Lpq spatial pyramid histogram, which significantly relies on Gabor wavelet, local phase quantization (LPQ) descriptor, and spatial pyramid histogram (SPH) method. First, the Gabor wavelet function with two scales and four orientations is used to capture the local structure in palmprint images. Second, the LPQ operator is applied to the response images of Gabor filter to get label LPQ images in order to fully explore the blur invariant property and the texture information. This is conducted in multiscale and multidirection space. Third, for each of them, the SPH of vertical decomposition is applied to obtain local palmprint feature descriptors. Next, the obtained histograms are normalized. Then the global representation of the palmprint image is obtained by concatenating all the local feature descriptors. After that, the discriminant representation of the palmprint image is constructed using whitened linear discriminant analysis. Finally, the K-nearest neighbor classifier is used for identification. Experiments, conducted on three palmprint databases (PolyU2D, PolyU 2D/3D, and IITD), show that the proposed method provides a significant performance improvement compared to the state-of-the-art and recently proposed methods in terms of accuracy.

Local apparent and latent direction extraction for palmprint recognition

Direction information of the palmprint provides one of the most promising features for palmprint recognition. However, more existing direction-based methods only extract the surface direction features from raw palmprint images and ignore the informative latent direction feature of the convolution layer of palmprint images. In this paper, we propose a novel double-layer direction extraction method for palmprint recognition. The method first extracts the apparent direction from the surface layer of a palmprint. Then, it further exploits the latent direction features from the energy map layer of the apparent direction. Lastly, by using the multiplication and addition schemes, the apparent and latent direction features are pooled as the histogram feature descriptor for palmprint recognition. The proposed method achieves state-of-the-art performance on four benchmark palmprint databases, namely the PolyU, IITD, GPDS and CASIA palmprint databases. In particular, the latent energy direction feature shows a promising performance for noisy palmprint image recognition.

A new approach for palmprint image encryption based on hybrid chaotic maps

Biometric data has gain great interests as being precise authenticators. Saving and transmitting biometric data in the form of images over public networks needs a fast, lossless and secure cryptosystem for keeping biometric images details while securing it. In this paper, a novel encryption scheme is presented for securing palm print images based on hybrid chaotic maps. The proposed hybrid scheme is designed to fit the challenges of diffusion, confusion, and to provide large key space. The scheme applies a mixture of different chaos maps for a particular selection of the control parameters. Both the encryption and decryption processes are explained, formulated, and graphically presented. The results of security analysis verified the strengths of the proposed cryptosystem against statistical, brute force and differential attacks. The evaluated running time, for both encryption and decryption processes, guarantee that the cryptosystem can work effectively in real-time applications compared to current available techniques.

Palmprint recognition using a modified competitive code with distinctive extended neighbourhood

In recent years, palmprint recognition has made great progress and many methods have been put forward. The extraction of robust orientation features and finding efficient matching strategies are two key points for palmprint recognition. Traditional coding methods usually only use a dominant filter response to extract orientation features of palmprint images while not taking into account the other useful filter responses. Without increasing the number of filers, this study presents a modified Competitive Code to extract orientation features more accurately, which makes use of the relation between the filter responses. Besides, a distinctive extended eight‐pixel neighbourhood method is proposed to select the sample points for matching by extracting the local features. At the matching stage, an effective fusion matching scheme with a double‐layer image pyramid is designed to calculate the similarity between two palmprint images. Extensive experiments on four types of public palmprint databases show that the proposed method has excellent performance compared with the other state‐of‐the‐art algorithms.

An approach to improve Palmprint Recognition Accuracy by Using Different Region of Interest Methods with Local Binary Pattern Techniques

Objective: To extract the Region of Interest (ROI) of palmprint image by using appropriate methods and to improve the accuracy of palmprint recognition system. Methods/Statistical Analysis: This piece of work is primarily addressing the different mechanisms for extracting ROI area. The techniques like Competitive Hand Valley Detection (CHVD), and Euclidean Distance (ED) were applied as the part of pre-processing, while the Feature Extraction mechanism LBP was utilized to extract the texture feature from different type of ROIs of palmprint image. Findings: The experimental results showed that CHVD with LBP gave best result with high accuracy reached to 96.10534% and Equal Error Rate (EER) of 3.894661%, while in ED the best result showed accuracy reached to 88.23611% and EER of 11.76389%. Application/Improvements: The study mainly concentrated on developing palmprint authentication system with less EER and high accuracy. Keywords: Analysis, Competitive Hand Valley Detection (CHVD), Euclidean Distance, Local Binary Pattern (LBP), Palmprint Recognition

Hierarchical palmprint feature extraction and recognition based on multi‐wavelets and complex network

This study presents a hierarchical palmprint feature extraction and recognition approach based on multi-wavelet and complex network (CN) since they can effectively decrease redundant information and enhance key points of main lines and wrinkles. The palmprint is first pre-filtered and decomposed once using multi-wavelet. Three components (LL 1,2,3 ) corresponding to the pre-filter except for diagonal component are extracted as the elementary features. Second, binary images (BLL 1,2,3 ) are obtained by the average window method using different thresholds. Third, three series of dynamic evolution CN models (the 1st, 2nd, 3rd CNs) are constructed from global to local, which is based on the mosaiced images obtained from BLL 1,2,3 , BLL 1 and four equally divided sub-images of BLL 1 , respectively. Fourth, statistical features are extracted from complex networks, in which average degree and standard deviation of the degrees are extracted for the 1st CNs and average degrees are extracted for the 2nd and 3rd CNs. Fifth, the fisher feature is extracted using the linear discriminate analysis method. Finally, the nearest neighbourhood classifier is used to recognise palmprint. Based on the CASIA Palmprint Image Database, experimental results show that the proposed method can effectively recognise palmprint with good robustness and overcome the problem of small training samples number.

An approach to improve Palmprint Recognition Accuracy by Using Different Region of Interest Methods with Local Binary Pattern Techniques

Objective: To extract the Region of Interest (ROI) of palmprint image by using appropriate methods and to improve the accuracy of palmprint recognition system. Methods/Statistical Analysis: This piece of work is primarily addressing the different mechanisms for extracting ROI area. The techniques like Competitive Hand Valley Detection (CHVD), and Euclidean Distance (ED) were applied as the part of pre-processing, while the Feature Extraction mechanism LBP was utilized to extract the texture feature from different type of ROIs of palmprint image. Findings: The experimental results showed that CHVD with LBP gave best result with high accuracy reached to 96.10534% and Equal Error Rate (EER) of 3.894661%, while in ED the best result showed accuracy reached to 88.23611% and EER of 11.76389%. Application/Improvements: The study mainly concentrated on developing palmprint authentication system with less EER and high accuracy.

An Improved Multispectral Palmprint Recognition System Using Autoencoder with Regularized Extreme Learning Machine.

Multispectral palmprint recognition system (MPRS) is an essential technology for effective human identification and verification tasks. To improve the accuracy and performance of MPRS, a novel approach based on autoencoder (AE) and regularized extreme learning machine (RELM) is proposed in this paper. The proposed approach is intended to make the recognition faster by reducing the number of palmprint features without degrading the accuracy of classifier. To achieve this objective, first, the region of interest (ROI) from palmprint images is extracted by David Zhang's method. Second, an efficient normalized Gist (NGist) descriptor is used for palmprint feature extraction. Then, the dimensionality of extracted features is reduced using optimized AE. Finally, the reduced features are fed to the RELM for classification. A comprehensive set of experiments are conducted on the benchmark MS-PolyU dataset. The results were significantly high compared to the state-of-the-art approaches, and the robustness and efficiency of the proposed approach are revealed.

Highly secure palmprint based biometric template using fuzzy vault

SummaryBiometric cryptosystem is an integrated system of cryptography with biometric features which is considered to be a promising component of template security. This paper presents a novel method for biometric cryptosystem based on fuzzy vault using palmprint features. Initially, pre‐processing is performed to remove noise in palmprint enrolled image and to pre‐align image automatically. Bottom‐hat filtering is employed for the feature extraction from the palmprint images. A polynomial is evaluated from the secret key and this polynomial is used to generate projections from palmprint features. Fuzzy vault is created by combining projected points with randomly generated dummy or chaff points. During decoding process, matching scores are calculated by comparing the query feature and stored vault. The vault is successfully decoded when the matched scores should be greater than a predefined threshold value. Experiments were conducted on CASIA database. Simulation result demonstrates that the proposed method gives better performance than other methods and improves the robustness against some security attacks.

An Effective Palmprint Recognition Approach for Visible and Multispectral Sensor Images.

Among several palmprint feature extraction methods the HOG-based method is attractive and performs well against changes in illumination and shadowing of palmprint images. However, it still lacks the robustness to extract the palmprint features at different rotation angles. To solve this problem, this paper presents a hybrid feature extraction method, named HOG-SGF that combines the histogram of oriented gradients (HOG) with a steerable Gaussian filter (SGF) to develop an effective palmprint recognition approach. The approach starts by processing all palmprint images by David Zhang’s method to segment only the region of interests. Next, we extracted palmprint features based on the hybrid HOG-SGF feature extraction method. Then, an optimized auto-encoder (AE) was utilized to reduce the dimensionality of the extracted features. Finally, a fast and robust regularized extreme learning machine (RELM) was applied for the classification task. In the evaluation phase of the proposed approach, a number of experiments were conducted on three publicly available palmprint databases, namely MS-PolyU of multispectral palmprint images and CASIA and Tongji of contactless palmprint images. Experimentally, the results reveal that the proposed approach outperforms the existing state-of-the-art approaches even when a small number of training samples are used.

Feature extraction in palmprint recognition using spiral of moment skewness and kurtosis algorithm

Because of their high recognition rates, coding-based approaches that use multispectral palmprint images have become one of the most popular palmprint recognition methods. This paper describes a new multispectral palmprint recognition method that aims to further improve the performance of coding-based approaches by focusing on the local binary pattern (LBP) filters and spiral moments features. The final feature map is derived through a staged process of creating a composite of spiral and LBP features by fusing them together and passing the features through the minimum redundancy maximum relevance transformers. Using Hamming distances, the inter- and intra-similarities of the palmprint feature maps are determined. The experimental technique was evaluated using the available data on the IITD, MSPolyU and PolyU PPDB databases. The results indicate that the method achieved high levels of accuracy in the identification and verification modes. Furthermore, this method outperforms the existing advanced techniques.

TECHNIQUE FOR PERSON IDENTIFICATION BY USING PALM PRINT METHOD

Palm print is a special and solid biometric trademark with high convenience. With the expanding interest of highly exact and strong palm print confirmation framework, multispectral imaging has been utilized to acquire more discriminative data and increment the anti-spoofing ability of palm print. The proposed work has utilize PolyU palm print Database. Palm print images were gathered from 250 volunteers, including 195 males and 55 females. Work extracting palm print includes by utilizing stockwel technique.

Fast and efficient minutia‐based palmprint matching

Using the palmprint in recognition systems has received a lot of interest during the last two decades. Some of these systems are based on first-level features, such as the existing lines and creases in palmprint images, and others use second-level features, such as minutiae, which are more reliable in comparison with the first group. Owing to a large number of minutiae in a palmprint, ∼1000 minutiae, the matching process is time consuming. In this study, a new minutia-based matching strategy is proposed to make the matching process faster and more efficient. First, an orientation field estimation algorithm based on region-growing is proposed, which emphasises selecting seed points with higher quality. Second, the estimated orientation field is used to align palmprint images to the same coordinate system, resulting in fewer computations during minutia matching. Finally, a new minutia descriptor based on the orientation field is designed to distinguish minutiae with different local orientation structures. This descriptor helps to find two mated minutiae much faster, speeding up the matching process. The proposed palmprint matching algorithm has been evaluated on the THUPALMLAB database, and the results show the superiority of the proposed algorithm over most of the state-of-the-art algorithms.

Palmprint gender classification by convolutional neural network

Palmprint gender classification can revolutionise the performance of authentication systems, reduce searching space and speed up matching rate. However, to the best of their knowledge, there is no literature addressing this issue. The authors design a new convolutional neural network (CNN) structure, fine‐tuning Visual Geometry Group Network, up to 19 layers to achieve a 20‐layer network, for palmprint gender classification. Experimental results show that the proposed structure could achieve good performance for gender classification. They also investigate palmprint images with 15 different kinds of spectra. They empirically find that a palmprint image acquired by the Blue spectrum could achieve 89.2% correct classification and could be considered as a suitable spectrum for gender classification. The neural network is able to classify a 224 × 224 × 3‐pixel palmprint image in <23 ms, verifying that the proposed CNN is an effective real‐time solution.