sdfgsdfg

Image Forgery Detection

In the current era of technology and multimedia we find ourselves surrounded by Digital Images. Considering the rapid advancement in Image Processing it has been very easy to manipulate any digital image using sophisticated tools like PhotoShop, CorelDraw etc. This proposed thesis shall provide some methods and algorithms that will attempt to restore authenticity and integrity in digital images.

Advance High Level Content Based Image Retrieval system for Logo and Trademark Distinctivity

Find out similar deceptive Trademarks from a database on the basis of their colour, texture, shape, spatial features and text within them.

De-skewing, De-warping and De-noising of perspectively distorted documents: A comparision between Geometrical Methods and Machine Learning Approach.

The main research directions in camera captured document analysis is to deal with the page curl and perspective distortions. Current document analysis and optical character recognition systems do not expect these types of artifacts and show poor performance ,when applied directly to camera-captured documents. The goal of text de-skewing is to re-orient the skewed text lines and the goal of text de-warping is to flatten a camera-captured document such that it becomes readable by currently OCR systems. We will use Support Vector Machine and Neural Network for our Machine Learning Approach.

User state modeling using EEG signals

Interface is the most crucial part of any system. Traditionally the evaluation of interface is done by surveys or think-aloud method which don't give valuable insight into the user's changing experience throughout the task. The motive is to develop objective technique to measure user states such as workload, emotion and fatigue in real time using EEG signals of UI evaluation.

Digital Image Forgery Detection

Three types of image forgery detection was done:
1. Image Composting.
2. Image Enhancement.
3. Copy Move Forgery.

Lesion detection in eye due Diabetic Retinopathy.

This system, with the help of morphological image processing techniques, we are trying to detect lesions in two categories i.e. dark and bright lesions

Content based Image Retrieval using Trademark Database

Content based image retrieval (CBIR) has been an active area of research due to its wide range of applicability and the major reduction in effort that it provides. This work focuses on the use of CBIR systems in the field of intellectual property rights. Preventing trademark infringement was the main objective of this project.

Graph Based Energy Efficient Data Gathering in Wireless Sensor Networks

Data gathering plays an important role in several applications of wireless sensor network. Better protocols are always needed to improve the lifetime of the network. This thesis analyses the existing approaches and proposes a Greedy Multilevel Clustering Algorithm (GMCA) of data collection which makes a dynamic path for multiple nodes to connect to the base station (BS) in a Wireless Sensor Network (WSN) which results in increment of the lifetime of the WSN.

A new multilinear discriminant analysis method for face recognition with fewer features under illumination and expression variations

This project makes use of fewer and effective features than traditional algorithms without relegation in recognition accuracy.