Open AccessPixort: A Novel Approach for Effective Photo Album Clustering
Author(s) -
Cheryl L Mathias,
Crystal F D’Souza,
Job Alexander,
Mariah S Hudson,
Renuka Tantry
Publication year - 2021
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/1099/1/012015
Subject(s) - computer science , cluster analysis , upload , image retrieval , artificial intelligence , feature extraction , content based image retrieval , the internet , feature (linguistics) , pattern recognition (psychology) , image (mathematics) , information retrieval , world wide web , linguistics , philosophy
Retrieving digital records on the internet and capturing, uploading images on social media or other platforms has become such a notable part of one’s daily existence. Digital content and videos constitute majority of this ever-growing deluge of information. Major problems are faced by people on a daily basis when they have to manually scan large databases to search for particular images thereby exhausting valuable time and resources. To overcome this, Content Based Image Retrieval in addition to Emotion Detection must be performed to efficiently store and retrieve very specific images. An effective image clustering application should be capable of accurately extracting images based on the given query image. It should also provide functionalities like Facial Detection and Recognition for customized usage and retrieval. In this study, texture-based feature extraction followed by the usage of a feed forward backpropagation neural network has been proposed for CBIR. For emotion detection, correlation has been proposed in addition to the Viola Jones algorithm which has been used for facial detection.