Thumbnail Image

Thesis

Browse

2014 - 20173
Reset filters
Subject: Computer Science and Engineering × Start date: 2010 × Type: Thesis × Thesis Type: M.Tech ×
Reset filters

Search Results

Now showing 1 - 3 of 3
  • Thumbnail Image
    ThesisItemRestricted
    Development of software to determine the maturity level of plum using image processing techniques
    (Punjab Agricultural University, Ludhiana, 2017) Harpuneet Kaur; Sawhney, B. K.
    Maturity is the key factor which determines the storage life and ripening quality of fruits. In order to provide marketing flexibility and to ensure the acceptable eating quality to the buyer it is very crucial to determine the right maturity stage. Maturity indices are also important for trade regulation, marketing strategy and for efficient use of labor and resources. The proposed system is based on the implementation of image processing techniques using digital images of different maturity stages of the plum. The external quality features like color, texture and size have been analyzed with MATLAB Image Processing Toolbox. Color feature has been extracted by using mean RGB values. Entropy, Local Binary Pattern and Discrete Cosine transformation have been used for extracting textural features. Correlation coefficients between images of various categories were recorded to determine the most dominant factor for classification. Multi Attribute Decision Making theory has been used for taking final decision. The developed system accurately determines the maturity level. Results indicated that color is the most dominant factor for classifying the plums according to maturity level. When the length and width computed from application were compared with the manual readings, the error percentage was less than 2.4%.
  • Thumbnail Image
    ThesisItemRestricted
    Development of software to estimate woody volume of a live tree
    (Punjab Agricultural University, Ludhiana, 2017) Singla, Nikita; Derminder Singh
    Tree volume is one of the oldest areas of interest and is a crucial task in tree management system. Estimating the woody volume of a live tree is important for economic, scientific purposes and provides a tool to researcher/grower. It provides the useful information about the commercial value of wood to the potential buyer/seller. Manual methods are being used largely to calculate woody volume of a tree. These methods are based on different log rules, cumbersome and laborious. The present work proposed a digital image processing technique to estimate the woody volume of a live tree. The developed program successfully determines the woody volume of standing tree trunk with MATLAB image processing techniques. In this method three parameters an individual tree were extracted from digital images of the tree. Calibration factor was also calculated to make the method independent of camera distance from the tree. The method was tested on several samples of trees and compared to experimental results. This software generates information about height, diameter and volume of the tree. The percentage error of height, diameter and volume of standing tree by proposed method was found to be less than 6.65%.
  • Thumbnail Image
    ThesisItemOpen Access
    Implementation and Performance Analysis of ClustalW Algorithm using Parallel and Distributed Computing
    (Punjab Agricultural University, Ludhiana, 2014) Jasrotia, Swati; Din, Salam
    Parallel version of bioinformatics applications can speed up the analysis of large scale sequence data, specially sequence alignment. sharing of distributed computing resources and data with the use of parallel version softwares on high performance computing is an emerging step. Clusters provide an excellent platform for solving a range of parallel and distributed applications in both scientific and commercial areas. Personal computer clusters are replacing the mainframe systems / super computers because of its cost effectiveness. In this paper, the performance result of parallel version ClustalW software is recorded on personal computer cluster with the help of an open source rocks toolkit. An easy to deploy, contract/expand, manage and scalable distributed environment is proposed and built for bioinformatics applications. The experimental results show that multiple sequence alignment of protein sequences with ClustalW software on rocks distributed environment gives high efficiency and speed up. It also shows cluster platforms are excellent alternative to access to super computing due to its price to performance ratio