Textural and color descriptor fusion for efficient content-based image retrieval algorithm
- PDF / 2,459,921 Bytes
- 15 Pages / 595.276 x 790.866 pts Page_size
- 106 Downloads / 228 Views
ORIGINAL ARTICLE
Textural and color descriptor fusion for efficient content-based image retrieval algorithm Yogita D. Mistry1 Received: 21 June 2019 / Accepted: 17 March 2020 © Springer Nature Switzerland AG 2020
Abstract In communication, medium images are used for various applications such as social websites, education, bio-medical field and industrial applications. Indexing and retrieval of such large image database pose a big problem. The content-based image retrieval (CBIR) approaches are used to select information from the input images using feature descriptors. In this article, CBIR algorithm is proposed with fusion of color and texture descriptors such as binary Gabor pattern (BGP), segmentation based fractal texture analysis (SFTA), edge histogram descriptor (EHD), Gabor wavelet texture features and fuzzy histogrambased descriptors. Laplacian score has employed to reduce feature vector dimensions. Retrieval rate is computed using different distance metrics. Experimentation was performed with ten different classes of 1000 images using Wang database. The simulation results indicate improvement in terms of precision compared with the existing techniques. Keywords Content based image retrieval (CBIR) · Binary Gabor pattern (BGP) · Segmentation based fractal texture analysis (SFTA) · Edge histogram descriptor (EHD)
1 Introduction Nowadays, various sources produce huge amount of image collections, generating a challenge to computer systems for storing, transmission and indexing of such large image data for easy access. For retrieval of desired image from large database, CBIR is used [45], [27]. All CBIR techniques work on extraction of query image attributes and matching these with database attributes. A traditional retrieval method which uses keywords for retrieval is text-based image retrieval. This is very sensitive to the keywords and not dependent on image contents [12]. In CBIR methods, image information in feature descriptors are extracted from the image. Low-level features such as color, shape and texture are extracted which assumes various perceptions. The main goal is to enhance efficient algorithm to improve the CBIR retrieval performance using the combination of features.
B 1
Yogita D. Mistry [email protected] Department of Electronics Engineering, Ramrao Adik Institute of Technology Nerul, Mumbai, Maharashtra 400706, India
Retrieval feature classification includes global features and local features. Global features extract information from the entire image, whereas local features operate locally which are more focused on the key points in images [54]. Still, because of the various reasons, the retrieval rate of CBIR algorithm is limited. One approach is to incorporate highlevel features using machine learning such as relevance feedback algorithms [37] and dictionary learning [53]. Color descriptor is the most important descriptor of the input image which is robust against the size change, direction change and resolution of image [11]. Color moments are represented in various spaces i
Data Loading...