Current Projects
Adao
ADAO is a collaborative project, funded through FP7-PEOPLE, between the Computer;Vision Centre and ITESOFT, a French company dedicated to the managing of information flow and the automatic processing of paper documents for businesses.The aim of this knowledge transfer and research program is to design a prototype of an innovative automatic system able to build up consolidated files by analyzing, understanding and gathering multimodal information from heterogeneous paper documents and electronic documents (e-mail, pad).
Docuread
The scientific objective or the project is the development of advanced image analysis and pattern recognition techniques. The final outcome can also be used in other application fields such as processing of administrative documents. The project follows three major research axes:Advanced OCR techniques, Graphics recognition and Handwriting analysis. In addition, the project aims at integrating the outcome of the research work in Interpretation of CAD documents and administrative documents in a banking framework.
Five centuries of marriages
This project is a long-term research initiative based on the data-mining of the “Llibres d’Esposalles” conserved at the Archives of the Barcelona Cathedral. This extraordinary data source comprises 244 books of marriage licenses records, with information of approximately 550.000 unions celebrated in over 250 parishes of the Diocese between 1451 and 1905. Its impeccable conservation is a miracle in a region where parish archives have undergone massive destruction.
Interactive Knowledge Extraction
The KEDIHC project is part of a coordinated project between the CVC and the Universidad de Valencia that aims at developing innovative technologies to implement such computer-assisted solutions that will be applied to difficult documents, specifically to historical documents, handwritten documents, unstructured documents, documents with heterogeneous contents, and handwritten music scores. In collaboration with Universidad Politécnica de Valencia
Scanplan
The main goal of this Eureka project is to extend a CAD software with a new functionality for scanning and recognizing paper floor plans in order to obtain an automatic 3D reconstruction of the structure of the building. Thus, the scientific objective or the project is the development of advanced image analysis and pattern recognition techniques that can be applied to floor plan interpretation. Within this context, the DOCUREAD project takes this objective as a starting point but extends it to a more generic framework so that the final outcome can also be used in other application fields such as processing of administrative documents
Past Projects
HuPerText
Over the last decade, electronic communication of information has become an important part of everyday life. Internet based applications such as e-mail and the World Wide Web, have provided the platforms needed to enable massive information exchange on a daily basis in an electronic format. Initially, most communications were based solely on encoded text, but as connection speeds became faster, images are nowadays habitually included in electronic documents along with the encoded text information transmitted.
Eina de Consulta d’ Expedients de frontera
Eina de Consulta d’ Expedients de frontera
Digital Mail Room
Efficient processing of digital (scanned) documents in phase of storage / before storing / in storage phase. To develop algorithms that allow the categorization of documents from the identification of visual elements (presence of logos, key words, structure, etc.).
SymbolRec
The main focus of the research of both groups is symbol recognition. Then, one of the goals of the project was to identify approaches or techniques where both groups had worked in the past and therefore, collaboration between researchers of both groups could lead to signifficant improvements in the field. Thus, this collaboration has yield to advances in 4 different approaches to shape and symbol recognition:
