Professor Vinod Chandran

Biography

Vinod Chandran is currently a Professor at Queensland University of Technology, Brisbane, Australia with a research and teaching continuing appointment. He has been educated in India (IIT Madras - Bachelors degree in Electrical Engineering ), USA (Texas Tech University and Washington State University - Masters degrees in Electrical Engineering and Computer Science and PhD degree in Electrical and Computer Engineering) and Australia (Queensland University of Technology - Graduate Certificate in Higher Education). He has teaching experience also in these three countries at the university level. He is a senior member of the IEEE. He is a citizen of Australia.

Broad area of research: Professor Vinod Chandran conducts research in Signal Processing, Image Processing and Pattern Recognition for Biometrics and Biomedical applications. In particular, he has focused on invariance properties, nonlinear transformations and higher order statistics (HOS)  for advancement of theoretical knowledge. Theoretical contributions include the derivation of principal domains of computation for HOS of arbitrary order, derivation of statistics of the bispectrum in the presence of leakage and estimation of the statistics of bicoherence and tricoherence. A new method of extracting similarity transformation invariant features using the bispectrum was discovered and developed by Professor Chandran in 1993.  HOS techniques have been applied by Professor Chandran and his team to Speech, Ground Penetrating Radar, EEG, and Heart rate variability.   He has supervised 13 PhD students to completion as the principal supervisor, 13 PhD students to completion as an associate supervisor, 3 Masters by research students as the principal supervisor and 3 as an associate supervisor. He is currently the principal supervisor of 8 PhD students and an associate supervisor for 6 PhD students. His students have investigated 2D and 3D face recognition, image coding and interpretability, fractal techniques applied to face recognition, super-resolution applied to face recognition from surveillance images, higher order spectral techniques applied to ground penetrating radar for finding coal rock interfaces, higher order spectral techniques applied to electron microscope images for identification of viruses, multi-classifier techniques in text dependent speaker recognition, higher order spectral techniques applied to image hashing and methods for iris recognition robust to light and drug induced pupil size changes. He has co-authored over 160 papers in international peer-reviewed journals and conferences. He has an h-index of 26 (Google Scholar).

Higher order spectral analysis: Professor Chandran has made seminal contributions to the theory and practice of higher order spectral analysis stretching back to 1988. He has published original work on 2D bispectral analysis, statistics of the bispectrum in the presence of leakage, a general procedure to derive principal domains of higher order spectra and statistics of the bicoherence and tricoherence, a new similarity transformation invariant feature extraction technique using bispectral integrals and an extension of the feature extraction technique to 2D images, all in Transactions of the IEEE. Professor Chandran and his research team at QUT have developed and applied higher order spectral techniques in many areas such as speaker verification, radar signal processing, sonar signal processing, document image processing, biomedical signal and image processing and biometric based cryptographic key generation. Professor Chandran has authored software for time-varying HOS analysis of EEG signals and has a few publications in this area.  

Image processing Professor Chandran and his team have investigated image processing techniques to enhance facial images from surveillance video using super-resolution methods, extracting features from facial images for automated face recognition, fractal iterated function systems, Gabor filter-based multi-resolution decompositions, part-face feature modeling, eye tracking to detect regions of interest (ROI) and ROI coding in JPEG2000, and higher order spectral invariant feature extraction from Radon transform projections. They have applied new image processing techniques to face recognition, clinically significant maculopathy detection in fundus images, logo recognition in document images, circuit diagram interpretation in document images and virus classification from electron microscope images. A variant of the method that extracts bispectral invariant features from histograms of regions has been applied to classification of cells (Indirect Immuno-Fluoroscopy cell classification).

Pattern recognition Professor Chandran and his research team have developed a number of innovative techniques in extracting robust features from higher order spectral line integrals and entropy parameters, and their use with Gaussian mixture models and Support Vector Machines. They have applied the new methodologies to biometrics (speaker verification, face recognition), biomedical signal and image processing (onset of epileptic seizure detection from EEG, heart condition monitoring using HRV, maculopathy detection in fundus images), radar signal processing (detection of coal rock interface using Ground Penetrating Radar) and sonar image processing (detection of sea mines). A decision fusion architecture was proposed by Professor Chandran in 2004 which combines multiple samples and multiple instances in a sequential decision system that allows multiple attempts. Closed form expressions were derived for detection errors for the multi-classifier framework. This architecture was applied and further developed by PhD students, Vishnu Nallagatla (for text-dependent speaker verification) and Inma Tomeo-Reyes (for iris recognition from parts and samples).

Biometrics Applications PhD students, Jamie Cook and Chris McCool, investigated methods of using 3D faces for recognition and combining them with 2D faces for improving performance. PhD student, Frank Lin, investigated the use of super-resolution technique to enhance faces in surveillance video frames such that they can be used for recognition. Post-doctoral fellow, Daryl Ning compared the performance of higher order spectral (HOS) invariant features with the more traditional Mel-frequency Cepstral Coefficient (MFCC) features in conjunction with GMM classifiers. The HOS features retain some Fourier phase information and are more robust to noise. PhD student, Vishnu Nallagatla investigated a system for text-dependent speaker verification in a multi-classifier framework using multiple utterances and multiple attempts. PhD student, Inma Tomeo Reyes, investigated robust techniques for iris recognition using a parts and samples approach that improves performance, especially when subject to pupil size variations (that may be induced as a sample presentation attack to increase false rejects). Professor Chandran developed a technique using time-varying higher order spectral features and Gaussian Mixture Models for online signature verification. He has also authored HOS and Classifier libraries in C. .     

Biomedical applications:  Professor Chandran’s feature extraction technique has been applied by Chua Kuang Chua and project students under his supervision in the development of a system for detection of clinically significant diabetic maculopathy in fundus images. Professor Chandran and Chua Kuang Chua have applied higher order spectral and nonlinear techniques to analyze and classify biomedical signals such as heart rate variability (HRV) and Electroencephalograph (EEG). They have successfully classified HRV signals into several cardiac condition classes. They have also classified EEG segments into normal, pre-ictal and epileptic seizure classes in a system to detect the onset of an epileptic seizure.

Research Grants and Contracts   Summary

• 2001-2004, 2004-2007, 2009-14: Chief investigator on 3 ARC Discovery Grants
• 1997-1999,1999-2001: Chief investigator on international research contracts from the USA
• 2007-2010: Chief Investigator on an ARC Linkage
• 1993-Present: A number of smaller external and internal grants

Details of some grants  

2009-2014: ARC Discovery Grant - Omniscient Face Recognition from Un-Cooperative Subjects, Role: Chief Investigator  Chief Investigators: Sridharan, Chandran, Fookes, Lucey and Tistarelli  Participating Agencies: ARC Amount: $269,110

2004-2007:  ARC Discovery Grant - Techniques to use stereo vision for improving person identification systems based on face recognition  Role: First Chief Investigator Chief Investigators: Chandran, Sridharan Participating agencies: ARC Amount: $150,000

2001-2004:  ARC Large Grant -Speech recognition using source dependent information from phases of higher-order spectra of quasi-periodic Role: First Chief Investigator Chief Investigators: Chandran, Sridharan, Roberts Participating agencies: ARC, DSTO Amount: $116,000

2000: ARC Small Grant -Speech recognition using source dependent information from phases of higher-order spectra of quasi-periodic speech components  Role: First Chief Investigator Chief Investigators: Chandran, Sridharan Participating agencies: ARC Amount: $13,000

2007-2010: ARC Linkage Grant -Progressive Transmission of Street Directory Assistance and Business Pages over 3G and 4G mobile channels Role: Chief Investigator  Chief Investigators: Senadji, Chandran, Mason, Wright and Chang  Participating Agencies: ARC, Telstra  Amount: $77,658

2007: National Security Science and Technology - Biometrics Vulnerability Assessment (Co-investigator, 4 of 6), with Moeller/Poulton/Dunstone/Foley/ Lyman, Participating Agencies: Biometrics Institute, Amount: $16,400 (of $59,650)

2000: ARC Small Grant -Multimodal speech processing and its application to speaker recognition Role: Chief Investigator Chief Investigators: Sridharan, Chandran Participating agencies: ARC Amount: $11,000

2003-2004: Office of Naval Research - 3D Face Identification and Verification Role: Investigator Chief Investigator: Chandran, Sridharan Participating agencies: ONR Amount: $100,000

1997-2001: Office of Naval Research (subcontracts from WSU and WHO) - Detection, Classification and Identification of Objects in Cluttered Acoustic Images Role: Chief Investigator at QUT on the Research Contract, Chief Investigators: Elgar (WHOI, USA), Chandran Participating agencies: Washington State University, Woods Hole Oceanographic Institution, ONR Amount: $257,581

2002-2003: Defense Science and Technology Organization - Speech and Image Recognition Technologies Role: Co Investigator Chief Investigators: Sridharan, Chandran, Boles Participating agencies: DSTO Amount: $100,000

2001-2002: Defense Science and Technology Organization - Image Recognition Technologies  Role: Co Investigator, Chief Investigators: Sridharan, Chandran, Boles Participating agencies: DSTO Amount: $30,000

2001-2002: Defense Science and Technology Organization -Importance Prioritized JPEG2000 Image Coder Role: Co Investigator Chief Investigator: Sridharan, Chandran Participating agencies: DSTO Amount: $25,000

2001: Australian Customs - 2D Face Recognition Trials
Role: Co Investigator Chief Investigator: Sridharan, Chandran, Boles Participating agencies: Australian Customs Amount: $10,000

1997: Australia Post - Voice and Vision, Role: Co-investigator (2 of 2), w Prof. Sridharan, Participating Agencies: Australia Post, Amount: $10000

 

 

Positions held
2009-Present: Professor, QUT, Australia
2004 - 2009: Associate Professor, QUT, Australia
997 - 2004: Senior Lecturer, QUT, Australia
1993 - 1997: Lecturer, QUT, Australia
1990 - 1993: Post doctoral Teaching Associate, Washington State University, USA
1986 - 1990: Teaching / Research Assistant,  Washington State University, USA
1984 - 1985: Teaching / Research Assistant, Texas Tech University, USA 1
1984: (3 months) Asst. Executive Engineer, Oil and Natural Gas Commission, India
1982 - 1984: Ad hoc Teaching Associate, Maulana Azad College of Technology, India