A Study on Image Retrieval Methods

Please enable javascript in your browser.

Ahmadinejad, Reyhaneh | 2018

1675 Viewed

Type of Document: M.Sc. Thesis
Language: Farsi
Document No: 51438 (02)
University: Sharif University of Technology
Department: Mathematical Sciences
Advisor(s): Razvan, Mohammad-Reza; Kamali-Tabrizi, Mostafa
Abstract:
Image retrieval refers to the task of finding images related to a query image within an image set. Due to ever-increasing volumes of data, it has become increasingly necessary to find suitable and efficient methods for searching in massive databases. In this thesis, modern image retrieval techniques developed within the last 15 years have been studied, with an aim to satisfy three primary constraints of efficiency, accuracy, and low memory usage. Our focus has been on content-based retrieval; meaning that instead of using text and other information, we directly utilize image features for analysis and processing. To achieve this, we studied two established techniques, the bag-of-words model, and vector of locally aggregated descriptors (VLAD) encoding, and the combination with Convolutional Neural Networks. Furthermore, we developed a new method of VLAD normalization which we call cluster norm. Our results show that this technique has significantly higher performance characteristics than previously used methods
Keywords:
Image Retrieval ; Words Bag Model ; Scale Invariant Feature Transform (SIFT)Algorithm ; Vector Of Locally Aggregated Descriptors (VLAD) Method