Lines of Research

• Automatic detection of lesions in digital mammographs
• Tecniques of computer vision and artificial intelligence in mammographs
• Automated search for clinical cases in public mammograph databases
• User interfaces in working with mammograph images
• Integration of echographic images and magnetic resonance

The Computer and Robotic Vision Group develops computer vision systems and image processing in real time that may be used in the recognition of objects, patterns, colours and shapes; in the monitoring of movements or the tri-dimensional reconstruction of objects and scenes. The main areas of application are: the quality control of industrial products, safety and biometric applications, and aid in medical diagnosis.

The computer vision systems are the main sources of information of the robots developed by the group. Some robots contribute to the mobility and independence of invalid people and others are independent and tele-directed vehicles used in the field of research and industry. As an example, the group has developed underwater observation robots that may be employed in underwater observation over long periods as well as in the collection of samples. The group has also designed automated ground vehicles that may be used in carrying goods and in security operations in industrial warehouses.

In the medical sphere the Computer and Robotic Vision Group has collaborated since 1996 with the radiological and oncological teams of the Dr Josep Trueta University Hospital of Girona as a result of the granting of a joint research project financed by the Health Research Fund (FIS), the aim being to work on the application of computer vision techniques in the detection of micro-calcifications in mammographies of doubtful malignity.

Since then the group has been developing other lines of research jointly with various medical teams, concentrating on the following:

     •    Automatic detection of lesions in digital mammographs

     •    Integration of techniques of computer vision and artificial intelligence in the    
          emission of a second computerised reading of mammographs

     •    Creation of a computer tool for the automated search for clinical cases in public
          mammograph databases

     •    Development of a user interface for radiographers working with mammographic
          images

     •    Integration of echographic and magnetic resonance images