Description: The research paper Retrieval of Trademark Images comments on retrieval of trademark images based on shape. The research paper talks about the need to have shape feature that is able to capture the human perceptual similarity in a better way. To achieve this, an automatic retrieval algorithm needs to be developed and it’s a very difficult task. Considering the amount of time and effort that goes into a manual retrieval, the automatic retrieval algorithm seems to be a better option in many ways.
In the research paper is suggested a method for trademark image database retrieval based on object shape information that would supplement traditional text based retrieval system. The proposed system achieves both the desired efficiency and accuracy using a two stage hierarchy: in the first stage simple and easily computable shape features are used to quickly browse through the database to generate a moderate number of possible retrievals when a query is presented and in the second stage, images from the first stage are screened using a deformable template matching process to discard spurious matches. It is suggested in the research paper that textual features such as filenames, captions, keywords have been used to annotate and retrieve images, but research says that there are many problems associated with these traditional methods. Various schemes have been proposed for shape representation and retrieval. These include shape representation using polygonal approximation of shape and matching using the polygonal vertices, image representation on the basis of strings , comparing the images using the hausdorff distance .
Conclusion: The research paper concludes upon a note that an effective trademark retrieval algorithm has been developed. Efficiency can be improved by designing a two tier hierarchical retrieval system.
Download Retrieval of Trademark Images Technical Paper Presentation.