Open AccessA Review Paper on Visual Vocabularies for Image Flower Classification
Author(s) -
C Rachana,
Vasudev Shahapur
Publication year - 2022
Publication title -
international journal of advanced research in science, communication and technology
Language(s) - English
Resource type - Journals
ISSN - 2581-9429
DOI - 10.48175/ijarsct-2831
Subject(s) - computer science , vocabulary , histogram , artificial intelligence , classifier (uml) , bag of words model , representation (politics) , bag of words model in computer vision , variety (cybernetics) , class (philosophy) , natural language processing , pattern recognition (psychology) , visual analytics , analytics , information retrieval , visual word , visualization , image (mathematics) , image retrieval , data mining , linguistics , philosophy , politics , political science , law
Detecting the existence of objects in photos is a difficult challenge for a machine, but it has improved dramatically in recent years.In particular, a text analytics representation known as bag-of-words has gained a lot of traction and has been successfully applied to a variety of problems in the flowers domain. Local floral features are computed in the first step of this approach, then clustered into K visual words, and ultimately each flower is represented as a K-dimensional histogram over the visual words.A classifier uses this fixed-size vector representation of flowers as input. The major purpose of this project is to look into ways to improve the visual vocabulary (the collection of all visual words) in the bag-of-words method. As a result of the aggregation and fast parallel processing of tiny class-specific vocabularies, this approach allows for the construction of huge vocabularies relatively quickly.