The group of Digital Signal Processing and Geoinformation aims to create a space oriented to theoretical and applied research, which guarantee the insertion of added value in the systems developed mainly by the GEySED development center, in addition to promoting the integration of lines and reinforce the scientific work and the visibility of the center.
Related research topics:
- Image processing techniques. It includes techniques that aim to improve the quality of the digital image, according to a given criterion. Emphasis is placed on image enhancement and image restoration techniques.
- Techniques of analysis, description, and classification of images. It includes techniques that return the set of objects that somehow represent a digital image. Emphasis is placed on segmentation techniques, especially clustering based segmentation. It addresses the description of objects and classifies an object into a dimensioned set of possible categories.
- Recovery and search for content in videos. It includes the techniques of recovering objects of interest from video streams as well as the segmentation of video from the automatic extraction of keyframes. Emphasis is given to example-based recovery techniques (query-by-example) and semantic retrieval techniques. Ontologies for content retrieval.
- Optical character recognition.Includes the techniques that identify the character represented in a digital image. Emphasis is placed on specialized fields that aim to recognize vehicle registrations (ANPR), recognize container identifiers (CCR), and recognize characters from scanned documents.
- Voice and speech processing. It includes the techniques that allow the recognition of the speaker, included in a limited list of possible speakers, who has pronounced a sample of spoken text. Emphasizes techniques that convert spoken text to written text.
- Representation of Geospatial data. They are the different techniques and topological forms of representing information on a map. Defines the relationships that may exist between different spatial objects. Tools for topographic representation and contouring. It covers the techniques and algorithms for the 3D representation of structures and forms of the ground. Models of raster and vectorial spatial representation.
- Geospatial data analysis. Includes interpretation and analysis of geological data. It emphasizes the analysis of diverse types of deposits or behaviors of the geographical structures. It also studies the methods that allow the analysis of physical surfaces: gradient, visibility or appearance of such surfaces. Analysis of man-made natural or artificial networks as well as spatial location analysis. Processing of geospatial data for data inference.
- Photogrammetry and remote sensing. Includes image processing and analysis on the ground. The processing of satellite images and the automatic detection of structures in them. Stereoscopic vision techniques applied to geographic information systems.
- Computational geometry. Some particular classes of problems addressed include geometric search and retrieval, convexity, and other related topics, proximity, intersection, geometric decomposition, geometric object arrays, shape approximation, and visibility. The main application areas are Computer Graphics, Computer Aided Design and Manufacturing (CAD / CAM), Pattern Recognition, Computational Morphology, VLSI Design, Artificial Vision, Robotics and Geographic Information Systems.