Analysis of Vision systems and Taxonomy Formulation: An abstract model for generalization
2011 (English)Report (Refereed)
Vision systems are increasingly used in many applications including optical character recognition, mechanical inspection, automotive safety, surveillance and traffic monitoring. The current trend in vision systems is to propose solutions for specific problems as each application has different requirements and constraints. There is no generalized model or benchmark, to the best of our knowledge, which can be used for providing generic solutions for different class of vision systems. Providing a generic model in vision systems is a challenging task due to number of influencing factors. However, common characteristic can be identified in order to propose an abstract model. The majority of vision applications focus on the detection, analysis and recognition of objects. These tasks are reduced to vision functions which can be used to characterize the vision systems. In this report, we have analysed different types of vision systems, both wire and wireless, individual vision systems as well as a vision node in a Wireless Vision Sensor Network (WVSN). This analysis leads to the development of a system taxonomy, in which vision functions are considered as characteristics of the systems. The taxonomy is evaluated by using a quantitative parameter which shows that it covers 95 percent of the investigated vision systems and its flow is ordered for 50 percent of the systems. The proposed taxonomy will assist designers to classify their systems and enable researchers to compare their results with a similar class of systems. Moreover, it will help designers/researchers to propose generic architectures for different class of vision systems.
Place, publisher, year, edition, pages
Sundsvall: Mid Sweden University , 2011. , 47 p.
Research report in electronics, 1
system taxonomy, vision systems, smart cameras, wireless visual sensor networks, wireless visual sensor node
Electrical Engineering, Electronic Engineering, Information Engineering
IdentifiersURN: urn:nbn:se:miun:diva-15977Local ID: STCISBN: 978-91-86694-79-1OAI: oai:DiVA.org:miun-15977DiVA: diva2:507222