Elsevier

Pattern Recognition

Volume 42, Issue 3, March 2009, Pages 386-394
Pattern Recognition

Trademark image retrieval using synthetic features for describing global shape and interior structure

https://doi.org/10.1016/j.patcog.2008.08.019Get rights and content

Abstract

A trademark image retrieval (TIR) system is proposed in this work to deal with the vast number of trademark images in the trademark registration system. The proposed approach commences with the extraction of edges using the Canny edge detector, performs a shape normalisation procedure, and then extracts the global and local features. The global features capture the gross essence of the shapes while the local features describe the interior details of the trademarks. A two-component feature matching strategy is used to measure the similarity between the query and database images. The performance of the proposed algorithm is compared against four other algorithms.

Introduction

With the rapid increase in the amount of registered trademark images around the world, trademark image retrieval (TIR) has emerged to ensure that new trademarks do not repeat any of the vast number of trademark images stored in the trademark registration system. As the traditional classification of trademark images is based on their shape features and types of object depicted by employing manually assigned codes, faults or slips may appear because of different subjective perception of the trademark images. Evidence has been provided that the traditional classification is not feasible in dealing with a large fraction of trademark images with little or no representational meanings [1].

Trademarks can be categorised into a few different types. A trademark can be a word-only mark, a device-only mark or a device-and-word mark. For a word-only mark, the design of the trademark consists purely of text words or phrases. However, for a device-only mark, the trademark only contains symbols, icons or images. If a trademark comprises both words and any iconic symbols or images, it can be regarded as a device-and-word mark [2]. Since different algorithms have to be used in describing different kinds of trademark images, a trademark image retrieval system can only be designed to accommodate one of the types. Although several trademark image retrieval systems have been designed to handle all kinds of trademark images, the performance of these systems are rather unfavourable when compared to those systems that are specifically designed to handle only one kind of trademark. Another challenge in trademark image retrieval is the difficulty in modeling human perception about similarity between trademarks. As human perception of an image involves collaboration between different sensoria, it is in fact difficult to integrate such human perception mechanisms into a trademark image retrieval system.

The contributions of this paper are summarized as follows: (1) novel algorithms are proposed to describe the shape of device-only marks and device-and-word marks; (2) a two-component feature matching strategy is applied to compare global and local features; (3) this study not only evaluates the proposed method, but also investigates the retrieval performance of another four algorithms.

The rest of this paper is organized as follows. Section 2 reviews the related studies regarding the existing trademark image retrieval systems and techniques. Section 3 provides an overview of the proposed system architecture. Section 4 presents the algorithms proposed for extracting global and local features of trademarks. Section 5 describes a two-component matching strategy for measuring similarity between trademarks. Section 6 evaluates the performance and analyses the results. Finally, conclusions are drawn in Section 7.

Section snippets

Existing trademark retrieval systems

There are several remarkable trademark image retrieval systems that have been developed in recent years. TRADEMARK [3], STAR [4] and ARTISAN [5] are three of the most prominent trademark image retrieval systems. Different methodologies have been employed in these trademark systems. The TRADEMARK system uses graphical feature vectors (GF-vector) to interpret the image content automatically and calculates the similarity based on human perception [3]. The STAR system adopts mainstream

Overview of the proposed system

Our trademark image retrieval system consists of an offline database construction part and an online image retrieval part as shown in Fig. 1. The offline database construction part is intended to ensure high retrieval efficiency by extracting a feature set for each of the images in the database in an offline manner and storing the feature set along with its corresponding image in the database so that when a query image is presented to the system, the system does not have to perform online

Feature extraction

As shown in Fig. 1, feature extraction has to be done in both offline database construction and online image retrieval processes. Feature extraction is about extracting from an image a set of attributes/features that can feasibly describe/represent the image in order to facilitate the ensuing feature matching and ranking processes. The feature extraction stage in the proposed algorithm involves image pre-processing and feature representation as shown in Fig. 2. The purpose of the image

Two-component feature matching

Feature matching is about measuring the similarity between the feature vectors of the query image and the database images. An appropriate choice of feature matching techniques can enhance the system performance while an inappropriate choice may lead to unexpected results from the system even though an effective approach has been employed for feature representation. As trademark images can be structurally and conceptually alike with different interior details, during the feature matching stage,

Performance evaluation

In this section, the results obtained using the proposed algorithm under various conditions are presented. Precision and recall were computed for performance evaluation. The performance of the proposed algorithm is compared against two region-based descriptors: moment invariants [31] and Zernike moments only and two contour-based descriptors: Fourier descriptors [32], [33] and curvature scaled space (CSS) descriptors [32]. Some prior studies [34], [35] have reported that the combined

Conclusions

In this work, we have proposed a novel content-based trademark retrieval system with a feasible set of feature descriptors, which is capable of depicting global shapes and interior/local features of the trademarks. We have also proposed an effective two-component feature matching strategy for measure the similarity between feature sets. By utilising the curvature feature and the distance to centroid, the proposed algorithm is robust against rotation, translation, scaling and stretching. As for

About the Author—CHIA-HUNG WEI is currently an assistant professor of the Department of Information Management at Ching Yun University, Taiwan. He obtained his Ph.D. degree in Computer Science from the University of Warwick, UK, in 2008, and Master's degree from the University of Sheffield, UK, in 2000 and Bachelor degree from the Tunghai University, Taiwan, in 1996. His research interests include content-based image retrieval, digital image processing, medical image processing and analysis,

References (35)

Cited by (0)

About the Author—CHIA-HUNG WEI is currently an assistant professor of the Department of Information Management at Ching Yun University, Taiwan. He obtained his Ph.D. degree in Computer Science from the University of Warwick, UK, in 2008, and Master's degree from the University of Sheffield, UK, in 2000 and Bachelor degree from the Tunghai University, Taiwan, in 1996. His research interests include content-based image retrieval, digital image processing, medical image processing and analysis, machine learning for multimedia applications and information retrieval. He has published over 10 research papers in those research areas.

About the Author—YUE LI is currently a Ph.D. student with the Department of Computer Science at the University of Warwick, UK. He obtained his Master's degree from the University of Nottingham, UK, in 2005 and Bachelor's degree from the Nankai University, China, in 2003. His research interests include steganography, watermarking, multimedia security and content-based image retrieval. He has published a number of research papers in those research areas.

About the Author—WING-YIN CHAU graduated with a first-class degree from the Department of Computer Science at the University of Warwick, UK in 2007. During the final year of her study, her research project was considered as one of the few best projects of the year. Her research interests include content-based image retrieval, digital image processing and information retrieval.

About the Author—CHANG-TSUN LI received the B.S. degree in Electrical Engineering from Chung-Cheng Institute of Technology (CCIT), National Defense University, Taiwan, in 1987, the M.S. degree in Computer Science from U.S. Naval Postgraduate School, USA, in 1992, and the Ph.D. degree in Computer Science from the University of Warwick, UK, in 1998. He was an Associate Professor of the Department of Electrical Engineering at CCIT during 1999–2002 and a Visiting Professor of the Department of Computer Science at U.S. Naval Postgraduate School in the second half of 2001. He is currently an Associate Professor of the Department of Computer Science at the University of Warwick, UK, the Editor-in-Chief of the International Journal of Digital Crime and Forensics and an associate editor of the International Journal of Applied Systemic Studies (IJASS). He was involved in the organisation of a number of international conferences and workshops and also served as member of the international program committees for several international conferences. His research interests include content-based image retrieval, multimedia security, computational forensics, bioinformatics, image processing, pattern recognition and computer vision.

View full text