Department of Electrical and Electronics Engineering
Permanent URI for this collectionhttp://localhost:4000/handle/123456789/1925
Browse
5 results
Search Results
Item Robust Multi-Spectral Palm-Print Recognition(Springer, 2023-07) Ajmera, Pawan K.In recent years, biometric has emerged as the most trustworthy and reliable technique for verifying or identifying humans. The distinctive and consistent qualities of palm prints have led to a rise in their use as a biometric attribute for user access and authentication. A palm-print template is proposed that stores the relative geometric information of the minutiae points. It is impossible to determine the orientation of the palm-print from the template since it lacks information on the orientation of the minutiae and their position. For template matching, an internal angle based on Delaunay triangulation is used, which then produces matching scores for various spectral bands. The multispectral data is finally integrated using a score-level fusion approach that minimizes the overlapping effects. The technique was evaluated on the widely used PolyU multispectral palm-print database, which provides more accurate results than a mono-spectral band. The total Correct Recognition Rate (CRR) is 96.28%, while the Equal Error Rate (EER) is 0.16%.Item Multi-view Feature Learning Based on Texture Description for Palm-Print Recognition(Springer, 2023-09) Ajmera, Pawan K.Biometric is the science of validating an individual’s identity while using behavioral and physiological characteristics. In unconstrained scenario, contactless palm-print recognition leads to better recognition accuracy of individuals. Most of the existing texture descriptors are fail to learn stable and discriminative features from palm-print images. The paper presents a multi-view feature learning method based on texture description for palm-print recognition. The multi-view features are simultaneously extracted by two complementary operators. We also learn how to use feature mapping to convert multi-view data into hash codes. Experiments are carried out on palm-print databases captured using a variety of devices and acquisition methods. We demonstrate that the proposed method has superior performance compared to the current methods.Item Palm-print recognition based on scale invariant features(IEEE, 2019) Ajmera, Pawan K.Over the past few years, palm-print recognition has proved to be one of the extensively used technology for human identification/verification in many aspects. This paper presents the implementation of five feature extraction algorithms such as Mean, AAD (Average Absolute Deviation), GMF (Gaussian Membership Function) along with SIFT (Scale Invariant Feature Transform) and SURF (Speeded Up Robust Feature) for effective recognition. SVM (Support Vector Machine) and KNN (K-Nearest Neighbor) are the machine learning algorithms used for the classification of data. Experimentations are carried out on the PolyU and IIT-Delhi palm-print databases. The scale invariant features of SURF provides the best performance with Correct Recognition Rate (CRR) of 99.56% and 97.95% for IIT-Delhi and PolyU palm-print database respectively.Item Palm-print identification based on deep residual networks(IEEE, 2021) Ajmera, Pawan K.Biometric recognition has been an inseparable part of security and authorization. In the last decade, palm-print has been widely used in security access and person authentication. However, for efficient identity management and access regulation neural network based classification algorithms are required as they provide an efficient means of adaptive feature extraction using back-propagation, leading to better classification results. This paper presents the implementation of various neural networks for an efficient palm-print classification. The model is trained using the ResNet-18, ResNet-50 and ResNet-101 architectures using the PolyU and IIT-Delhi palm-print databases. The evaluation of the performance parameters indicate that the ResNet with SURF features provides the best results in lesser number of epochs. The results obtained are significantly better than the traditional methods.Item Upgrading security and protection in ear biometrics(IET, 2019-02) Ajmera, Pawan K.Biometrics is being widely accepted for user authentication across the globe. Integration of biometrics in the daily life provokes the need to design secure authentication systems. This study proposes the use of outer ear images as a biometric modality. The comparable complexity between the human outer ear and face in terms of its uniqueness and permanence has increased interest in the use of ear as a biometric. However, similar to face recognition, it poses challenges of variation in illumination, contrast, rotation, scale and pose. Owing to the extensive work in the field of computer vision using convolutional neural networks (CNNs), its feasibility in the field of ear biometrics has been presented in this work. The proposed technique uses a CNN as a feature extractor and a support vector machine (SVM) for the classification task. The joint CNN-SVM framework is used for mapping ear images to random base-n codes. The codes are further hashed using the secure hash algorithm SHA-3 to generate secure ear templates. The feasibility of the proposed technique has been evaluated on annotated web ears dataset. This work demonstrates 12.52% average equal error rate without any image pre-processing, which shows that the proposed approach is promising in the field of secure ear biometrics.