Environmental Art Design Element Identification and Mining Based on Deep Learning

Abstract

Identifying different design elements is a crucial step in environmental art design. These elements may include visual elements such as lines, shape, color, and texture or immaterial elements such as culture, history, and society. Data mining (DM) can help us find useful information hidden in a large amount of data. Deep learning (DL) possesses the capability to autonomously extract pertinent features from extensive datasets and discover the underlying structure and patterns within the data through hierarchical representations. From the change curve of the training accuracy and loss rate, the environmental art image classification method based on the integrated one-dimensional CNN showed good performance. This processing method not only helps to improve the quality and beauty of images but also provides more flexible and efficient solutions for a variety of image processing applications. Computer-aided design (CAD) systems can store and manage large amounts of design data, including design solutions, design elements, user feedback, etc. The combination with CAD technology further expands the application scope of DM in environmental art design.