Soumitra Kar

A Multi-Algorithmic Face Recognition System

The importance of utilising biometrics to establish personal authenticity and to detect impostors is growing in the present scenario of global security concern. Development of a biometric system for personal identification, which fulfills the requirements for access control of secured areas and other applications like identity validation for social welfare, crime detection, ATM access, computer security, etc. is felt to be the need of the day. Face recognition has been evolving as a convenient biometric mode for human authentication for more than last two decades. Several vendors around the world claim the successful working of their face recognition systems. However, the Face Recognition Vendor Test (FRVT) conducted by the National Institute of Standards and Technology (NISI), USA, indicates that the commercial face recognition systems do not perform up to the mark under the variations ubiquitously present in a real-life situation. Availability of a largely accepted robust face recognition system has proved elusive so far. Keeping in view the importance of indigenous development of biometric systems to cater to the requirements at BARC and elsewhere in the country, the work was started on the development of a face-based biometric authentication system. In this paper, we discuss our efforts in developing a face recognition system that functions successfully under a reasonably constrained set-up for facial image acquisition. The prototype system built in our lab finds facial match by utilizing multi-algorithmic multi-biometric technique, combining gray level statistical correlation method with Principal Component Analysis (PCA) or Discrete Cosine Transform (DCT) techniques in order to boost our system performance. After automatic detection of the face in the image and its gross scale correction, its PCA and DCT signatures are extracted. Based on a comparison of the extracted signature with the set of references, the set of top five hits are selected. Exact scale of the face is ascertained w.r.t. each of these hits by first locating the eyes employing template matching technique and then finding the inter-ocular distance. After interpolating the face to the exact scale, matching scores are computed based on gray level correlation of a number of features on the face. Final identification decision is taken amongst this set of five faces, depending on the highest score. We have tested the technique on a set of 109 images belonging to 43 subjects, both male and female. The result on this image-set indicates 89% success rate of our technique.

Potential of carbon nanotubes in water purification: an approach towards the development of an integrated membrane system

The problems of water shortages and lack of access to safe drinking water have been and will continue to grow as major global problems. To alleviate these problems, water purification technologies are being updated. Recent years have witnessed impressive breakthroughs towards practical application of nanostructured materials such as Carbon Nanotubes (CNTs) in the field of water purification owing to their unique thermal, electrical and mechanical properties. These nanoscale structures need to be arranged into well-defined configurations in order to build integrated systems with high efficiency (the nanotubes being reusable, whereas the traditional membranes foul easily and require frequent replacements), high flux (owing to the hydrophobic super smooth inner surface of nanotubes), and with improvements in chemical selectivity (through suitable chemical functionalisation of the CNTs), so that the idea of using CNTs in separation technology can be made realistic and the potential benefits of practical application of these unique materials can be exploited. This paper assesses the CNTs as an emerging technology in water purification system, particularly with respect to its potential for the removal of arsenic, fluoride, heavy metals and toxic organic components.

An Automated Vehicle License Plate Recognition System

License plate recognition is a computer vision method that identifies vehicles from their license plates. The most crucial step of such a system is accurate localization of the plate. The authors propose a system for automatic recognition that has three phases: image capture, plate localization, and license plate number recognition. They tested their methodology on 40 different car models with different types of license plates.

Preparation, characterisation and performance evaluation of anti-biofouling property of carbon nanotube-polysulfone nanocomposite membranes

Membrane based process assumes a significant importance among all the separation practices being followed in the present time. Among the limitations faced by the membrane process, the most critical one is the bio-fouling, which results in lesser throughput as well as deterioration in membrane life. To enhance the anti-biofouling behaviour of the membrane surface, nanostructured materials offer a tremendous opportunity. One potential candidate among them is the carbon nanotubes (CNTs). With the objective of having improvement over the existing anti-biofouling tendency of polysulfone membranes, nanocomposites were developed with impregnation of both single walled as well as multi walled CNTs. The performance of the membranes was evaluated in terms of pure water permeability and solute rejection studies. The membranes were characterised using scanning electron microscopy, atomic force microscopy and contact angle studies. The anti-biofouling performance of the membrane surfaces was examined using E. Coli culture and a comparison of antibiofouling tendency obtained with the nanocomposites over the pure polysulfone membranes has been made. It was confirmed that membranes with impregnation of single walled CNTs possess better antibiofouling behaviour as compared to pure polysulfone as well as polysulfone membrane embedded with multi walled CNTs.

Reinforcement of nanostructured reduced graphene oxide: a facile approach to develop high-performance nanocomposite ultrafiltration membranes minimizing the trade-off between flux and selectivity

The salient features of a nanostructured carbonaceous material like graphene or graphene oxide have provided innovative alternatives for the development of nanocomposite membranes with better selectivity without having a compromise in throughput, which as a result have a promising role to play in desalination and water purification. Here, nanostructured reduced graphene oxide (nRGO) is synthesized from graphite powder and characterized. Using non-solvent induced phase inversion technique, a series of nanocomposite ultrafiltration (UF) membranes are developed by in situ impregnation of the as synthesized nRGO in polysulfone (Ps) polymer matrix with variation of nRGO from 1 to 8 w/w%. The physicochemical features and transport properties offered by the membranes are evaluated. Structural characterization of the Ps–nRGO composite UF membranes is done by X-ray photoelectron spectroscopy and Fourier transform infrared spectroscopy. The variation in porous morphology of the membranes upon impregnation of nRGO is evaluated by scanning electron microscopy. Variation in skin surface topography is analyzed by atomic force microscopy. The change in surface hydrophilicity is evaluated by contact angle studies. The thermal and mechanical properties of the membranes are assessed by thermogravimetric analysis and tensile strength measurements, respectively. The studies reveal that an optimum loading of nRGO (2 w/w%) in the Ps matrix resulted in membranes with elimination of the trade-off between the flux and selectivity that exists with the conventional UF membranes. In addition, the optimum loading of nRGO resulted in membranes with improved thermal and mechanical stability. Thus, nRGO as an emerging potential nanofiller can lead to the development of an ideal membrane with desirable attributes.

Using Radon Transform to Recognize Skewed Images of Vehicular License Plates

An algorithm that projects image intensity along a radial line oriented at a specific angle enables the recognition of vehicular license plates at odd angles. The algorithm determines orientation angle, rotates the image to a horizontal perspective, and removes image noise from rotation to achieve recognition accuracy as high as 98 percent.

Ring artifact correction in gamma-ray process tomography imaging

Ring artifacts have been studied for X-ray based Computed Tomography (CT) systems but not on γ-ray based in-situ applications. This paper discusses application of recently proposed automatic ring artifact reduction method (Yoon et al., 2016) on previously obtained experimental projection data from a γ-ray based Industrial Process Tomography (IPT) system for a prototype catalytic column. Studies include qualitative and quantitative evaluation of the method. It is observed that ring artifacts are suppressed without loss of significant information in γ-ray PT images.

A review of thermochemical cycles for hydrogen production: analysis of potential of membrane technology in I-S, UT-3 and CuCl cycles

Global warming and climate change necessitate a serious move away from fossil-based systems toward hydrogen economy. A study has been carried out on different thermochemical cycles for hydrogen production to bring about their important aspects. The study involves process description of different routes followed by thermochemical cycles. These include metal oxide processes, sulphur family processes like iodine sulphur cycle and Westinghouse cycle, halide family processes like UT-3 and Ispra Mark cycle, copper chlorine cycle and others. This paper gives an insight into the advantages and disadvantages associated with the processes. The review is done keeping in view the relevant and useful aspects in research to present information in terms of species, operating parameters, reactors, costs involved, safety, etc. This will provide a platform for further research to save effort by referring the basic information on thermochemical cycles, so that an appropriate and satisfactory cycle may be chosen.

Mixed-matrix membranes with enhanced antifouling activity: probing the surface-tailoring potential of Tiron and chromotropic acid for nano-TiO 2

Mixed-matrix membranes (MMMs) were developed by impregnating organofunctionalized nanoadditives within fouling-susceptible polysulfone matrix following the non-solvent induced phase separation (NIPS) method. The facile functionalization of nanoparticles of anatase TiO2 (nano-TiO2) by using two different organoligands, viz. Tiron and chromotropic acid, was carried out to obtain organofunctionalized nanoadditives, FT-nano-TiO2 and FC-nano-TiO2, respectively. The structural features of nanoadditives were evaluated by X-ray diffraction, X-ray photoelectron spectroscopy, Raman and Fourier transform infrared spectroscopy, which established that Tiron leads to the blending of chelating and bridging bidentate geometries for FT-nano-TiO2, whereas chromotropic acid produces bridging bidentate as well as monodentate geometries for FC-nano-TiO2. The surface chemistry of the studied membranes, polysulfone (Psf): FT-nano-TiO2 UF and Psf: FC-nano-TiO2 UF, was profoundly influenced by the benign distributions of the nanoadditives enriched with distinctly charged sites (−SO3−H+), as evidenced by superior morphology, improved topography, enhanced surface hydrophilicity and altered electrokinetic features. The membranes exhibited enhanced solvent throughputs, viz. 3500–4000 and 3400–4300 LMD at 1 bar of transmembrane pressure, without significant compromise in their rejection attributes. The flux recovery ratios and fouling resistive behaviours of MMMs towards bovine serum albumin indicated that the nanoadditives could impart stable and appreciable antifouling activity, potentially aiding in a sustainable ultrafiltration performance.

Vehicular License Plate Localization Using Principal Component Analysis

ABSTRACT The need of vehicular license plate recognition system (VLPR) has arisen based on the need to implement traffic control on transportation systems, since early 1970s. Since then, researchers are continuously proposing various approaches and solutions. One of the significant and challenging tasks is to localize the license plate of the moving car. Since the license plate standards are not strictly practiced in world, a large amount of variations are obtained like, size, location, type of font used, background and foreground colour, and so on. The principal component analysis (PCA) is one of the widely used and most successful techniques that have been used in image recognition and compression. In this paper, we propose a novel approach to identify the license plate using PCA.