Building a Recognition Process of Cooking Actions for Smart Kitchen System

摘要

Smart kitchen should be focusing its development on the actual interaction with users and the environmental objects rather than emphasizing on complicated instructions and feedback. Unfortunately, the current techniques can only be designed to identify motions and basic actions. The main purpose of this paper is to analyze and research user motions and actions involved in the process of cooking, including ingredient preparation, and to discover multiple action identification characteristics for the user and cooking utensils. By using the video analysis, ultimately, the project will use these characteristics to establish a reliable cooking-action database. Our study can distinguish between similar actions. The model is primarily used to identify, understand and differentiate the extent of the intellectuality of user motions. This model may be used in the future in the application to cooking support systems or other smart kitchen developments.