Please use this identifier to cite or link to this item:
http://hdl.handle.net/10603/334538
Title: | Morphological filter and object oriented techniques for shadow detection and removal from urban high resolution images |
Researcher: | Vicky Nair |
Guide(s): | Parimala Geetha, K and Meenakshi, M |
Keywords: | Morphological filter Shadow detection Colour histogram |
University: | Anna University |
Completed Date: | 2019 |
Abstract: | newline This work presents a method for shadow detection and removal from high resolution single image. The study is organized in three parts namely. Segmentation, Shadow detection and shadow elimination. An algorithm is proposed for each of the three parts and compared with the standard existing algorithms. The shadow is formed because of blocking of the light source by some object. Self-shadow and cast shadow are the two outstanding forms of shadows. Self-shadows occur on the sides not facing the source of illumination. When the shadow of one object falls on another object, cast shadow is created. Cast shadows incorporate Umbra and Penumbra. The entire blocking of light creates the umbra region in the shadow and partial blocking creates the penumbra. The light source is completely concealed in umbra region and partially concealed in penumbra region. The first part of the study deals with segmentation. Segmentation is important process for any image recognition system as it helps to extract features from the system. The aim of image segmentation is to find and extract regions which make up an image. The extracted features are used to perform other image processing applications. The work introduces an algorithm using morphological processing for colour segmentation of the image. The two basic morphological operations, erosion and dilation are applied on the image. The difference between erosion and dilation gives the gradient. The edges are detected using the image gradient. The colour histogram of the image is generated. Smoothing operation on the histogram is performed by generating two new histogram using the gaussian function and the difference of the two gives a smoothened histogram. newline newline |
Pagination: | xx,137p. |
URI: | http://hdl.handle.net/10603/334538 |
Appears in Departments: | Faculty of Information and Communication Engineering |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
01_title.pdf | Attached File | 104.18 kB | Adobe PDF | View/Open |
02_certificates.pdf | 71.69 kB | Adobe PDF | View/Open | |
03_vivaproceedings.pdf | 583.21 kB | Adobe PDF | View/Open | |
04_bonafidecertificate.pdf | 290.19 kB | Adobe PDF | View/Open | |
05_abstracts.pdf | 98.32 kB | Adobe PDF | View/Open | |
06_acknowledgements.pdf | 185.23 kB | Adobe PDF | View/Open | |
07_contents.pdf | 203.68 kB | Adobe PDF | View/Open | |
08_listoftables.pdf | 91.48 kB | Adobe PDF | View/Open | |
09_listoffigures.pdf | 30.17 kB | Adobe PDF | View/Open | |
10_listofabbreviations.pdf | 221.64 kB | Adobe PDF | View/Open | |
11_chapter1.pdf | 359.18 kB | Adobe PDF | View/Open | |
12_chapter2.pdf | 367.84 kB | Adobe PDF | View/Open | |
13_chapter3.pdf | 1.28 MB | Adobe PDF | View/Open | |
14_chapter4.pdf | 938.28 kB | Adobe PDF | View/Open | |
15_chapter5.pdf | 707.92 kB | Adobe PDF | View/Open | |
16_chapter6.pdf | 125.51 kB | Adobe PDF | View/Open | |
17_conclusion.pdf | 125.51 kB | Adobe PDF | View/Open | |
18_references.pdf | 430.07 kB | Adobe PDF | View/Open | |
19_listofpublications.pdf | 108.94 kB | Adobe PDF | View/Open | |
80_recommendation.pdf | 180.04 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: