Please use this identifier to cite or link to this item: http://hdl.handle.net/10603/596651
Title: Comprehensive Study of Applications of Supervised and Unsupervised Techniques on Spectral Data
Researcher: Modi, Anitha
Guide(s): Naik, Amisha
Keywords: Computer Science
Computer Science Software Engineering
Engineering and Technology
Precision
Thermal
University: Nirma University
Completed Date: 2024
Abstract: Recent advancements in computer vision, equipped with artificial intelligence-based models, have played a vital role in several areas, such as object detection, surveillance, autonomous vehicles, and augmented reality. These advancements have significantly improved the accuracy and efficiency of analyzing and interpreting data, expanding computer vision capabilities and applications across various sectors. Satellite image data comprises spatial and spectral information captured using different imaging techniques. Several research studies have been conducted, and challenges in the Precision Agriculture sector have been addressed using artificial intelligence (AI) based classification and segmentation approaches. Satellite data are available in multiple spectra, and they comprise information that can be extracted and can be used to address Precision Agriculture challenges. newlineThe agriculture sector faces numerous challenges, such as improper resource management, shrinking agricultural land, and diverse environmental conditions. These requirements are addressed using computer vision, AI, and integrated approaches. To address these issues, researchers have conducted numerous experiments with multiple data types comprising numeric, text, image, and video data. A significant focus was placed on obtaining advance crop yield estimates, which could be further used to monitor production and reduce agricultural losses, subsequently assisting in addressing broader challenges such as food security and sustainable agriculture. newlineIn our research, we have considered multispectral, hyperspectral, and thermal image data captured from different parts of the electromagnetic spectrum, each serving a unique purpose in observation and analysis. We have addressed challenges related to crop production, crop classification, and intrusion detection. newline There is an intrinsic requirement for extensive labeled data to train AI algorithms. The algorithm performance is substantially impacted without a significant quantity of labeled data. Therefore,
Pagination: 
URI: http://hdl.handle.net/10603/596651
Appears in Departments:Institute of Technology

Files in This Item:
File Description SizeFormat 
01_title.pdfAttached File162.58 kBAdobe PDFView/Open
02_prelim pages.pdf2.38 MBAdobe PDFView/Open
03_content.pdf834.32 kBAdobe PDFView/Open
04_abstract.pdf2.64 MBAdobe PDFView/Open
05_chapter1.pdf6.16 MBAdobe PDFView/Open
06_chapter2.pdf16.72 MBAdobe PDFView/Open
07_chapter3.pdf11.71 MBAdobe PDFView/Open
08_chapter4.pdf21.44 MBAdobe PDFView/Open
09_chapter5.pdf8.53 MBAdobe PDFView/Open
10_chapter6.pdf8.58 MBAdobe PDFView/Open
11_annexures.pdf23.68 MBAdobe PDFView/Open
80_recommendation.pdf1.83 MBAdobe PDFView/Open
Show full item record


Items in Shodhganga are licensed under Creative Commons Licence Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0).

Altmetric Badge: