Resumen de Geometric and Structural-based Symbol Spotting. Application to Focused Retrieval in Graphic Document Collections
Marçal Rusiñol 
Usually pattern recognition systems consist in two main parts. On the one hand, the data acquisition and, on the other hand, the classification of this data on a certain category. In order to recognize which category a certain query element belongs to, a set of pattern models must be provided beforehand. An off-line learning stage is needed to train the classifier and offer a robust classification of the patterns. Within the pattern recognition field, we are interested in the recognition of graphics and, in particular, on the analysis of documents rich in graphical information. In the particular case of graphical symbol recognition, descriptors are extracted from the symbol to recognize and are subsequently matched with the set symbol models. In this context, one of the main concerns is to see if the proposed systems remain scalable with respect to the data volume so as it can handle growing amounts of symbol models. In order to avoid to work with a database of reference symbols, symbol spotting and on-the-fly symbol recognition methods have been introduced in the past years. Generally speaking, the symbol spotting problem can be defined as the identification of a set of regions of interest from a document image which are likely to contain an instance of a certain queried symbol without explicitly applying the whole pattern recognition scheme. Our application framework consists on indexing a collection of graphic-rich document images. This collection is queried by example with a single instance of the symbol to look for and, by means of symbol spotting methods we retrieve the regions of interest where the symbol is likely to appear within the documents. This kind of applications are known as focused retrieval methods. In order that the focused retrieval application can handle large collections of documents there is a need to provide an efficient access to the large volume of information that might be stored. We use indexing strategies in order to efficiently retrieve by similarity the locations where a certain part of the symbol appears. In that scenario, graphical patterns should be used as indices for accessing and navigating the collection of documents. These indexing mechanism allow the user to search for similar elements using graphical information rather than textual queries. Along this thesis we present a spotting architecture and different methods aiming to build a complete focused retrieval application dealing with a graphic-rich document collections. Different symbol descriptors encoding geometric and structural information are proposed in this thesis. These descriptors aim to describe parts of the symbols in a very compact and efficient way. Vectorial signatures, attributed strings and off-the-shelf shape descriptors are used to cluster parts of the symbols by similarity. Several strategies aiming to search for graphical information by similarity are used in this thesis. In order to retrieve locations from the document collection where parts of the symbols appear we use lookup tables and grid files indexed by graphical patterns. A final validation phase is introduced to validate the hypothetic locations where a symbol is likely to be found. This validation stage is formulated in terms of spatial and relational information. In addition, a protocol to evaluate the performance of symbol spotting systems in terms of recognition abilities, location accuracy and scalability is proposed. We show that the evaluation measures allow to determine the weaknesses and strengths of the methods under analysis. All the proposed contributions have been tested with an experimental scenario consisting of a collection of architectural drawings with its corresponding ground-truth
