Garima Joshi

Effect of Temperature Variation on Gate Tunneling Currents in Nanoscale MOSFETs

In this paper, an analytical model for gate tunneling current has been deployed by solving Schrodinger equation using Wentzel-Kramer-Brillouin (WKB) approximation method for trapezoidal potential barrier. The gate tunneling current has been computed for direct tunneling from channel to gate as well as for tunneling from source drain extension (SDE) region to gate. Effect of temperature variation on gate tunneling current with SiO2 thickness of 4 nm to 1 nm range has been studied at various gate voltages. To study the effect of temperature on gate tunneling current, the related parameters have been modeled based on physics. Effect of variation of substrate doping concentration (Na) on gate tunneling current in n-MOSFET has also been studied. These studies have been used to bring out the design margins available in equivalent oxide thickness (EOT) and Na.

Static hand gestures recognition system using shape based features

In this paper, an easy and a simple approach using shape based features to recognize alphabets of Indian Sign Language (ISL) has been proposed. Geometrical features and Zernike Moments (ZM) of hand shapes are extracted. Covariance and minimum Euclidean Distance are used for recognition of sign. For 19 ISL alphabets, high recognition rate is achieved.

Analysis of shape recognition capability of Krawtchouk moments

In this paper, Krawtchouk invariant moments are used as features for object recognition. For hand images, the performance of Krawtchouk moments in terms of recognition accuracy, rotational invariance, scale invariance, computational time and feature vector size, has been analysed. A user independent dataset for 21 subjects under varying illumination conditions is created. A comparative analysis with Zernike moments is also done. Results confirm that Krawtchouk moments have a better shape recognition capability even for a very small feature set.

Histograms of orientation gradient investigation for static hand gestures

In this paper, Histograms of Orientation Gradient (HOG) algorithm is used to identify the static hand gestures. Experimental results show that HOG descriptor is a better shape descriptor than existing feature sets for gesture recognition. The overall algorithm has only three main steps; pre-processing, feature extraction and classification. It completely omits the segmentation phase. SVM is used for recognition of gestures. High recognition accuracy is achieved for 11 hand gestures.

Indian sign language recognition using Krawtchouk moment-based local features

ABSTRACT In this paper, Krawtchouk moment-based shape features at lower orders are proposed for Indian sign language (ISL) recognition system which gives local information about the shape from a specific region of interest. The shape recognition capability of Krawtchouk moment-based local features is verified on two databases: the standard Jochen Triesch’s database and 26 ISL alphabets which are collected from 72 different subjects, with variations in position, scale and rotation. Feature selection is performed to minimise redundancy. The effect of order and feature dimensionality for different classifiers is studied. Results show that Krawtchouk moment-based local features are found to exhibit user, scale, rotation and translation invariance. Moreover, they have shape identification capability.

CFS- InfoGain based Combined Shape-based Feature Vector for Signer Independent ISL Database.

In Sign language Recognition (SLR) system, signs are identified on the basis of hand shapes. Zernike Moments (ZM) are used as an effective shape descriptor in the field of Pattern Recognition. These are derived from orthogonal Zernike polynomial. The Zernike polynomial characteristics change as order and iteration parameter are varied. Observing their behaviour gives an insight into the selection of a particular value of ZM as a part of an optimal feature vector. The performance of ZMs can be improved by combining it with other features, therefore, ZMs are combined with Hu Moments (HM) and Geometric features (GF). An optimal feature vector of size 56 is proposed for ISL dataset. The importance of the internal edge details to address issue of hand-over-hand occlusion is also highlighted in the paper. The proposed feature set gives high accuracy for Support Vector Machine (SVM), Logistic Model Tree (LMT) and Multilayer Perceptron (MLP). However, the accuracy of Bayes Net (BN), Nave Bayes (NB), J48 and kNearest Neighbour (k-NN) improves significantly for Info Gain based normalized feature set.

Identification of ISL Alphabets Using Discrete Orthogonal Moments

In this paper, discrete orthogonal moment-based shape features up to 5th order are proposed for Indian sign language (ISL) recognition system. The shape recognition capability of discrete orthogonal moment-based local features is verified on two databases. These include the standard Jochen-Triesch’s database and 26 ISL alphabets. The ISL alphabets are collected on both uniform and complex backgrounds, with variations in position, scale and rotation. The feature-set is increased for 26 ISL alphabets by varying Region of Interest (ROI) and extracting features from each ROI. A minimum possible feature-set with least redundancy is selected that gives the best recognition accuracy. The effect of order and feature dimensionality for different classifiers is studied. Results show that both Dual-Hahn and Krawtchouk moments are found to exhibit user, scale, rotation and translation invariance. Moreover, they have shape identification capability, thus achieving good recognition accuracy.

Analysis of ternary multiplier using booth encoding technique

This paper introduces a new approach to multiplication of ternary numbers. The whole multiplication is based on the efficient Booth Encoding technique that multiplies both positive as well as negative ternary numbers. Verilog HDL has been used to implement the ternary multipliers of 3bit, 8bit and 12bit. The HDL design is based on the Finite State Machine (FSM) and multiplexing techniques. The design is simulated using ModelSim SE 6.5 and synthesized using Xilinx ISE Design Suite 14.1. The results obtained from the proposed design in terms of delay, power and area have been compared with the conventional multiplier design.

Analysis of shape and orientation recognition capability of Complex Zernike Moments for signed gestures

In this paper, exact behavior of Complex Zernike Moments is analyzed. Zernike Moments contain two parameters: magnitude and orientation. In literature, mostly magnitude is considered to recognize the shape, because magnitude is orientation invariant. On the other hand, orientation of an image has its own significance in case of sign language. This work is dedicated to the study of the capability of Zernike Moments to recognize shape and orientation of Indian Sign Language gestures. Database of total 720 images of five signs (C, I, L, T, V) is used here. Test sets are designed such that they are shape specific, orientation variant and orientation invariant. Experiments are performed on these test sets for magnitude, phase and their combination. High accuracy is achieved even at lower order of Zernike Moments when both magnitude and phase are used as a feature set.

Comparative Analysis of Movement and Tracking Techniques for Indian Sign Language Recognition

Sign Language is considered as a way of communication for hearing handicapped persons. We can make the communication of deaf people easier by building a translation system of this language. To realize these systems, the identification of words and gestures in sign language is very important. Indian Sign Language (ISL) is used in major parts of India that includes gestures. Most of the gestures include movements of a part of body. Here, in this paper, the focus is to track the movement of hand, identifying its shape and direction of motion. The tracking techniques are compared on some factors and analysis is done. Preprocessing for extracting the region of interest (a hand) is done on image sequences. Tracking is done through Mean-shift and Kalman filter. The performance of the above mentioned algorithms are compared on the basis of precision, tracking time, affect of velocity change and recognition. Different shape based features are extracted based on different region based shape models. The preprocessing and feature extraction is done in MATLAB. After extracting these features are applied as input to a classifier. Classification is done in WEKA. Performance of the system is analyzed by identification of hand shape with direction.