Thumbnail Image

Theses

Browse

2016 - 201820
Reset filters
Subject: Information Technology × Start date: 2015 × Thesis Type: M.Tech ×
Reset filters

Search Results

Now showing 1 - 9 of 20
  • Thumbnail Image
    ThesisItemOpen Access
    Performance evaluation off content based image retrieval using dwt, modified k means and neural network classification
    (G.B. Pant University of Agriculture and Technology, Pantnagar - 263145 (Uttarakhand), 2018-10) Aeri, Manisha; Ashok Kumar
    The increasing need for searching relevant images in large scale databases is an active area of interest now days. The current method of text based image retrieval has many challenges as context sensitive images cannot be retrieved, the effort required to annotate each and every image manually as well as the difference in human perception while describing an image gives inaccurate and inefficient results during the retrieval process. So, Content Based Image retrieval came into picture in which the retrieval of images can be done by using the contents of an image such as color, shape or texture. In this research work we have proposed a novel approach for content based image retrieval by applying an efficient clustering algorithm i.e. Modified k means for image segmentation and we have used the concept of discrete wavelet transformation, color moments and HSV histogram for extracting image features. Artificial neural network is trained about the extracted features from the database images. The testing phase involves the querying and retrieval task in which the query images features are compared with the trained features of database images and the best matched images are retrieved from the database similar to the query image. This technique is tested by conducting experiments on WANG image dataset containing 1000 general purpose colored images in terms of precision and recall in which the proposed technique gave better results as compared to the existing techniques.
  • Thumbnail Image
    ThesisItemOpen Access
    A novel approach recognizing objects from images by using SIFT and HMM
    (G.B. Pant University of Agriculture and Technology, Pantnagar - 263145 (Uttarakhand), 2018-08) Verma, Aanchal; Srivastava, Ratnesh Prasad
    Recognizing Objects in images and finding a particular image from the set of images have always been a challenging task. There has been significant progress in this area as discussed in literature survey part of the thesis. This section proposed a novel approach for recognizing such objects by the use of algorithms SIFT and HMM which are used for the purpose of feature extraction and classification. The survey papers of last few years clearly shows that there has been a missing in the use of SIFT and HMM for recognizing the images. Therefore, The proposed framework will perform the recognition of an object as human or something by using database images and store the derived features or Keypoints from image sequence for measurement in recognition stage. An automated extraction of relevant feature point from given image is provided to automate the recognition procedure using SIFT. This process improves the recognition accuracy. The human facial features extracted through SIFT are utilized for the recognition of human. The SIFT feature will be created for each given images and the key points are computed, and then HMM is applied for recognition. HMM uses SIFT feature to perform recognition process on MATLAB. The recognition outcomes demonstrate that our proposed framework gives more accurate performance when comparing to tradition procedure.
  • Thumbnail Image
    ThesisItemOpen Access
    An additive noise suppression in noisy signal using spectral subtraction
    (G.B. Pant University of Agriculture and Technology, Pantnagar - 263145 (Uttarakhand), 2017-07) Malik, Medha; Negi, Chetan Singh
  • Thumbnail Image
    ThesisItemOpen Access
    Fetching of texture from image using enhanced morphological component analysis
    (G.B. Pant University of Agriculture and Technology, Pantnagar - 263145 (Uttarakhand), 2017-08) Giri, Kapil; Pandey, Binay Kumar
    The Morphological Component Analysis (MCA) is a method which allows us to separate features contained in an image when these features present different morphological aspects. MCA can be used for image inpainting and image separation task. MCA can be very useful for decomposing images into texture and piecewise smooth (cartoon) parts. MCA followed by TV regularization scheme described by J-L Starck, M.Elad, D. L. Donoho is a very efficient method for separating image into its piecewise smooth content and its texture. Due to the use of curvelet dictionary in MCA piecewise smooth content part suffers from ringing artifact . TV regularization scheme was used to remove ringing artifact from the piecewise smooth part. In this work we aim at improving the MCA algorithm so that it can separate the piece wise smooth part and texture part more efficiently. The TV regularization scheme is approximated with Daubechies wavelet. Daubechies wavelet transform is applied over the cartoon part of the image then soft thresholding of the coefficient is done and then again the image is reconstructed by taking inverse Daubechies wavelet transform. Simulation result has shown that proposed method is giving better performance than the previous method in terms of various qualities metric such as Peak Signal to Noise Ratio (PSNR), Structural Similarity (SSIM).
  • Thumbnail Image
    ThesisItemOpen Access
    Analysis of fraudulent in graph database for identification and prevention
    (G.B. Pant University of Agriculture and Technology, Pantnagar - 263145 (Uttarakhand), 2017-08) Rawat, Deepak Singh; Singh, Rajesh Shyam
    Because of the amazing addition of extortion which results in incredible loss of billions of cash far and wide consistently, a couple of current systems in perceiving distortion are reliably made and associated with various business fields. Extortion revelation incorporates watching the direction of customers with a particular ultimate objective to assess, distinguish, or counteract undesirable lead. Undesirable direct is a wide term including wrongdoing, deception, intrusion, and record defaulting. This examination displays a graph-based technique executed with graph database itself used protection from fraudulent. The purpose of this investigation is to perceive and maintain a strategic distance from deception if there ought to be an event of on the web and disconnected keeping money from the net transaction with a record using graph database. Meanwhile, we have endeavored to ensure that real exchanges are not dismissed by adjusting them to past example of fraudulent. These case studies are also providing information by matching pattern of fraudulent case with database entry. Banks are looking to minimize big data through misrepresentation identification and prevention frameworks. A wide range of cutting edge fraud innovations are being connected to fraudulent Internet banking transactions recognition and protection. In any case, they have no successful identification instrument to distinguish honest to goodness clients and follow their unlawful exercises. We consider a prototype to vanquish each one of these difficulties using the graph database.
  • Thumbnail Image
    ThesisItemOpen Access
    Analysis of scheduling algorithms in cloud computing
    (G.B. Pant University of Agriculture and Technology, Pantnagar - 263145 (Uttarakhand), 2017-08) Yadav, Ashwani Kumar; Mandoria, H.L.
    Cloud computing is a rapidly growing paradigm, which has gained tremendous interest in researchers at the industrial level and at the academic level. As a result, many researchers have been enthusiastically involved in research projects related to cloud computing. Researcher's major challenge in cloud computing is to provide a solution to the scheduling problem. We have analyzed various scheduling algorithms for user’s performance criteria. We have proposed an enhanced conductance based scheduling algorithm, which has been implemented to bind the cloudlets on cloud resources minimizing the overall execution time and resource utilization and to increase system’s QoS. The purpose of our proposed research is to analyze the scheduling algorithm in cloud computing, keeping in view VM’s MIPS and length of tasks. Using the CloudSim simulation toolkit, we have evaluated the performance of existing algorithms like FCFS. We have analyzed scheduling algorithms for the time shared environment in cloudlet provisioning policies. The results of our simulation proved the efficiency of the new proposed approach in reducing the overall execution time and in the makespan. With our extensive study of cloudsim library we have analyzed these algorithms by extending the datacenter broker class of CloudSim.
  • Thumbnail Image
    ThesisItemOpen Access
    Fast retrieval of images in secured content based image retrieval system
    (G.B. Pant University of Agriculture and Technology, Pantnagar - 263145 (Uttarakhand), 2017-07) Pareek, Shreela; Mandoria, H.L.
    Content Based Image Retrieval is referred to a technique for retrieving images from database similar to query image. Retrieval is based on automated matching of features of query image with that of image database through some similarity evaluation. Due to the increment in advancement of technologies data has been increased to huge amount, hence securing these data is an important concern and to retrieve data from these large dataset in less amount of time is one of the challenging task for Content Based Image Retrieval System. Here we have implemented a method of image retrieval to overcome these challenges. Thus this research work focuses on analyzing the performance of CBIR based on GPU and using Daubechies Wavelet Transform and Arnold’s Cat Map. Analysis is performed in four phases, first phase compares different image feature descriptors. Second phase compares some members of Wavelet Family, then in third phase comparison of CBIR process using GPU CUDA and without using GPU CUDA is performed, final phase compares and analyze the performance of CBIR after applying GPU CUDA, Daubechies Wavelet-db35 for feature extraction and securing images using Arnold’s Cat Map algorithm. Results are evaluated based on the performance matrices i.e precision and recall and time consumed to retrieve images, encryption algorithm is tested on the basis on PSNR and MSE. This research work has been simulated in MATLAB.
  • Thumbnail Image
    ThesisItemOpen Access
    Simulation of color and textural based method for image encryption in image processing
    (G.B. Pant University of Agriculture and Technology, Pantnagar - 263145 (Uttarakhand), 2017-07) Singh, Samridhi; Mandoria, H.L.
    These days secured data storage and transmission have become an important issue in this digital world due to the increased use of Internet for communication purpose. Information security is becoming more importance as the amount of sensitive data being exchanged on the Internet increases. The service like confidentiality and data integrity are required to protect data against unauthorized usage and modification. In these years several image encryption methods are introduced by various researchers to secure multimedia information while transmit via public networks. A novel Image encryption method based on color and textural feature is suggested in this research work. In color feature image is divided into RGB component and each component is encrypted by shuffling pixels and this shuffling is decided by three different keys to encrypt each component. To transmit data whole image is divided into blocks then confusion and diffusion means shuffling of pixels according to the pattern and pattern is decided on the basis of key which is generated after extracting feature of the image through GLCM and feature is selected through PCA in texture feature. The results confirm that the proposed method resists the statistical analysis. Also attains acceptable entropy value and has a robust performance against attacks. The simulation of the above algorithm is done on Matlab.
  • Thumbnail Image
    ThesisItemOpen Access
    Study and analyze detection and isolation of Sybil attack in vehicular Ad-Hoc network
    (G.B. Pant University of Agriculture and Technology, Pantnagar - 263145 (Uttarakhand), 2017-07) Chauhan, Prachi; Mandoria, H.L.
    Vehicular ad-hoc network (VANET) have been impartially a hot research area over the recent years. Due to increase in vehicle crashes and fatalities, Vanet fascinate so much attention of both academia and industry. Vehicular ad-hoc networks (VANETs) are the dynamic subcategory of mobile ad-hoc network (MANET) in which mobile node presents set of smart vehicles on road. The vehicular ad-hoc network is the self-configuration and decentralized type of network with no central controller. In the recent times, various techniques have been examined for the detection of malicious nodes in the network. Path establishment from source to destination is the major issue in the network due to the high mobility of vehicle nodes and many security prevailing in Vanet for both safety and navigational purposes of vehicles on the road. The proposed technique is signal strength and monitor mode based technique. The simulation has been performed on Network Simulator version2 (NS2). The comparison between existing and proposed technique have been done with the help of several parameters such as throughput, end to end delay, packet loss and routing overhead. The results display that the preferred technique demonstrated expressive results as compare to existing technique.