Automatic Document Logo Detection

Zhu, Guangyu; Doermann, David

doi:10.1109/icdar.2007.4377038

Cited by 93 publications

(56 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…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). For the experiments the University of Maryland Logo Dataset from the Laboratory for Language and Media Processing (LAMP) of the University of Maryland was used (University of Maryland 2013; Doermann et al 1996;Zhu and Doermann 2007). The database contains 105 logo images which, for our experiments, were affinely converted to create the test patterns.…”

Section: Resultsmentioning

confidence: 99%

“…In other words, the answer how resistant the system is to rotation, change of scale, occlusions, as well as additive noise of the prototype patterns is examined. Doermann et al 1996;Zhu and Doermann 2007) In the paper by Yang et al (2007) only 20 images from the database were investigated. However, in our experiments all images from the University of Maryland were used.…”

Section: Resultsmentioning

confidence: 99%

“…However, this method is augmented by computation of the morphological scalespace, as will be discussed. Figure 4 depicts examples of the structural tensor components: T x x , T xy , T yy , as well as coherence c, phase ϕ, and magnitude m, computed for one of the test logo patterns belonging to the logo database prepared by the University of Maryland (University of Maryland 2013; Doermann et al 1996;Zhu and Doermann 2007). …”

Section: Finding Local Structures With the Structural Tensormentioning

confidence: 99%

See 2 more Smart Citations

Hybrid ensemble of classifiers for logo and trademark symbols recognition

Cyganek

2014

Soft Comput

View full text Add to dashboard Cite

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.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Resultsmentioning

confidence: 99%

Section: Finding Local Structures With the Structural Tensormentioning

confidence: 99%

See 1 more Smart Citation

Hybrid ensemble of classifiers for logo and trademark symbols recognition

Cyganek

2014

Soft Comput

View full text Add to dashboard Cite

show abstract

“…As noted in [8], one of the big challenges for the next years for the GR community is the localization/recognition of graphic symbols appearing in complete documents without any previous segmentation. To our best knowledge, in the literature, only Zhu and Doerman addressed in [11] the problem of logo spotting (i.e. the recognition and localization of logos in real documents) by means of a cascade of classifiers.…”

Section: Introductionmentioning

confidence: 99%

Logo Spotting by a Bag-of-words Approach for Document Categorization

Rusiñol

Lladós

2009

2009 10th International Conference on Document Analysis and Recognition

View full text Add to dashboard Cite

“…For example, Zhu and Doermann [20] propose a method for logo detection in document images that combines contextual information about the expected logo location in the document with geometrical features of extracted components (aspect ratio, spatial density and area). They use a Fisher classifier at different scales of the image and calculate performance by comparing the overlapping of the resulted logo regions with ground truth ones.…”

Section: Introductionmentioning

confidence: 99%

A polar-based logo representation based on topological and colour features

Nourbakhsh

Karatzas

Valveny

2010

Proceedings of the 9th IAPR International Workshop on Document Analysis Systems

View full text Add to dashboard Cite

In this paper, we propose a novel rotation and scale invariant method for colour logo retrieval and classification, which involves performing a simple colour segmentation and subsequently describing each of the resultant colour components based on a set of topological and colour features. A polar representation is used to represent the logo and the subsequent logo matching is based on Cyclic Dynamic Time Warping (CDTW). We also show how combining information about the global distribution of the logo components and their local neighbourhood using the Delaunay triangulation allows to improve the results. All experiments are performed on a dataset of 2500 instances of 100 colour logo images in different rotations and scales. KeywordsColour logo retrieval and logo representation

show abstract

Automatic Document Logo Detection

Cited by 93 publications

References 10 publications

Hybrid ensemble of classifiers for logo and trademark symbols recognition

Hybrid ensemble of classifiers for logo and trademark symbols recognition

Logo Spotting by a Bag-of-words Approach for Document Categorization

A polar-based logo representation based on topological and colour features

Contact Info

Product

Resources

About