Please use this identifier to cite or link to this item:
http://hdl.handle.net/10603/468622
Title: | An efficient content based satellite image retrieval system using optimized feature selection technique |
Researcher: | Sunitha, T |
Guide(s): | sivarani, T S |
Keywords: | Engineering and Technology Computer Science Imaging Science and Photographic Technology Image retrieval Feature selection Feature extraction |
University: | Anna University |
Completed Date: | 2021 |
Abstract: | Remote sensing satellite images are used for rural development newlineplanning, land usage calculation, climate change observations, agriculture, newlinedisaster management, and many more. in the case of disaster management, the newlineimages help to analyze natural hazards like earthquakes, cyclones, agricultural newlinedroughts, landslides, floods, forest fires, etc. an increase in forest fire incidents newlineacross the globe poses threat to the entire ecosystem of the world. so, newlinemonitoring the forest regularly using satellite images is essential to secure the newlinelife of the animals, living beings, and natural resources in the forest. newlinecontent based satellite image retrieval (cbsir) system is meant to newlineretrieve forest fire images from the remote sensing satellite image datasets. the newlineprecision of the retrieved images in existing cbsir is impacted by the lowlevel newlinefeatures such as shape, color, and texture referred to during the feature newlineextraction stage. also, the existing system has a semantic gap problem where newlineirrelevant images are retrieved for the input query image. most importantly, the newlineimage retrieval time is high and accuracy is low in the existing system due to newlinethe increased number of features used for similarity matching. these problems newlineprevent the existing cbsir system from timely detection of forest fires newlineaccurately.this research focused on the updated framework for the cbsir system newlineto retrieve the highly relevant fire images from the huge volume of fire images. newlineadjusted intensity based variant of adaptive histogram equalization (aiva) newlinealgorithm is proposed in the pre-processing stage to improve the contrast of the newlineimage, eliminate the unwanted noise, and avoid the over enhancement of the newlineimages. a hybrid feature extraction technique is proposed in the feature newlineextraction stage to extract the desirable features of satellite images with lower newlineretrieval time and high accuracy. newline newline |
Pagination: | xix, 121p. |
URI: | http://hdl.handle.net/10603/468622 |
Appears in Departments: | Faculty of Information and Communication Engineering |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
01_title.pdf | Attached File | 23.7 kB | Adobe PDF | View/Open |
02_prelim pages.pdf | 1.13 MB | Adobe PDF | View/Open | |
03_content.pdf | 22.94 kB | Adobe PDF | View/Open | |
04_abstract.pdf | 9.21 kB | Adobe PDF | View/Open | |
05_chapter 1.pdf | 1.28 MB | Adobe PDF | View/Open | |
06_chapter 2.pdf | 924.52 kB | Adobe PDF | View/Open | |
07_chapter 3.pdf | 1.99 MB | Adobe PDF | View/Open | |
08_chapter 4.pdf | 1.48 MB | Adobe PDF | View/Open | |
09_chapter 5.pdf | 645.02 kB | Adobe PDF | View/Open | |
10_chapter 6.pdf | 1.64 MB | Adobe PDF | View/Open | |
11_annexures.pdf | 126.39 kB | Adobe PDF | View/Open | |
80_recommendation.pdf | 70.98 kB | Adobe PDF | View/Open |
Items in Shodhganga are licensed under Creative Commons Licence Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0).
Altmetric Badge: