Skip to Main Content
Cognitive aspects implemented in computational processes establish frames to model human actions relative to the human corresponding senses. Vision is one of them. To conceive appropriated artificial architectures, behavior manifestations and brain structural proprieties are intensively studied. Physiology is modeled in simplified schemas in order to be implemented in virtual systems and machine models. Eye tracking models offer clues for implementing artificial vision on robots, to give salient information for marketing, for art, or forensic research. If the task might be relatively simple for following one precise target in an image, the issues are more complicated when more objects have to be identified, in a multitask perception simulation. Another aspect is to take into consideration the semantic content and this depends on the detailed, rule-frames, formulation of the problem. A number of direct and reverse processes are identified: from real image to human vision and understanding, and from observation of the human behavior, to computer modeling and image processing. The paper makes a quick overview on the relevant studies connected to these topics and proposes a new schema to identify salient points in an image, based both on image processing techniques and on cognitive aspects, mainly from vision and multi-task perception studies.