Rotary Sushi Chain Online Real - Time Management System

With the rapid development of service industry, we are in a big trend of modernization with advance information technology and modern management pattern adopted in this industry. Meanwhile, Service Science, Management and Engineering (SSME), as a new subject focusing on scientific methodology, has raised great concern. As the cornerstone of SSME, IT is playing a big part in the modernization of service sector.This thesis focus on a Japanese Sushi-go-round restaurant project, and study how we can make use of a combination of knowledge from computer science, network technology, image processing to improve the quality of their service.Based on the investigation in the sushi restaurants, we designed a ’Sushi real-time online management system’ to realize the automatic personalized management of the sushi chain.The system is composed of two parts:a server, which is responsible for database operations and data analysis, and several distributed image processing subsystems. The function description of each module and the key points of design are provided in details. It is pointed out that the effectiveness of such system is mainly affected by the image processing subsystems, which therefore became the main part of this thesis. In the experimental section, we compared three image segmentation algorithms with a sample set of640images. Based on the results, we proposed a new hybrid algorithm with combination of improved fast level set method and the global histogram thresholding method that can not only get better segmentation of the sushi images, but also meet the real-time requirement. Finally the segmentation results are passed to the pattern recognition module to verify the total performance of the image processing subsystems. The simulation results shows that the proposed image processing algorithm can make a good tradeoff between processing speed and recognition accuracy.