Computer Vision and Image Processing

Computer vision is a discipline that encompasses methods for analyzing images and, in general, visual artefacts from the real world in order to produce numerical or symbolic information that could support understanding, enable to take decisions, etc.

Analysis of images is performed in several ways by the GRAINS group. For instance, natural images are considered for identifying subjects of interests, improve their contrast, etc. Images taken from microscope, telescope and, in particular, industrial cameras are used to extract features that can be useful to characterize specific details. The video flow originating from Unmanned Aerial Vehicles (UAV) and other robots is used to develop algorithms for adjustable autonomy.

Activity in the industrial field has produced the filing of several international patents, mainly in putting together algorithms of computer vision and special illumination techniques. Also images in the infrared part of the spectrum are considered, both in the near-infrared (NIR) and in the thermal infrared.

In particular, the GRAINS group has carried out significant activities on target tracking in forward looking infrared (FLIR) images. An overview of the many challenges that need to be faced in such a domain may be found in the Special Issue on Detection and Tracking of Targets in Forward Looking Infrared (FLIR) Imagery published by Sensors.