Research on structure identification and design optimization of intelligent furniture products based on ResNet network

With the rapid development of smart home technology, smart furniture, as an important part of it, has gradually become a key factor in improving the quality of life and living comfort. The design and structural optimization of intelligent furniture products face many challenges in practical applications, such as how to realize efficient functional layout according to user needs, how to realize modularization and intelligence of products while ensuring beauty and practicality, etc. In order to improve the efficiency and accuracy of intelligent furniture design, this study proposes a method of intelligent furniture product structure identification and design optimization based on ResNet network. This method automatically identifies the structure of smart furniture products through deep learning technology, thus providing effective data support for subsequent design optimization. Firstly, ResNet network is used to perform deep feature extraction on existing intelligent furniture product images, and combined with the processing capabilities of convolutional neural network, the main structural features of furniture products are successfully identified. In terms of data sets, this study collected a training data set containing 5,000 images of different smart furniture products, and annotated and classified these images. By comparing ResNet models with different depths, the ResNet-50 furniture structure recognition reaches 92 %, which is obviously better than the traditional image processing method. By combining the identified structural features with user demand data, a design optimization algorithm is proposed, optimize the product on basis ensuring functionality of furniture. The optimized intelligent furniture design improves the space utilization rate by 15 % in function realization, and meets the needs of more than 90 % users in aesthetics. The novelty of this study lies in its dual approach that combines deep learning-based structure recognition with functional and cost-sensitive design optimization. Unlike previous research that focused solely on image classification or structural modeling, this work creates a practical feedback loop between data-driven recognition and adaptive design improvements in intelligent furniture.