Research Interests:
Visually impaired people face problems in navigation in an unknown environment. There are many portable devices and technologies to enhance the blind persons movement. Technology has already replaced the traditional methods like simple... more
Visually impaired people face problems in navigation in an unknown environment. There are many portable devices and technologies to enhance the blind persons movement. Technology has already replaced the traditional methods like simple guiding canes with GPS based voice help, intelligent canes with inbuilt detectors and QR codes to name a few. This advancement in technology helps visually impaired people to achieve independence. This research developed an android based application for assisting blind people in navigation in any indoor or outdoor environment using a pre-generated map, GPS for self localization and vision based algorithms. The work makes use of Google voice recognition to acquire user voice input for the navigator system. Visual data acquired using the mobile camera is processed and analyzed for destination environment survey. To implement this functionality, application creates a SVM (Support Vector Machine) and train data set onto four categories like “Close Door”, ...
Research Interests:
Hands play an important part in expressing one’s actions and ideas thus Hand Gesture Recognition (HGR) is very significant in computer vision based gesture recognition for Human Computer Interaction (HCI). In our work, the dataset has... more
Hands play an important part in expressing one’s actions and ideas thus Hand Gesture Recognition (HGR) is very significant in computer vision based gesture recognition for Human Computer Interaction (HCI). In our work, the dataset has been generated for five hand gestures (Close Hand, Open Hand, Victory Hand, Thumb Down and Thumb Up), by making videos of 10 different users doing the gestures with all possible variations resulting in total 16,240 entries. Firstly we have used image processing algorithms like Bilateral Filter, Median Blur and Gaussian Threshold for smoothing the images and then compared the performance of different Support Vector Machine (SVM) kernels i.e. rbfdot, vanilladot, polydot, tanhdot, laplacedot and besseldot, for HGR. The accuracy achieved with different SVM kernels varied from 24.17% to 85.07% with training-testing ratio of 70–30% for 16,240 entries in the dataset. The 10-fold cross validation is performed to prove the robustness of the kernel with SVM.
Research Interests:
Security in banking transactions is major concern. In this paper we propose a method to use face as a biometric password along with PIN number in ATM machines. For face recognition in controlled environment Wavelet, LBP and PCA are used.... more
Security in banking transactions is major concern. In this paper we propose a method to use face as a biometric password along with PIN number in ATM machines. For face recognition in controlled environment Wavelet, LBP and PCA are used. This can be used to perform more secure ATM transactions. Even after authorizing the access to a user the algorithm continuously monitors the user. If a user chooses to leave the ATM machine without completing his/her ongoing transaction or moves his/her head away from the camera for 10 s then his/her session will be logged out automatically and he/she will have to restart a new transaction. Our technique can easily be combined with regular ATM machines and people do not even require any further knowledge to use ATM machines. This technique can also be combined with banking websites to provide more safe and secure online transactions for online users.
Research Interests:
Research Interests:
Research Interests:
Coins have been the integral part of our day to day life since the ancient civilizations. In comparison to paper currency coin are easy to carry, requires less maintenance and can be used for a longer period of time. Moreover,... more
Coins have been the integral part of our day to day life since the ancient civilizations. In comparison to paper currency coin are easy to carry, requires less maintenance and can be used for a longer period of time. Moreover, commemorative coins issued by the governments were a common practice to mark important events and personalities. The coins are used frequently in places like grocery stores, banks, trains, buses, etc. With increase in coins usage and introduction of new automated machines that can use coin as a token to provide services, the need for coins to be recognized, counted, sorted automatically arises. In this thesis A Robust Rotation Invariant Coin Recognition System is proposed that takes into consideration various features of the coins (radius, color, rotation and texture) for the recognition. The system can recognize Indian coins of denominations of Rs.1, 2, 5 and 10. The system takes as input a RGB coin image of single side i.e. tail. The image is then pre-proces...
Research Interests:
Research Interests:
Research Interests:
To tackle the problem of range anxiety of a driver of an electric vehicle (EV), it is necessary to accurately estimate the power/energy consumption of EVs in real time, so that drivers can get real-time information about the vehicle’s... more
To tackle the problem of range anxiety of a driver of an electric vehicle (EV), it is necessary to accurately estimate the power/energy consumption of EVs in real time, so that drivers can get real-time information about the vehicle’s remaining range. In addition, it can be used for energy-aware routing, i.e., the driver can be provided with information that on which route less energy consumption will take place. In this paper, an integrated system has been proposed which can provide reliable and real-time estimate of the energy consumption for an EV. The approach uses Deep Auto-Encoders (DAE), cross-connected using latent space mapping, which consider historical traffic speed to predict the traffic speed at multiple time steps in future. The predicted traffic speed is used to calculate the future vehicle speed. The vehicle speed, acceleration along with wind speed, road elevation, temperature, battery’s SOC, and auxiliary loads are used as input to a multi-channel Convolutional Neu...
Research Interests:
Research Interests:
Research Interests:
In our proposed model a hybrid method of tone mapping is presented. In this technique, an image of High Dynamic Range (HDR) is employed as input and split up into two layers i.e. base and detail layer. The details of HDR image will be... more
In our proposed model a hybrid method of tone mapping is presented. In this technique, an image of High Dynamic Range (HDR) is employed as input and split up into two layers i.e. base and detail layer. The details of HDR image will be preserved because only the base layer's contrast is brought down. A bilateral filter is used to obtain the base layer. Along the base layer the logarithmic compression is used and a bias power function is employed so that the logarithmic base can be varied adaptively and the contrast remains preserved. This tone mapping technique produces a high quality tone mapped image as specified in given experimental results.
Research Interests:
Coins are frequently used in everyday life at various places like in banks, grocery stores, supermarkets, automated weighing machines, vending machines etc. So, there is a basic need to automate the counting and sorting of coins. For this... more
Coins are frequently used in everyday life at various places like in banks, grocery stores, supermarkets, automated weighing machines, vending machines etc. So, there is a basic need to automate the counting and sorting of coins. For this machines need to recognize the coins very fast and accurately, as further transaction processing depends on this recognition. Three types of systems are available in the market: Mechanical method based systems, Electromagnetic method based systems and Image processing based systems. This paper presents an overview of available systems and techniques based on image processing to recognize ancient and modern coins.
Research Interests:
Coins are integral part of our day to day life. We use coins everywhere like grocery store, banks, buses, trains etc. So it becomes a basic need that coins can be sorted and counted automatically. For this it is necessary that coins can... more
Coins are integral part of our day to day life. We use coins everywhere like grocery store, banks, buses, trains etc. So it becomes a basic need that coins can be sorted and counted automatically. For this it is necessary that coins can be recognized automatically. In this paper we have developed an ANN (Artificial Neural Network) based Automated Coin Recognition System for the recognition of Indian Coins of denomination Rs. 1, 2, 5 and 10 with rotation invariance. We have taken images from both sides of coin. So this system is capable of recognizing coins from both sides. Features are extracted from images using techniques of Hough Transformation, Pattern Averaging etc. Then, the extracted features are passed as input to a trained Neural Network. 97.74% recognition rate has been achieved during the experiments i.e. only 2.26% miss recognition, which is quite encouraging.