" I believe Teaching is a work of Heart "

Cocoa-Care

---

Project Guide: Harivinod N
Team Members : Pooja P, Nithesh Kumar H, Gururaj G Hegde, Ashritha B S
Period : Jan-May 2017

Abstract

Cocoa is an economically important crop that nowadays enlarges its production in southern India. To assist the farmers growing cocoa, This project involves development of android application "Cocoa-Care" for cocoa disease identification. This application automatically identifies the diseases of cocoa crops, thereby helps the farmers who have little or no information about the disease. This application is developed by applying digital image processing techniques on the diseased cocoa images. The approach replaces the manual disease inspection by the android application that identifies the cocoa disease from the captured image and suggests the possible remedies for the farmer. The texture features are used for the image representation and description. The matching is performed by nearest neighbor classifier. The results obtained are promising.

Architecture

Close

Leaf-Id

---

Project Guide: Harivinod N
Team Members : Ramya H, Roopa R N, Sandhya A, Ashwija Rai A
Period : Jan-May 2016

Abstract

Regardless of its vast usefulness, many important species of plants are facing extinction. Botanists need to identify near extinct plants, before climate change and development erase their living record. Reasons for extinction include global warming, introduction of exotic and hybrid species, habitat loss, disease, pollution, lack of knowledge and over exploitation. Out of these reasons, lack of knowledge contribute more, which necessitates the need for automated tools that can help botanists, researchers and students to identify plants from leaf images.

In this project, we developed an Automatic Medicinal Plant Leaf Identification System . The system is intended to identify the plant/leaf based on given leaf image and provide useful information about the recognized plant/leaf. The system will be developed by applying digital image processing techniques on the shape and texture features of the leaf. The system uses image segmentation technique to extract the leaf from the image, various feature extraction techniques to represent the leaf and matching based on nearest neighbor classifier.

Architecture

Close

Cheque-Process

---

Project Guide: Harivinod N
Team Members : Sharanya K, Shreevani K, Sumadhura Rao, Vyshali K N
Period : Jan-May 2015

Abstract

Machine simulation of human reading has become a promising area of research after the arrival of digital computers. The main reason for that is not only the challenge in simulating the human reading but also its utility in developing document processing systems capable of transferring data present on documents like bank cheques, commercial forms, government records and envelops into machine readable format. Paper cheques still play a big role in the non-cash transactions in the world even after the arrival of credit cards, debit cards and other electronic means of payment. In many developing countries, the present cheque processing procedure requires a bank employee to read and manually enter the information present on a cheque (or its image) and also verify the entries. As a large number of cheques have to be processed every day in a bank, an automatic reading system can save much of the work. Even with the success achieved in character recognition over the last few decades, the recognition of handwritten information and the verification of signatures present on bank cheques still remain a challenging problem in document image analysis. The system we are developing, accepts image of a bank cheque. The objective of the proposed system is to design and develop a computer vision based automatic bank cheque clearing system which prevents physical movement of the cheques and validates the contents electronically. It also minimizes transaction costs and provides better verification process.

Sample Cheque Image

Close

Video Synopsis

---

Project Guide: Harivinod N
Team Members : Chaithra P V, Harshitha R, Nikita P V, Prajna S K
Period : Jan-May 2016

Abstract

Video synopsis aims at providing condensed representations of video datasets that can be easily captured from digital cameras, especially for daily surveillance videos. Previous work in video synopsis usually moves active objects along the time axis, which inevitably causes collisions among the moving objects if compressed much. Here, we propose a novel approach for compact video synopsis using a unified spatiotemporal optimization. Our approach globally shifts moving objects in both spatial and temporal domains, which shifting objects temporally to reduce the length of the video and shifting colliding objects spatially to avoid visible collision artifacts. Our experimental results have shown that the compact video synopsis we produced can be browsed quickly, preserves relative spatiotemporal relationships, and avoids motion collisions.

Architecture of the Project

Close