Please use this identifier to cite or link to this item:
https://www.um.edu.mt/library/oar/handle/123456789/92409
Full metadata record
DC Field | Value | Language |
---|---|---|
dc.date.accessioned | 2022-03-25T10:30:38Z | - |
dc.date.available | 2022-03-25T10:30:38Z | - |
dc.date.issued | 2012 | - |
dc.identifier.citation | Callus, T. (2012). Child abuse image database (Bachelor's dissertation). | en_GB |
dc.identifier.uri | https://www.um.edu.mt/library/oar/handle/123456789/92409 | - |
dc.description | B.SC.ICT(HONS)ARTIFICIAL INTELLIGENCE | en_GB |
dc.description.abstract | Child abuse is a huge problem that effects children all around the world. With the aid of computer science and the advancements in computer vision we can help in tackling this worldwide problem. The main principle is to build a reliable tool that is made up of various features which address this ever growing problem. This study focuses on detecting abuse by detecting child nudity in images and by performing image similarity. Child nudity is considered as a form of abuse and the possession of such images is considered against the law. Image similarity will say how similarity two images are. A high percentage of similarity in images often means that the images where taken in the same place. The child nudity algorithm uses an age classifier to determine whether the image contains a child/adolescent or not. The age classifier takes as input a face that was detected in the image. If nudity detection algorithm detects a high level of nudity and the age classifier detects children the image is considered to contain child nudity. The image similarity uses the SURF algorithm to extract and match interest points from images. The more points matched the more similar are the images and the more probability that the images were taken in the same place. The proposed system will be evaluated using two methods. The first method is performing some manual tests and analyzing the results. The second method is performing automated tests that returns percentages of true and false positives and compare them to other results from similar published work. | en_GB |
dc.language.iso | en | en_GB |
dc.rights | info:eu-repo/semantics/restrictedAccess | en_GB |
dc.subject | Child abuse | en_GB |
dc.subject | Image processing | en_GB |
dc.subject | Algorithms | en_GB |
dc.title | Child abuse image database | en_GB |
dc.type | bachelorThesis | en_GB |
dc.rights.holder | The copyright of this work belongs to the author(s)/publisher. The rights of this work are as defined by the appropriate Copyright Legislation or as modified by any successive legislation. Users may access this work and can make use of the information contained in accordance with the Copyright Legislation provided that the author must be properly acknowledged. Further distribution or reproduction in any format is prohibited without the prior permission of the copyright holder. | en_GB |
dc.publisher.institution | University of Malta | en_GB |
dc.publisher.department | Faculty of Information and Communication Technology. Department of Artificial Intelligence | en_GB |
dc.description.reviewed | N/A | en_GB |
dc.contributor.creator | Callus, Thorsten (2012) | - |
Appears in Collections: | Dissertations - FacICT - 2012 Dissertations - FacICTAI - 2002-2014 |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
BSC(HONS)ICT_Callus Thorsten_2012.PDF Restricted Access | 10.61 MB | Adobe PDF | View/Open Request a copy |
Items in OAR@UM are protected by copyright, with all rights reserved, unless otherwise indicated.