Embedded Files
I did my PhD at IIT (Indian Institute of Technology) Roorke. My researches are focused on developing a Mobile Vision System for Plant Biometric System - using plant leaf images in Computer Science and Engineering (CSE). Dr. P. Sateesh Kumar, Associate Professor, CSE, IIT Roorkee and Dr. Debashis Ghosh, Professor, ECE, IIT Roorkee are my supervisors under whom I am completed my thesis work.
Volumetric Tumor Detection
Volumetric Tumor Detection
This work is on segmentation of brain pathological tissues (Tumor, Edema an Narcotic core) and visualize it in 3D for their better physiological understanding. We propose a novel approach which combines threshold and region grow algorithm for tumor detection. In this proposed system, FLAIR and T2 modalities of MRI are used due to their unique ability to detect the high and low contrast lesions with great accuracy. Here, first the tumor is segmented from an image which is a combination of FLAIR and T2 image using a threshold value, selected automatically based on the intensity variance of tumor and normal tissues in 3D MR images. Then the tumor part is extracted from the actual 3D MRI of brain by selecting the largest connected volume. To correctly detect tumor 26 connected neighbors are used.
Artistic Image-Video Style Transfer
Artistic Image-Video Style Transfer
In today's world of digitization image stylization dealing with colorization, oil painting, cartoon style generation, and so on has emerged since past decade. As a definition, style transfer is process where a source image I_s and a template image I_t are the inputs to the machine to generate a new source image I_s^t which has the content of I_s with the style of I_t. Image style transfer is information of some images can be perceptually enhanced like color transfer enhancing the detail of grayscale image.
Image Aesthetics and Interestingness
Image Aesthetics and Interestingness
Today, we human beings have traveled many years of evolution and has reached to the present modern world of digitization where we are populated with digital ubiquitous devices. Due to these low-cost and high resolution devices such as smartphones, tables or phablets make our lifestyle ubiquitous and plays a major role in daily routines. However,due to such increase of digital medias the possibility of redundancy and noisy multimedia data are increasing.
Now the major questions in image quality assessment (IQA) in front of digital camera users are:
Is it a good or bad image?
Is it blurred?
Is it a duplicate image?
Is this a near duplicate image?
Is this image important?
Is this image interesting?
Is to keep this image or delete?
Real-time computing.
The main focus is on a novel low-computing image representation for aesthetic and quality measurement, suitable for low-computing devices.
Which one is good?
Natural Image Understanding - Plant Image
Natural Image Understanding - Plant Image
Currently working with natural image understanding with complex objects. Designing a plant biometric system.
Our contribution is threefold:
Complex background segmentation.
Species identification.
Species pattern analysis.
Real-time computing.
Computer/Mobile Vision for Low-Processing Devices - A Smart Classroom
Computer/Mobile Vision for Low-Processing Devices - A Smart Classroom
We develop a method allowing to track object in a physical world with a color coded ribbon.
Presenting an automated intelligent computer vision based HCI system to control and interact without skin-color MAP algorithm to detect motion with more accurate and more natural and efficient way. The experiment involves a very simple mathematics for color tolerance and for motion detection used a trigonometric concept further for action performance based on the gesture definition to compute on low-computing machine such as mobile devices. Here, hand gesture is used to operate presentation by the presenters for easiness while presenting in front of a large gathering and practically sounds good in performance.
The method proposes the following advantages:
Real-time computation.
Transform invariant.
illumination change tolerance.
More details can be found in the below publications:
“MOBILE AUGMENTED REALITY BASED INTERACTIVE TEACHING AND LEARNING SYSTEM WITH LOW COMPUTATION APPROACH”, IN IEEE SYMPOSIUM SERIES ON COMPUTATIONAL INTELLIGENCE 2013, (IEEE SSCI 2013),
APRIL 16TH – 19TH, 2013, SINGAPORE, PP. 97-103.
“CONTROL OF COMPUTER PROCESS USING IMAGE PROCESSING AND COMPUTER VISION FOR LOW-PROCESSING DEVICES”, IN INTERNATIONAL CONFERENCE ON ADVANCE IN COMPUTING, COMMUNICATIONS AND INFORMATICS, ICACCI-2012, ACM, 3 - 5 AUGUST 2012 CHENNAI, INDIA, PP. 1169-1174. (BEST PAPER AWARD)
A Mobile Vision System for Palmistry
A Mobile Vision System for Palmistry
We develop a mobile vision system to detect and analysis hand prints and predict the personality and nature of human.
An automated intelligent mobile vision based technique for extracting principal lines from colored images of human palm to predict the personality of users. A robust and efficient algorithm is designed to work on low processing devices such as mobile phones with low resolution camera. A mobile client-server architecture is involved in this project.
Won First prize of worth 15k in National Student Project Contest (SPC) in 2nd International Conference on Intelligent Interactive Technologies and Multimedia (IITM-2013) held at Indian Institute of Information Technology, Allahabad.
Was second runner up at IBM I-CARE 2013, IBM Research Center, New Delhi.
The method proposes the following advantages:
Capture the real world on a small screen.
Real-time computation.
Mobile as Palmist.
More details can be found in the below publication:
“IMPROVED CONNECTED REGION BASED APPROACH FOR EXTRACTION OF PRINCIPAL LINES FROM PALM IMAGES”, IN INTERNATIONAL CONFERENCE ON IMAGE AND SIGNAL PROCESSING (ICISP-2013), ELSEVIER, AUGUST 9TH – 11TH, 2013, BANGALORE, PP. 104-112. http://searchdl.org/index.php/book_series/view/942, (BEST PAPER AWARD)
Handwriting based Human Personality Prediction
Handwriting based Human Personality Prediction
Handwriting analysis is a method to predict personality of author and to better understand the writer. Allograph and their combination analysis is a scientific method of writer identification and evaluating the behavior. To make this computerized we considered six main different types of features: (i) size of letters, (ii) slant of letters and words, (iii) baseline, (iv) pen pressure, (v) spacing between letters and (vi) spacing between words in a handwritten document to identify their personality. This can also be used for writer identification for further security.
Prasad, S., Singh, V. K., & Sapre, A. (2010). Handwriting Analysis based on Segmentation Method for Prediction of Human Personality using Support Vector Machine. International Journal of Computer Application
(0975-8887), (12Volume), 8.
Hand Data Glove - Virtual Reality
Hand Data Glove - Virtual Reality
HCI is moving more and more natural and intuitive way to be used. One of the important parts of our body is our hand which is most frequently used for the Interaction in Digital Environment and thus complexity and flexibility of motion of hands are the research topics. To recognize these hand gestures more accurately and successfully data glove is used.
We have used hand data glove to record the hand gestures and then identify the patterns and map to the various objects in digital world.
“HAND DATA GLOVE: A WEARABLE REAL-TIME DEVICE FOR HUMAN COMPUTER INTERACTION” ININTERNATIONAL JOURNAL OF ADVANCE SCIENCE AND TECHNOLOGY (IJAST), BY SCIENCE AND ENGINEERING RESEARCH SUPPORT SOCIETY, Vol. 43, JUNE 2012. PP. 15-26.
Google Sites
Report abuse