Abstract: The broad access to digital cameras, personal computers and Internet, has lead to the generation of large volumes of data in digital form. If we want an effective usage of this huge amount of data, we need automatic tools to allow the retrieval of relevant information. Image data is a particular type of information that requires specific techniques of description and indexing. The computer vision field that studies these kind of techniques is called Content-Based Image Retrieval (CBIR). Instead of using text-based descriptions, a system of CBIR deals on properties that are inherent in the images themselves. Hence, the feature-based description provides a universal via of image expression in contrast with the more than 6000 languages spoken in the world.
Nowadays, the CBIR is a dynamic focus of research that has derived in important applications for many professional groups. The potential fields of application can be such diverse as: the medical domain, the crime prevention, the protection of the intel- lectual property, the journalism, the graphic design, the web search, the preservation of cultural heritage, etc.
The definition on the role of the user is a key point in the development of a CBIR application. The user is in charge to formulate the queries from which the images are retrieved. We have centered our attention on the image retrieval techniques that use queries based on pictorial information. We have identified a taxonomy composed by four main query paradigms: query-by-selection, query-by-iconic-composition, query- by-sketch and query-by-paint. Each one of these paradigms allows a different degree of user expressivity. From a simple image selection, to a complete painting of the query, the user takes control of the input in the CBIR system.
Along the chapters of this thesis we have analyzed the influence that each query paradigm imposes in the internal operations of a CBIR system. Moreover, we have proposed a set of contributions that we have exemplified in the context of a final application.

Abstract: Sketch recognition is a discipline which has gained an increasing interest in the last
20 years. This is due to the appearance of new devices such as PDA, Tablet PC’s
or digital pen & paper protocols. From the wide range of sketched documents we
focus on those that represent structured documents such as: architectural floor-plans,
engineering drawing, UML diagrams, etc. To recognize and understand these kinds
of documents, first we have to recognize the different compounding symbols and then
we have to identify the relations between these elements. From the way that a sketch
is captured, there are two categories: on-line and off-line. On-line input modes refer
to draw directly on a PDA or a Tablet PC’s while off-line input modes refer to scan
a previously drawn sketch.
This thesis is an overlapping of three different areas on Computer Science: Pattern
Recognition, Document Analysis and Human-Computer Interaction. The aim of this
thesis is to interpret sketched documents independently on whether they are captured
on-line or off-line. For this reason, the proposed approach should contain the following
features. First, as we are working with sketches the elements present in our input
contain distortions. Second, as we would work in on-line or off-line input modes, the
order in the input of the primitives is indifferent. Finally, the proposed method should
be applied in real scenarios, its response time must be slow.
To interpret a sketched document we propose a syntactic approach. A syntactic
approach is composed of two correlated components: a grammar and a parser. The
grammar allows describing the different elements on the document as well as their
relations. The parser, given a document checks whether it belongs to the language
generated by the grammar or not. Thus, the grammar should be able to cope with
the distortions appearing on the instances of the elements. Moreover, it would be
necessary to define a symbol independently of the order of their primitives. Concerning to the parser when analyzing 2D sentences, it does not assume an order in the
primitives. Then, at each new primitive in the input, the parser searches among the
previous analyzed symbols candidates to produce a valid reduction.
Taking into account these features, we have proposed a grammar based on Adjacency Grammars. This kind of grammars defines their productions as a multiset
of symbols rather than a list. This allows describing a symbol without an order in
their components. To cope with distortion we have proposed a distortion model.
This distortion model is an attributed estimated over the constraints of the grammar and passed through the productions. This measure gives an idea on how far is the
symbol from its ideal model. In addition to the distortion on the constraints other
distortions appear when working with sketches. These distortions are: overtracing,
overlapping, gaps or spurious strokes. Some grammatical productions have been defined to cope with these errors. Concerning the recognition, we have proposed an
incremental parser with an indexation mechanism. Incremental parsers analyze the
input symbol by symbol given a response to the user when a primitive is analyzed.
This makes incremental parser suitable to work in on-line as well as off-line input
modes. The parser has been adapted with an indexation mechanism based on a spatial division. This indexation mechanism allows setting the primitives in the space
and reducing the search to a neighbourhood.
A third contribution is a grammatical inference algorithm. This method given a
set of symbols captures the production describing it. In the field of formal languages,
different approaches has been proposed but in the graphical domain not so much work
is done in this field. The proposed method is able to capture the production from
a set of symbol although they are drawn in different order. A matching step based
on the Haussdorff distance and the Hungarian method has been proposed to match
the primitives of the different symbols. In addition the proposed approach is able to
capture the variability in the parameters of the constraints.
From the experimental results, we may conclude that we have proposed a robust
approach to describe and recognize sketches. Moreover, the addition of new symbols
to the alphabet is not restricted to an expert. Finally, the proposed approach has
been used in two real scenarios obtaining a good performance.

Abstract: Despite the “paperless office” ideal that started in the decade of the seventies, businesses still strive against an increasing amount of paper documentation. Companies still receive huge amounts of paper documentation that need to be analyzed and processed, mostly in a manual way. A solution for this task consists in, first, automatically scanning the incoming documents. Then, document images can be analyzed and information can be extracted from the data. Documents can also be automatically dispatched to the appropriate workflows, used to retrieve similar documents in the dataset to transfer information, etc.
Due to the nature of this “digital mailroom”, we need document representation methods to be general, i.e., able to cope with very different types of documents. We need the methods to be sound, i.e., able to cope with unexpected types of documents, noise, etc. And, we need to methods to be scalable, i.e., able to cope with thousands or millions of documents that need to be processed, stored, and consulted. Unfortunately, current techniques of document representation, classification and retrieval are not apt for this digital mailroom framework, since they do not fulfill some or all of these requirements.
Through this thesis we focus on the problem of document representation aimed at classification and retrieval tasks under this digital mailroom framework. We first propose a novel document representation based on runlength histograms, and extend it to cope with more complex documents such as multiple-page documents, or documents that contain more sources of information such as extracted OCR text. Then we focus on the scalability requirements and propose a novel binarization method which we dubbed PCAE, as well as two general asymmetric distances between binary embeddings that can significantly improve the retrieval results at a minimal extra computational cost. Finally, we note the importance of supervised learning when performing large-scale retrieval, and study several approaches that can significantly boost the results at no extra cost at query time.

Abstract: Pattern recognition is the task that aims at distinguishing objects among different classes. When such a task wants to be solved in an automatic way a crucial step is how to formally represent such patterns to the computer. Based on the different representational formalisms, we may distinguish between statistical and structural pattern recognition. The former describes objects as a set of measurements arranged in the form of what is called a feature vector. The latter assumes that relations between parts of the underlying objects need to be explicitly represented and thus it uses relational structures such as graphs for encoding their inherent information. Vector spaces are a very flexible mathematical structure that has allowed to come up with several efficient ways for the analysis of patterns under the form of feature vectors. Nevertheless, such a representation cannot explicitly cope with binary relations between parts of the objects and it is restricted to measure the exact same number of features for each pattern under study regardless of their complexity. Graph-based representations present the contrary situation. They can easily adapt to the inherent complexity of the patterns but introduce a problem of high computational complexity, hindering the design of efficient tools to process and analyse patterns.
Solving this paradox is the main goal of this thesis. The ideal situation for solving pattern recognition problems would be to represent the patterns using relational structures such as graphs, and to be able to use the wealthy repository of data processing tools from the statistical pattern recognition domain. An elegant solution to this problem is to transform the graph domain into a vector domain where any processing algorithm can be applied. In other words, by mapping each graph to a point in a vector space we automatically get access to the rich set of algorithms from the statistical domain to be applied in the graph domain. Such methodology is called graph embedding.
In this thesis we propose to associate feature vectors to graphs in a simple and very efficient way by just putting attention on the labelling information that graphs store. In particular, we count frequencies of node labels and of edges between labels. Although their locality, these features are able to robustly represent structurally global properties of graphs, when considered together in the form of a vector. We initially deal with the case of discrete attributed graphs, where features are easy to compute. The continuous case is tackled as a natural generalization of the discrete one, where rather than counting node and edge labelling instances, we count statistics of some representatives of them. We encounter how the proposed vectorial representations of graphs suffer from high dimensionality and correlation among components and we face these problems by feature selection algorithms. We also explore how the diversity of different embedding representations can be exploited in order to boost the performance of base classifiers in a multiple classifier systems framework. An extensive experimental evaluation finally shows how the methodology we propose can be efficiently computed and compete with other graph matching and embedding methodologies.

Abstract: Searching for text objects in real scene images is an open problem and a very active computer vision research area. A large number of methods have been proposed tackling the text search as extension of the ones from the document analysis field or inspired by general purpose object detection methods. However the general problem of object search in real scene images remains an extremely challenging problem due to the huge variability in object appearance. This thesis builds on top of the most recent findings in the visual attention literature presenting a novel computational model of eye guidance aiming to better describe text object search in real scene images.
First are presented the relevant state-of-the-art results from the visual attention literature regarding eye movements and visual search. Relevant models of attention are discussed and integrated with recent observations on the role of top-down constraints and the emerging need for a layered model of attention in which saliency is not the only factor guiding attention. Visual attention is then explained by the interaction of several modulating factors, such as objects, value, plans and saliency. Then we introduce our probabilistic formulation of attention deployment in real scene. The model is based on the rationale that oculomotor control depends on two interacting but distinct processes: an attentional process that assigns value to the sources of information and motor process that flexibly links information with action.
In such framework, the choice of where to look next is task-dependent and oriented to classes of objects embedded within pictures of complex scenes. The dependence on task is taken into account by exploiting the value and the reward of gazing at certain image patches or proto-objects that provide a sparse representation of the scene objects.
In the experimental section the model is tested in laboratory condition, comparing model simulations with data from eye tracking experiments. The comparison is qualitative in terms of observable scan paths and quantitative in terms of statistical similarity of gaze shift amplitude. Experiments are performed using eye tracking data from both a publicly available dataset of face and text and from newly performed eye-tracking experiments on a dataset of street view pictures containing text. The last part of this thesis is dedicated to study the extent to which the proposed model can account for human eye movements in a low constrained setting. We used a mobile eye tracking device and an ad-hoc developed methodology to compare model simulated eye data with the human eye data from mobile eye tracking recordings. Such setting allow to test the model in an incomplete visual information condition, reproducing a close to real-life search task.