Kelime eşleme yöntemi için yeni bir niteleme

ABSTRACTA NEW REPRESENTATION FOR MATCHING WORDSEsra AtaerM.S. in Computer EngineeringSupervisor: Assist. Prof. Pınar Duygulu SahinşJuly, 2007Large archives of historical documents are challenging to many researchers allover the world. However, these archives remain inaccessible since manual index-ing and transcription of such a huge volume is difficult. In addition, electronicimaging tools and image processing techniques gain importance with the rapidincrease in digitalization of materials in libraries and archives. In this thesis,a language independent method is proposed for representation of word images,which leads to retrieval and indexing of documents. While character recogni-tion methods suffer from preprocessing and overtraining, we make use of anothermethod, which is based on extracting words from documents and representingeach word image with the features of invariant regions. The bag-of-words ap-proach, which is shown to be successful to classify objects and scenes, is adaptedfor matching words. Since the curvature or connection points, or the dots areimportant visual features to distinct two words from each other, we make use ofthe salient points which are shown to be successful in representing such distinc-tive areas and heavily used for matching. Difference of Gaussian (DoG) detector,which is able to find scale invariant regions, and Harris Affine detector, whichdetects affine invariant regions, are used for detection of such areas and detectedkeypoints are described with Scale Invariant Feature Transform (SIFT) features.Then, each word image is represented by a set of visual terms which are obtainedby vector quantization of SIFT descriptors and similar words are matched basedon the similarity of these representations by using different distance measures.These representations are used both for document retrieval and word spotting.The experiments are carried out on Arabic, Latin and Ottoman datasets,which included different writing styles and different writers. The results show thatthe proposed method is successful on retrieval and indexing of documents even ifwith different scripts and different writers and since it is language independent,iiiit can be easily adapted to other languages as well. Retrieval performance of thesystem is comparable to the state of the art methods in this field. In addition,the system is succesfull on capturing semantic similarities, which is useful forindexing, and it does not include any supervising step.Keywords: word matching, document retrieval, bag-of-features.