Image retrieval based on AND/OR-construction models
- PDF / 3,048,552 Bytes
- 28 Pages / 439.37 x 666.142 pts Page_size
- 8 Downloads / 171 Views
Image retrieval based on AND/OR-construction models Yin-Fu Huang 1
& Yun-Shin Hsieh
1
Received: 7 August 2019 / Revised: 22 June 2020 / Accepted: 26 June 2020 # Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract
With the rapid development of the Internet, finding desired images from numerous images has become an important research topic. In this paper, we propose an image retrieval system facilitating retrieval time and accuracy. Since the performance of image retrieval is deeply influenced by image features and retrieval methods. Five different types of features and five different methods are used to find the best combination for an image retrieval system. First, we segment out the main object in an image and then extract its features. Next, relevant features are selected from the original feature set for facilitating image retrieval, using the SAHS algorithm. Then, five methods based on AND/ORconstruction are proposed to build the image retrieval model, using the relevant features. Finally, the experimental results not only show that our methods are more effective than the other state-of-the-art methods but also present some observations never explored by the previous research. Keywords Data mining . Image retrieval . Locality sensitive hashing . Kmeans . Deep learning
1 Introduction With the rapid development of the Internet, the number of images on the Internet is increasing by millions. How to find desired images from numerous images has become an important research topic. In the early days, most image retrieval systems used text descriptions as search keys to find out desired images (i.e., text-based image retrieval). However, because text descriptions could be arbitrary, the manual classification of images and their text descriptions
* Yin-Fu Huang [email protected] Yun-Shin Hsieh [email protected]
1
Department of Computer Science and Information Engineering, National Yunlin University of Science and Technology, 123 University Road, Section 3, Touliu, Yunlin 640 Taiwan, Republic of China
Multimedia Tools and Applications
takes more time. Therefore, these search methods have gone out of date. Instead, content-based image retrieval methods were proposed based on the low-level features extracted from images. In this paper, we propose an image retrieval system for querying similar images on three large-scale image databases. The system is divided into an offline phase and an online phase. In the offline phase, we employ segmentation and main object detection to separate the main object from an image. Then, we extract features from the object and select relevant features using the SAHS algorithm. Next, five methods called LSH-based AND/OR-construction, Kmeans-based AND/OR-construction, Softmax classification, SDH + AND/OR-construction simulation, and DPLM+AND/OR-construction simulation are proposed to build the image retrieval model. Besides, we also use advanced image retrieval 1) to retrieve the extra images for the LSH-based and Kmeans-based methods and 2) to reduce t
Data Loading...
