Abstract
Abstract The paper presents a hybrid ensemble of diverse classifiers for logo and trademark symbols recognition. The proposed ensemble is composed of four types of different member classifiers. The first one compares color distribution of the logo patterns and is responsible for sifting out images of different color distribution. The second of the classifiers is based on the structural tensor recognition of local phase histograms. A proposed modification in this module consists of tensor computation in the space of the morphological scale-space. Thanks to this, more discriminative histograms describing global shapes are obtained. Next in the chain, is a novel member classifier that joins the Hausdorff distance with the correspondence measure of the log-polar patches computed around the corner points. This sparse classifier allows reliable comparison of even highly deformed patterns. The last member classifier relies on the statistical affine moment invariants which describe global shapes. However, a real advantage is obtained by joining the aforementioned base classifiers into a hybrid ensemble of classifiers, as proposed in this paper. Thanks to this a more accurate response and generalizing properties are obtained at reasonable computational requirements. Experimental results show good recognition accuracy even for the highly deformed logo patterns, as well as fair generalization properties which support human search and logo assessment tasks.
Highlights
Logo and trademark recognition concerns classification of usually artificially created images which content typically reflects name of a company, an organization, goods for sell, etc
The member classifiers were implemented with help of the composite design pattern which allows abstraction on building hierarchical structures of the contained objects (Cyganek and Siebert 2009)
Four member classifiers are used in the ensemble
Summary
Logo and trademark recognition concerns classification of usually artificially created images which content typically reflects name of a company, an organization, goods for sell, etc. The form of a logo or a trademark is commonly designed to attract people, and they constitute a specific group of images. These are characterized by specific colors, contrast or other attributes. The system architecture is based on plug-in modules which can be flexibly configured in either order Such interface allows setting serial, parallel or even hybrid output-input like connections of the classifiers. Some of the plug-ins can be turned off or new types of plug-ins can be added if necessary Such flexible architecture allows high recognition accuracy which compares or in some cases outperforms the known solutions, as will be discussed. Thanks to the used fast image processing methods, the whole system can perform in real-time, which is one of requirements of modern search tools
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
