Online product search using gray level co-occurrence matrix, color moments, and histogram of oriented gradients for content based image retrieval. Issue 1 (July 2020)

• Record Type: Journal Article
• Authors: Halim, Henry; Hendryli, J; Herwindiati, D E
• Is Part Of: IOP conference series. Volume 852:Issue 1(2020)
• Publication Date: 2020-07
• Subjects: Materials science -- Periodicals; 620.1105
• DOI: 10.1088/1757-899X/852/1/012140
• Languages: English
• ISSNs: 1757-8981

Abstract

Abstract: Usually an information retrieval system uses text as its query for retrieving the results. However, the system can also use images, as opposed to text, for the search query. This technique commonly is called content based image retrieval. In this paper, we explore the retrieval of fashion products such as hats, sneakers, t-shirts, flat shoes, and dresses from images using color, texture, and shape features to represent the characteristics of the query images. The color, texture, and shape features are color moments, gray level co-occurrence matrix, and histogram of oriented gradients, respectively. The feature extraction process produces vectors with statistical feature value which can then be clustered to retrieve images with high degree of similarity. Comparing those feature representations for the retrieval system, it is shown that shape feature is the best representation.