The Journal of Pattern Recognition Research (JPRR) provides an international forum for the electronic publication of high-quality research and industrial experience articles in all areas of pattern recognition, machine learning, and artificial intelligence. JPRR is committed to rigorous yet rapid reviewing. Final versions are published electronically
(ISSN 1558-884X) immediately upon acceptance.
News
 
Dr. Alex Pappachen James has been appointed as the new Assistant Editor
Alex Pappachen James

Alex Pappachen James is currently a faculty in the Electrical and Electronic Engineering discipline, School of Engineering, Nazarbayev University that is a partner school of University College London. He finished his PhD from School of Engineering, Griffith University in the area of developing neuron inspired face recognition methods. After completing PhD in 2 years, he was working with Queensland Micro and Nanotechnology facility at Griffith University on Intelligent systems research using memory devices, and was also involved as a lecturer in teaching courses in the Griffith School of Engineering and Queensland Institute of Business Technology. He accepted a faculty position at the School of Computer Science, Indian Institute of Information Technology and Management (IIIT), where he led the Intelligent Machines and Neuromorphic Engineering Laboratory and founded the Center for Excellence in Applied Machine Intelligence and Pattern Analysis, where he is currently an adjunct professor.

In press: Partially Occluded Object Recognition Based on Assumption based Truth Maintenance System
JPRR Article #319
We present a new approach to recognize and locate partially occluded rigid objects from a given scene and generate a belief about the scene using assumption based truth maintenance (ATM) system. The ATM system is basically a tool for belief revision. It explores multiple potential solutions and can workout efficiently with inconsistent information. In practice, sometimes occlusion of objects in a 2D-scene may occur due to the presence of objects which are not described in our primary knowledge base and which may appear to be an object, in addition to the model objects of our primary knowledge base. Hence, after detection of such an event, question of revising belief about the scene may arise to establish a new belief. The present approach to recognize and locate an occluded scene is completely different from the existing paradigm based on the concept of hypothesis generation and verification. The new paradigm of recognition is hypothesis generation and belief revision.
In press: Kernel Non-Negative Matrix Factorization for Seismic Signature Separation
JPRR Article #463
A supervised learning algorithm for the separation of seismic sources in a single channel is presented. The proposed algorithm employs non-negative matrix factorization (NMF) technique in the feature space, called Kernel NMF (KNMF). KNMF is based on factorizing the magnitude spectrogram, a time frequency representation of an input signal, into a sum of components. Each of which has a fixed magnitude spectrum and a time-varying gain. KNMF extracts discriminative spectral features from the spectrogram of seismic data and separates signatures of a human from a horse. The main benefit of the proposed technique is its ability to decompose a complex signal automatically into objects that have a meaningful interpretation. KNMF is developed and implemented on seismic data of human and horse footsteps. The performance of this method is very promising as demonstrated by the experimental results.
Just Published: A System for Handwritten Script Identification from Indian Document
JPRR Article #485
In a country like India a number of scripts (a total of 13) are used to write official languages (a total of 23). For development of General Optical Character Recognizer (OCR) for a particular language in India, the script by which the document is written is to be identified first. The task is more challenging when it comes about handwritten documents. So identification of the script from a document may be written with any of these 13 scripts is a very challenging task. In this paper we have identified scripts written by popular six official languages of India. Here we have used very simple and efficient features at document level for the same. Using some Abstract/ Mathematical features, Structure based features and Script dependent features, series of classifiers were used. Overall accuracy of the proposed system is at present 92.8% on the test set without rejection. Read Article
Dr. Yanhui Guo has been appointed as the new Assistant Editor
Yanhui Guo

Dr. Yanhui Guo is an Assistant Professor in the School of Science, Technology and Engineering Management at Saint Thomas University in Florida. He actively researches image processing, computer vision, medical information processing, parallel algorithms, computer-aided detection/diagnosis, and neutrosophic theory. For graduate school he attended the Harbin Institute of Technology, receiving his master’s degree in computer science. He obtained his doctorate at Utah State University, conducting further research as a fellow scholar in the University of Michigan. He has also published over 40 scholarly works in academic journals and conferences. JPRR wholeheartedly welcomes Dr. Guo, an experienced and vigilant editor.

Invitation: Associate/Prospective JPRR Editors in Bioinformatics: 2013

JPRR maintains its annual editor enrollment tradition. In the term of 2013, the call for the enrollment is focused on the Bioinformatics Section. JPRR welcomes motivated talents, scholars, as well as individual contributors to voluntarily invest their expertise, wit, and passion into the Pattern Recognition community by assisting in conducting manuscript reviews, identifying trustworthy reviewers, and by learning from the JPRR magnificent Editorial constellation. New editorial candidates are welcome to provide their CVs to confirm their expertise in Bioinformatics and Pattern Recognition domains.

In order to apply, please send your cover letter and CV to apply@jprr.org.

Professor Raymond DeCarlo has accepted the title of JPRR Distinguished Editor
Raymond A. Decarlo

Dr. DeCarlo is a world-renowned authority in Control Theory. The topics of his interest are hybrid optimal control, hybrid electric vehicle modeling and control, Lyapunov stability, decentralized control, analog fault analysis and diagnosis of circuits. His practical academic work has been crystallized in a number of monographs and co-authored formidable volumes.

Vol 8, No 1 (2013)
Applications
A System for Handwritten Script Identification From Indian Document 1-12
In a country like India a number of scripts (a total of 13) are used to write different official languages (a total of 23). For development of Optical Character Recognizer (OCR) for a particular language, the script by which the document is written is to be identified first.
Vol 7, No 1 (2012)
Pattern Recognition Theory
Distance Between Distributions With Special Topologies of Cost Matrices 1-25
Comparing two distributions plays important role in many problems. The traditional minimum cost flow problem has been utilized as a distance measure between two distributions (transportation problem) such as the earth mover’ distance (EMD).
Design of a Decision Tree to Classify Similar Looking Characters Using Subimages for Kannada Script 42-55
Kannada script has large number of vowels, consonants, conjuncts and combination of these in inflectional and agglutinative manner.
Biometric Authentication and Identification Using Keystroke Dynamics: A Survey 116-139
Dependence on computers to store and process sensitive information has made it necessary to secure them from intruders. A behavioral biometric such as keystroke dynamics which makes use of the typing cadence of an individual can be used to strengthen existing security techniques effectively and cheaply.
Integrating Stereo Disparity and Optical Flow by Closely-Coupled Method 175-187
As a convergence method for stereo matching and motion estimation, this paper presents an equation, called Disparity-Optical flow Equation, that relates disparity with optical flow in rectified images.
Applications
Recognizing Thai Broken Characters Based on Set-Partitions and N-Grams Graphs 26-41
Automatic recognition of broken Thai characters represents one of the biggest challenges in some applications such as computerized restoration of Thai text documents.
Legend Extraction From E-Learning Video Streams 56-71
Teachers usually illustrate major pedagogical concepts with graphics and/or images and/or tables and, in doing so, take a considerable amount of time in explanation.
Investigation of Shoeprints Using Radon Transform With Reduced Computational Complexity 80-89
Biometric traits along with various evidences left by the offender at the place of crime found useful in crime investigation.
Image Registration Using Log Polar Transform and Phase Correlation to Recover Higher Scale 90-105
Image registration is an important and fundamental task in image processing used to match two different images. Given two or more different images to be registered, image registration estimates the parameters of the geometrical transformation model that maps the sensed images back to its reference image.
Clustering Financial Time Series via Information-Theory Analysis and Rank Statistics 106-115
A method of clustering of a time series set is described. Each cluster includes time series containing the same amounts of information about other time series.
Robust Facial Feature Extraction and Matching 140-154
Automated feature extraction and correspondence determination is an extremely important problem in the face recognition community as it often forms the foundation of the normalisation and database construction phases of many recognition and verification systems.
Pattern Recognition in Bioinformatics
Facial Feature Points Tracking Based on AAM With Optical Flow Constrained Initialization 72-79
A facial feature points tracking method is proposed by adding Lucas-Kanade optical flow constraint on the face alignment algorithm, Active Appearance Model (AAM).
Guest Papers
A Multi-Level Model for Fingerprint Image Enhancement 155-174
Fingerprint has remained a very vital index for human recognition. In the field of security, series of Automatic Fingerprint Identification System (AFIS) have been developed.