Research on The Identification of Building Feature Using Image Recognition Technology

Abstract

Each building has its own features. To find the impact on attracting customers and tenant demand, from each feature of buildings such as facade, image recognition technology is useful to determine the number of features from image of buildings. We used convolutional neural networks of AI to learn and separate the correct and incorrect images for the three categories of commercial facilities, office buildings, and hotels predicted and conducted a comparative study for the correct and incorrect images. We found that there are unique characteristics of buildings by use, such as the ratio of glass area, the number of colors, the shape of the building, and the shape of the windows.