{"title":"A Novel and Efficient CBIR using CNN for Flowers","authors":"Subash. S. I, Muthiah. M. A., N. Mathan","doi":"10.1109/WiSPNET57748.2023.10134508","DOIUrl":null,"url":null,"abstract":"Image processing is vital to extract the required data from images. Machine learning is an efficient tool used for penetration in most of the classification and identification tasks performed by a computer. This project proposes the identification of a flower after the classification of flower images using a successful artificial intelligence tool named the Convolutional Neural Network (CNN). Models similar to this project have been used in most search engines for a long time, but CBIR (content-based image retrieval) still runs with less accuracy and produces outputs with fewer specifications due to the use of convolutional feed-forward networks for image retrieval. System performance depends a lot on the drawn-out features extracted from images. So, it is required to develop a CBIR system that retrieves similar images without explicit feature extraction and classification by using CNN, which accepts images as input. For experimentation, images from the Oxford-102 flower dataset are used.","PeriodicalId":150576,"journal":{"name":"2023 International Conference on Wireless Communications Signal Processing and Networking (WiSPNET)","volume":"66 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2023-03-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2023 International Conference on Wireless Communications Signal Processing and Networking (WiSPNET)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/WiSPNET57748.2023.10134508","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
Image processing is vital to extract the required data from images. Machine learning is an efficient tool used for penetration in most of the classification and identification tasks performed by a computer. This project proposes the identification of a flower after the classification of flower images using a successful artificial intelligence tool named the Convolutional Neural Network (CNN). Models similar to this project have been used in most search engines for a long time, but CBIR (content-based image retrieval) still runs with less accuracy and produces outputs with fewer specifications due to the use of convolutional feed-forward networks for image retrieval. System performance depends a lot on the drawn-out features extracted from images. So, it is required to develop a CBIR system that retrieves similar images without explicit feature extraction and classification by using CNN, which accepts images as input. For experimentation, images from the Oxford-102 flower dataset are used.