Moussavi, MahmoodFar, Behrouz HomayounSalehian-Dardashti, Soheil2018-04-302018-04-302018-04-25Salehian-Dardashti, S. (2018). Performance and Portability of Cloud-based Computer vision: A Software Case Study (Master's thesis, University of Calgary, Calgary, Canada). Retrieved from https://prism.ucalgary.ca. doi:10.11575/PRISM/31860http://hdl.handle.net/1880/106574The proliferation of visual sensors in terms variety and magnitude, has commenced an era where rich extensive visual data can be extracted and analyzed with the goal of a better understanding of the world around us. The forefront of this visual data analysis, is the field of computer vision that traditionally has been utilized in centralized systems but in recent years, has experienced an evolution towards a distributed cloud environment. The objective of this work is to analyze the software engineering processes that allow a rapid conversion of a centralized computer vision system to a cloud based distributed one. The work entails in developing the centralized algorithm in the specific realm of driver fatigue detection, analyzing its properties and designing an engineering methodology along with implementing the conversion from centralized to distributed computer vision. The focus of the thesis is on the implementation and real world metrics that represent the challenges in performance and portability attributes of most distributed computer vision systems. The proposed methodology’s merit is demonstrated through the blink detection use case study. Using requirement gathering techniques, the metrics are defined and compared at different stages of development to ensure minimal effort in porting the blink detection system. A generalizable message based architecture is proposed and implemented for a subclass of computer vision algorithms as a stepping stone for future analysis of other computational patterns that are associated with computer vision algorithms and systems.engUniversity of Calgary graduate students retain copyright ownership and moral rights for their thesis. You may use this material in any way that is permitted by the Copyright Act or through licensing that has been assigned to the document. For uses that are not allowable under copyright legislation or licensing, you are required to seek permission.Software EngineeringComputer ScienceEngineeringPerformance and Portability of Cloud-based Computer vision: A Software Case Studymaster thesis10.11575/PRISM/31860