David Pradas Fernandez

ON THE INFLUENCE OF WORD REPRESENTATIONS FOR HANDWRITTEN WORD SPOTTING IN HISTORICAL DOCUMENTS

Word spotting is the process of retrieving all instances of a queried keyword from a digital library of document images. In this paper we evaluate the performance of different word descriptors to assess the advantages and disadvantages of statistical and structural models in a framework of query-by-example word spotting in historical documents. We compare four word representation models, namely sequence alignment using DTW as a baseline reference, a bag of visual words approach as statistical model, a pseudo-structural model based on a Loci features representation, and a structural approach where words are represented by graphs. The four approaches have been tested with two collections of historical data: the George Washington database and the marriage records from the Barcelona Cathedral. We experimentally demonstrate that statistical representations generally give a better performance, however it cannot be neglected that large descriptors are difficult to be implemented in a retrieval scenario where word spotting requires the indexation of data with million word images.

Handwritten word spotting in old manuscript images using a pseudo-structural descriptor organized in a hash structure

There are lots of historical handwritten documents with information that can be used for several studies and projects. The Document Image Analysis and Recognition community is interested in preserving these documents and extracting all the valuable information from them. Handwritten word-spotting is the pattern classification task which consists in detecting handwriting word images. In this work, we have used a query-by-example formalism: we have matched an input image with one or multiple images from handwritten documents to determine the distance that might indicate a correspondence. We have developed an approach based in characteristic Loci Features stored in a hash structure. Document images of the marriage licences of the Cathedral of Barcelona are used as the benchmarking database.

VoIP cross-layer load control for hybrid satellite-WiMAX networks

Hybrid architectures including satellites are of undoubted interest for wireless networks deployment in rural and remote areas. We focus on VoIP capacity service optimization over a hybrid network and we compare performance under temperate and tropical propagation conditions. We have chosen a (GEO) satellite and wireless network combination, based on DVB-S2/RCS and WiMAX standards. This combination has already proven to be a relatively simple hybrid combination with enormous potential at both the service and architecture levels. In order to tackle the design of the hybrid network on a difficult propagation/topographical environment, we propose a cross-layer design across both layers and networks. Our proposal is based on commercially available VoIP codecs, and the cross-layer design does not require breaking the protocol stacks of both networks. We propose a centralized scenario where VoIP load control is performed through a physical-transport-application cross- layer information flow (inherently cross-layer across both networks). In particular, VoIP rate is centrally controlled in order to match the satellite link capacity drops due to physical layer adaptation to channel conditions. Realistic modeling is considered for both the hybrid architecture and the channels. We present numerical results of the proposed design showing that our cross-layer load control technique is able to maintain the active WiMAX number of connections for a wide range of rain attenuation in both temperate and tropical areas.

Unsupervised Wall Detector in Architectural Floor Plans

Wall detection in floor plans is a crucial step in a complete floor plan recognition system. Walls define the main structure of buildings and convey essential information for the detection of other structural elements. Nevertheless, wall segmentation is a difficult task, mainly because of the lack of a standard graphical notation. The existing approaches are restricted to small group of similar notations or require the existence of pre-annotated corpus of input images to learn each new notation. In this paper we present an automatic wall segmentation system, with the ability to handle completely different notations without the need of any annotated dataset. It only takes advantage of the general knowledge that walls are a repetitive element, naturally distributed within the plan and commonly modeled by straight parallel lines. The method has been tested on four datasets of real floor plans with different notations, and compared with the state-of-the-art. The results show its suitability for different graphical notations, achieving higher recall rates than the rest of the methods while keeping a high average precision.

A Coarse-to-Fine Approach for Handwritten Word Spotting in Large Scale Historical Documents Collection

In this paper we propose an approach for word spotting in handwritten document images. We state the problem from a focused retrieval perspective, i.e. locating instances of a query word in a large scale dataset of digitized manuscripts. We combine two approaches, namely one based on word segmentation and another one segmentation-free. The first approach uses a hashing strategy to coarsely prune word images that are unlikely to be instances of the query word. This process is fast but has a low precision due to the errors introduced in the segmentation step. The regions containing candidate words are sent to the second process based on a state of the art technique from the visual object detection field. This discriminative model represents the appearance of the query word and computes a similarity score. In this way we propose a coarse-to-fine approach achieving a compromise between efficiency and accuracy. The validation of the model is shown using a collection of old handwritten manuscripts. We appreciate a substantial improvement in terms of precision regarding the previous proposed method with a low computational cost increase.

Contextual word spotting in historical manuscripts using Markov logic networks

Natural languages can often be modelled by suitable grammars whose knowledge can improve the word spotting results. The implicit contextual information is even more useful when dealing with information that is intrinsically described as one collection of records. In this paper, we present one approach to word spotting which uses the contextual information of records to improve the results. The method relies on Markov Logic Networks to probabilistically model the relational organization of handwritten records. The performance has been evaluated on the Barcelona Marriages Dataset that contains structured handwritten records that summarize marriage information.

On Influence of Line Segmentation in Efficient Word Segmentation in Old Manuscripts

The objective of this work is to show the importance of a good line segmentation to obtain better results in the segmentation of words of historical documents. We have used the approach developed by Manmatha and Rothfeder to segment words in old handwritten documents. In their work the lines of the documents are extracted using projections. In this work, we have developed an approach to segment lines more efficiently. The new line segmentation algorithm tackles with skewed, touching and noisy lines, so it is significantly improves word segmentation. Experiments using Spanish documents from the Marriages Database of the Barcelona Cathedral show that this approach reduces the error rate by more than 20%.

Multicast Transmission Optimization over Hybrid DVB-SH Systems

Research interest related to hybrid networks for mobile users has rapidly evolved in order to extend service to all users. In particular, DVB-SH provides an efficient way of carrying multimedia services over hybrid satellite and terrestrial networks for a big diversity of terminals. Also a variety of possible hybrid system architectures can be chosen due to the two different physical layers specified: ODFM and TDM. Although DVB-SH has been designed for broadcast transmission, we propose to use it for multicast applications, which is more competitive over satellite systems. New system techniques implemented in the hybrid scheme, such as ACM and resource allocation, make the system strongly dependent on the channel conditions. Thus the capacity optimization in a multicast scenario becomes a powerful research challenge, mainly in mobile scenarios. In this paper, we aim at optimizing the multicast transmission over the satellite link and the terrestrial link of a DVB-SH based system. The architecture proposed includes resource allocation techniques depending on the scenarios considered (mobility and transmission mode). We first investigate a Hierarchical Allocation algorithm based on SC (HA-SC) for the OFDM air interface of DVB-SH, which is constrained to serve a minimum rate to all users. We later propose a Time Scheduling (TS) algorithm, more appropriate for TDM satellite-based systems, which finds the optimal number of users to be served in order to achieve the highest throughput in the multicast group. We show that with the TS algorithm, the spectral gain is 20-40% with respect to the worst case user. Moreover, the HA-SC is more suitable for the satellite since it outperforms TS in 300%.

Cross-layer rate allocation of multicast transmission over hybrid DVB-SH

DVB-SH has been demonstrated to provide an efficient way of carrying multimedia services over hybrid satellite and terrestrial networks for a big diversity of terminals. Moreover, a variety of possible hybrid system architectures can be chosen due to the two different physical layers specified: ODFM and TDM. Although DVB-SH has been designed for broadcast transmission, we propose to use it for multicast applications, which is more competitive over satellite systems. New system techniques implemented in the hybrid scheme, such as ACM and resource allocation, make the system strongly dependent on the channel conditions. Thus the capacity optimization in a multicast scenario becomes a powerful research challenge, mainly in mobile scenarios. In this paper, we aim at optimizing the multicast transmission over the satellite link and the terrestrial link of a DVB-SH based system. The architecture proposed includes resource allocation techniques adapted according to cross-layer information, and depending on the scenarios considered (users mobility and transmission mode). We first investigate a Hierarchical Allocation algorithm based on SC (HA-SC) for the OFDM air interface of DVB-SH, which is constrained to serve a minimum rate to all users. We later propose a Time Scheduling (TS) algorithm, more appropriate for TDM satellite-based systems, which finds the optimal number of users to be served in order to achieve the highest throughput in the multicast group. We show that TS algorithm performs successfully compared to the worst case user (broadcast). On the other hand, HA-SC is more suitable for the satellite since it outperforms TS.

Cost-Efficient Design of Hybrid Network for Video Transmission in Tropical Areas

This paper aims at providing different cost-efficient solutions for the channel impairments in tropical areas. In order to extend service to isolated areas, we propose hybrid architecture based on DVB-S2/RCS+Wi-Fi networks. In this scenario the satellite channel is affected by deep rain events that do not allow ACM modes to protect data. Moreover, the delay in the ACM reaction to fade changes can affect the quality of the video transmission. In order to avoid the QoS reduction, we focus on different fading mitigation techniques (FMT), designing a multi-layer scheme protection, using LL-FEC and AL-FEC in higher layers, and optimizing the intervention threshold and super-frame length in the physical layer.