Design of Multi-Species First Aid Kit Folding Gesture Library Based on Computer Vision Technology
As one of the general products in the field of medical equipment production,first aid kit is characterized by many varieties,wide range of uses and surge in demand.Aiming at the technical problems such as large labor load of manual folding,obvious quality fluctuation,low production efficiency,and lack of operation standards,etc.,in the production process of first aid kits,the characteristics of various types of first aid instruments and auxiliary materials yardage process are analyzed,which is mapped to the combination of technical processes of folding gestures.The innovations include:the introduction of Leap Motion binocular computer vision technology,the construction of two-handed spatial model and gesture model;convolutional neural network(CNN)is used to build a gesture key point recognition model to form an ordered set of folding gestures for first aid kits.Adopt graphical database technology to design and store the folding gesture library of each model of first aid kit.The feasibility of the design is verified through the production case of a certain model of first aid kit.The designed folding gestures of the first aid kit can be converted into a collection of folding operation commands for the robot,which realizes the intelligent manufacturing mode of machine-for-man in the field of first aid kit production.
First aid kitGesture recognitionLeap MotionComputer visionGesture libraryConvolutional neural network(CNN)
NETL
NSTL
万方数据
