Bow image retrieval method based on SSD target detection. Issue 17 (25th February 2021)

• Record Type: Journal Article
• Authors: Liao, Kaiyang; Fan, Bing; Zheng, Yuanlin; Lin, Guangfeng; Cao, Congjun
• Is Part Of: IET image processing. Volume 14:Issue 17(2020)
• Page Start: 4441
• Page End: 4449
• Publication Date: 2021-02-25
• Subjects: vectors -- feature extraction -- image representation -- object detection -- image classification -- image matching -- image retrieval -- convolutional neural nets; bow image retrieval method -- query image -- reference image -- database image -- single‐shot MultiBox detector target detection -- SSD target detection model -- labelled training gallery -- target object SSD model -- similar target regions -- target region information -- convolutional features -- image similarity -- Paris6k -- Paris106k -- convolution features -- 512‐dimensional vector characterisation images -- retrieval mean average precision -- 256‐dimensional features -- recent optimal CroW method -- query expansion rearrangement method -- Paris5k -- image retrieval framework -- image retrieval accuracy; Image processing -- Periodicals; 621.36705
• DOI: 10.1049/iet-ipr.2020.0478
• Languages: English
• ISSNs: 1751-9659

Abstract

Abstract : The query image is usually a simple and single object in image retrieval, and the reference images in the database usually have many distractions. The precision of image retrieval can be greatly improved If the target regions in the database image are extracted during retrieval. So this paper proposes a Bow image retrieval method based on SSD target detection. First, the training gallery is manually annotated to record the location and size information. Second, the SSD target detection model is trained with the labeled training gallery to obtain the target object SSD model. Third, the SSD model is used to locate the similar target regions of the reference image and the query graph. Finally, the target region information is mapped into the convolutional features, and these feature vectors are used for image similarity matching. The performance of the proposed method is evaluated on Paris6k, Oxford5k, Paris106k and Oxford105k databases. The experimental results show that the accuracy of image retrieval will be greatly improved by adding optimization methods in the proposed image retrieval framework. The image retrieval accuracy of this method is higher than that of similar methods in recent years.