Anush Sankaran

On matching latent to latent fingerprints

This research presents a forensics application of matching two latent fingerprints. In crime scene settings, it is often required to match multiple latent fingerprints. Unlike matching latent with inked or live fingerprints, this research problem is very challenging and requires proper analysis and attention. The contribution of this paper is three fold: (i) a comparative analysis of existing algorithms is presented for this application, (ii) fusion and context switching frameworks are presented to improve the identification performance, and (iii) a multi-latent fingerprint database is prepared. The experiments highlight the need for improved feature extraction and processing methods and exhibit large scope of improvement in this important research problem.

Hierarchical fusion for matching simultaneous latent fingerprint

Simultaneous latent fingerprints are a cluster of latent fingerprints that are concurrently deposited by the same person. Inherent challenges of latent fingerprints such as partial and smudgy ridge flow information, presence of background noise, and availability of less number of features makes it challenging to develop an automated system for simultaneous latent fingerprint matching. This research attempts to fill this gap by developing a fusion framework. The contribution of this paper is two-fold: (i) an automated hierarchical fusion approach is proposed for fusing evidences from multiple latent impressions and (ii) IIITD simultaneous latent fingerprint database is prepared to drive further research in this area. The proposed algorithm yields promising results on the simultaneous latent fingerprint database.

Latent Fingerprint Matching: A Survey

Latent fingerprint has been used as evidence in the court of law for over 100 years. However, even today, a completely automated latent fingerprint system has not been achieved. Researchers have identified several important challenges in latent fingerprint recognition: 1) low information content; 2) presence of background noise and nonlinear ridge distortion; 3) need for an established scientific procedure for matching latent fingerprints; and 4) lack of publicly available latent fingerprint databases. The process of automatic latent fingerprint matching is divided into five definite stages, and this paper discusses the existing algorithms, limitations, and future research directions in each of the stages.

On latent fingerprint minutiae extraction using stacked denoising sparse AutoEncoders

Latent fingerprint identification is of critical importance in criminal investigation. FBIs Next Generation Identification program demands latent fingerprint identification to be performed in lights-out mode, with very little or no human intervention. However, the performance of an automated latent fingerprint identification is limited due to imprecise automated feature (minutiae) extraction, specifically due to noisy ridge pattern and presence of background noise. In this paper, we propose a novel descriptor based minutiae detection algorithm for latent fingerprints. Minutia and non-minutia descriptors are learnt from a large number of tenprint fingerprint patches using stacked denoising sparse autoencoders. Latent fingerprint minutiae extraction is then posed as a binary classification problem to classify patches as minutia or non-minutia patch. Experiments performed on the NIST SD-27 database shows promising results on latent fingerprint matching.

Multisensor Optical and Latent Fingerprint Database

Large-scale fingerprint recognition involves capturing ridge patterns at different time intervals using various methods, such as live-scan and paper-ink approaches, introducing intraclass variations in the fingerprint. The performance of existing algorithms is significantly affected when fingerprints are captured with diverse acquisition settings such as multisession, multispectral, multiresolution, with slap, and with latent fingerprints. One of the primary challenges in developing a generic and robust fingerprint matching algorithm is the limited availability of large data sets that capture such intraclass diversity. In this paper, we present the multisensor optical and latent fingerprint database of more than 19000 fingerprint images with different intraclass variations during fingerprint capture. We also showcase the baseline results of various matching experiments on this database. The database is aimed to drive research in building robust algorithms toward solving the problem of latent fingerprint matching and handling intraclass variations in fingerprint capture. Some potential applications for this database are identified and the research challenges that can be addressed using this database are also discussed.

Adaptive latent fingerprint segmentation using feature selection and random decision forest classification

Abstract Latent fingerprints are important evidences used by law enforcement agencies. However, current state-of-the-art for automatic latent fingerprint recognition is not as reliable as live-scan fingerprints and advancements are required in every step of the recognition pipeline. This research focuses on automatically segmenting latent fingerprints to distinguish between ridge and non-ridge patterns. There are three major contributions of this research: (i) a machine learning algorithm for combining five different categories of features for automatic latent fingerprint segmentation, (ii) a feature selection technique using modified RELIEF formulation for analyzing the influence of multiple category features on latent fingerprint segmentation, and (iii) a novel SIVV based metric to measure the effect of the segmentation algorithm without the requirement to perform the entire matching process. The image is tessellated into local patches and saliency based features along with image, gradient, ridge, and quality based features are extracted. Feature selection is performed to study the contribution of the various category features towards foreground ridge pattern representation. Using these selected features, a trained Random Decision Forest based algorithm classifies the local patches as background or foreground. The results on three publicly available databases demonstrate the efficacy of the proposed algorithm.

Group sparse autoencoder

Unsupervised feature extraction is gaining a lot of research attention following its success to represent any kind of noisy data. Owing to the presence of a lot of training parameters, these feature learning models are prone to overfitting. Different regularization methods have been explored in the literature to avoid overfitting in deep learning models. In this research, we consider autoencoder as the feature learning architecture and propose 2,1-norm based regularization to improve its learning capacity, called as Group Sparse AutoEncoder (GSAE). 2,1-norm is based on the postulate that the features from the same class will have a common sparsity pattern in the feature space. We present the learning algorithm for group sparse encoding using majorizationminimization approach. The performance of the proposed algorithm is also studied on three baseline image datasets: MNIST, CIFAR-10, and SVHN. Further, using GSAE, we propose a novel deep learning based image representation for minutia detection from latent fingerprints. Latent fingerprints contain only a partial finger region, very noisy ridge patterns, and depending on the surface it is deposited, contain significant background noise. We formulate the problem of minutia extraction as a two-class classification problem and learn the descriptor using the novel formulation of GSAE. Experimental results on two publicly available latent fingerprint datasets show that the proposed algorithm yields state-of-the-art results for automated minutia extraction. Group Sparse AutoEncoder (GSAE) learns better discriminative features compared to an unsupervised autoencoder.Class label based 2,1-regularization is incorporated to squared error reconstruction loss function using a majorization-minimization approach.The proposed GSAE is used to learn minutia representation from noisy latent fingerprint images.Results on standard image datasets, MNIST, CIFAR-10, and SVHN and latent fingerprint image datasets, NIST SD-27 and MOLF, show effectiveness of the proposed GSAE feature extraction approach.

Automated clarity and quality assessment for latent fingerprints

Clarity of a latent impression is defined as the discern-ability of fingerprint features while quality is defined as the amount (number) of features contributing towards matching. Automated estimation of clarity and quality at local regions in a latent fingerprint is a research challenge and has received limited attention in the literature. Local clarity and quality helps in better extraction of features and assessing the confidence of matches. The research focuses on (i) developing an automated local clarity estimation algorithm, (ii) developing an automated local quality estimation algorithm based on clarity, and (iii) understanding the correlation between clarity and quality in latent fingerprints. Local clarity assessment is performed using a 2-D linear symmetric structure tensor. The goodness of orientation field is proposed to estimate the local quality of a latent fingerprint. Experiments on the NIST SD-27 database show that incorporating local clarity information in the quality assessment improves the performance of the matching system.

Class sparsity signature based Restricted Boltzmann Machine

Abstract Restricted Boltzmann Machines (RBMs) have been extensively utilized in machine learning as core units in constructing deep learning architectures such as Deep Boltzmann Machines (DBMs) and Deep Belief Networks (DBNs). However, they are prone to overfitting and several regularization techniques have been proposed to mitigate this effect. In this paper, we propose the semi-supervised class sparsity signature based RBM formulation by combining unsupervised generative training of the RBM with a supervised sparsity regularizer. The proposed approach, termed as cssRBM, enforces sparsity at the class level to ensure that coherent and discriminative representations are learnt during training. Combining unsupervised learning with supervised learning allows the model to utilize external training data to learn better generative features while the supervised learning enables fine-tuning for discrimination using the learned features. We construct both DBMs and DBNs with cssRBM units and evaluate the performance on multiple publicly available benchmark datasets. Experiments on the MNIST and CIFAR-10 databases demonstrate that the proposed approaches are comparable with state-of-the-art deep learning architectures in the literature. We also evaluate the performance on one of the most challenging face databases, i.e., the Point and Shoot Challenge dataset. The results show that the proposed approaches improve state-of-the-art results by 15% on the PaSC database.

On smartphone camera based fingerphoto authentication

Authenticating fingerphoto images captured using a smartphone camera, provide a good alternate solution in place of traditional pin or pattern based approaches. There are multiple challenges associated with fingerphoto authentication such as background variations, environmental illumination, estimating finger position, and camera resolution. In this research, we propose a novel ScatNet feature based fingerphoto matching approach. Effective fingerphoto segmentation and enhancement are performed to aid the matching process and to attenuate the effect of capture variations. Further, we propose and create a publicly available smartphone fingerphoto database having three different subsets addressing the challenges of environmental illumination and background, along with their corresponding live scan fingerprints. Experimental results show improved performance across multiple challenges present in the database.