Research On The Design Of Glasses Modeling Intelligent Recommendation System Matching Face Features

In the backdrop of high integration of information system and physical system,great changes have taken place in the industrial mode.Intelligent product design,manufacturing and sales mode have become an important way for the manufacturing industry to find breakthroughs.More and more manufacturing industries expect to rely on the Internet to improve the added value and competitiveness of their products.Glasses are a kind of common product in the market.They not only have the functions of correcting vision,shading,also can beautify the user's face.In order to improve the competitiveness of products,this paper studies the intelligent recommendation system of glasses modeling to match face features.The system can intelligently analyze the facial features and needs of users,establish the correlation model of facial features,eyeglasses features and perceptual evaluation.Furthermore,it can recommend the eyeglasses that meet the physiological and psychological needs of users and carry out personalized recommendation process relying on the Internet.First of all,this paper studies the face feature extraction based on image,and the qualitative evaluation is used to obtain the classification of the human face images.It compares the prediction effect of supervised learning and unsupervised learning neural network on the feature classification of the face images.The prediction model of face classification is constructed by the broader GRNN with good performance.Next,the recognition and classification methods of eyewear features are studied,and the prediction model of eyewear classification is built.Then,we use the Semantic Differential method in perceptual engineering to analysis the perceptual evaluation of the matching effect between glasses and faces.Meanwhile,we create the mapping model of face features,glasses features and perceptual evaluation with the quantitative theory I.After that,according to the user's ability to describe the abstract aesthetic,two input methods are proposed:perceptual word input method.Finally,we structure information of the glasses recommendation system,create a few wireframes and develop this system.