Font Size: a A A

Research On Home Layout And Material Matching System Based On Deep Learning And Collaborative Filtering

Posted on:2020-04-08Degree:MasterType:Thesis
Country:ChinaCandidate:B Z LiuFull Text:PDF
GTID:2492306242954879Subject:Instrumentation engineering
Abstract/Summary:PDF Full Text Request
In the traditional interior home design field,the designer needs to match the owner with the complete set of household materials and layout according to the owner’s preferences and the actual size of the owner.So the designer is required to have a good professional training background and rich industry experience,which will inevitably lead to the high threshold of entry and long design period.In recent years,with the deepening of artificial intelligence theory research,it has become possible to design an intelligent home layout and material matching system.Based on interior design software and artificial intelligence technology,an intelligent home layout and material matching system is researched,which enables it to realize the intelligent layout of space and the intelligent matching of home materials.The main contents are as follows:(1)According to the characteristics of indoor home design work,the concept of the apartment segment has been studied,and expresses the complex home layout problem has been expressed through the form of the home material constraint tree,thence the indoor layout problem has been simplified.(2)In this thesis,the deep learning technique is studied,and the data enhancement method of expanding the dataset of home design scheme is proposed.The two-way LSTM neural network model for wall classification and recognition is constructed.The model training parameters are determined by contrast experiment method,and finally the pair is obtained.The single-sided segment recognition rate reaches 99% of the neural network,thus realizing the intelligent layout of the home scene.(3)In this thesis,the collaborative filtering algorithm has been studied,and the home material matching degree table is built according to the characteristics of the home material matching,the intelligent home material matching module is realized.Thus the problem of home material matching can be solved in an effectively way.By testing the explicit feedback and invisible feedback information obtained by the user after using the intelligent home layout matching system,an objective and comprehensive evaluation of the intelligent home layout matching system is made in this thesis which shows that the system can effectively bear part of the home design work.
Keywords/Search Tags:Smart home, Layout, Artificial neural network, Material matching
PDF Full Text Request
Related items