Vijayakumar Bhagavatula

Illumination Tolerant Face Recognition Using a Novel Face From Sketch Synthesis Approach and Advanced Correlation Filters

Current state-of-the-art approach for performing face sketch recognition transforms all the test face images into sketches, and then performs recognition on sketch domain using the sketch composite. In our approach we propose the opposite; which has advantages in a real-time system; we propose to generate a realistic face image from the composite sketch using a hybrid subspace method and then build an illumination tolerant correlation filter which can recognize the person under different illumination variations from a surveillance video footage. We show how effective proposed algorithm works on the CMU PIE (pose illumination and expression) database

Investigating Correlation-Based Fingerprint Authentication Schemes for Mobile Devices Using the J2ME technology

This paper investigates correlation-based fingerprint authentication schemes that can be used for mobile devices. The investigated algorithms were implemented with a J2ME environment on the application layer. In order to reduce the resources demanded for the mobile device environment, we also propose a new hierarchical correlation-based scheme based on the idea that the overall authentication can be decomposed into partial autocorrelations. The algorithms have been tested on a J2ME CDC 1.0 emulator of a smart mobile phone.

A Method for Simultaneous Position and Timing Error Detection for Bit-Patterned Media

This paper presents a method for simultaneously detecting the position and timing error in bit patterned media (BPM). Conventionally the position error is determined only in the servo area while the timing error is determined in the data area. However, the regular spatial arrangement of bits in BPM allows position information and timing information to be extracted from the data. The new method exploits interference between adjacent tracks by employing a read head wider than the track pitch. The new method also provides the capability of reading data from two tracks at the same time.

Automatic segmentation of cardiosynchronous waveforms using cepstral analysis and continuous wavelet transforms

The cardiosynchronous signal obtained through Radio Frequency Impedance Interrogation (RFII) is a non-invasive method for monitoring hemodynamics with potential applications in combat triage and biometric identification. The RFII signal is periodic in nature dominated by the heart beat cycle. The first step in both of these applications is to segment the signal by identifying a fiducial point in each heart beat cycle. A continuous wavelet transform was utilized to locate the fiducial points with high temporal resolution. Cepstral Analysis was used to estimate the average heart rate to focus on the appropriate portion of the time-frequency spectrum. Robust heartbeats from RFII signals collected from four subjects were segmented using this method.

Simultaneous PES Generation, Timing Recovery, and Multi-Track Read on Patterned Media: Concept and Performance

This paper presents new work on the performance of the method for simultaneously detecting the position and timing error in bit patterned media (BPM). The regular spatial arrangement of bits in BPM allows position information and timing information to be extracted from the data. Our method exploits interference between adjacent tracks by employing a read head wider than the track pitch. The method also provides the capability of reading data from two tracks at the same time. Here we employ a more realistic read head field and more realistic shapes of the bit islands in simulations than in prior work. We determine the sensitivity of the position detection, timing recovery, and bit error rate of the method to manufacturing variations in the locations and sizes of bits.

Low density parity check (LDPC) codes for optical data storage

Low-density parity check (LDPC codes, first introduced by Gallager (1962), are error-correcting codes based on very sparse parity check matrices. Recently, in the wake of excellent performance of turbo codes, LDPC codes were rediscovered as another category of random codes approaching the Shannon capacity limit with practical decoding complexity. Extensive investigation of the applications of turbo codes and LDPC codes for data storage channels has been reported. While LDPC codes can be iteratively decoded using the sum-product algorithm with moderate complexity, the memory required to specify the nonzero elements in the random parity check matrix can be a major challenge for hardware implementation. Construction of good LDPC codes with structure is particularly desired in data storage channels to facilitate low complexity implementation. We address this issue by introducing a class of low complexity structured LDPC codes which are useful for optical data storage.

Second-degree correlation surface features from Optimal Trade-off Synthetic Discriminant Function filters for subject identification using radio frequency cardiosynchronous waveforms

Radio Frequency Impedance Interrogation (RFII) measures hemodynamic function via resonance frequency coupling to a hydrophilic protein molecule. The RFII device generates a cardiosynchronous waveform from the identification of blood movement in the time, frequency, and voltage domains. This paper examines RFII signals with the end goal of allowing confirmation of the identity of a subject in an operational setting. An Optimal Trade-off Synthetic Discriminant Function (OT-SDF) was applied to filter the data stream for subject identification. Preliminary results using the OT-SDF Filters demonstrate 63.3% successful single-heartbeat subject identification. However, each individuals correlation surfaces appear to have a unique waveform morphology that is visually distinct from the other individuals in the data set. Improved identification was seen with second-degree correlation suggesting that a second-degree correlation may hold great potential as a biometric feature extraction identifier. We show that using correlation plane outputs as features actually provide a robust biometric identifier and significant higher identification accuracy.