In recent years, scientists have developed a new type of algorithm called convolutional neural network algorithm in the field of neural networks. This algorithm not only has a powerful image recognition function, but also can distinguish and arrange data images. At the same time, the algorithm's recognition and processing functions are also very powerful, able to identify relatively hidden images and process a very large image library in a short time. The research content of this article is the application and development of ceramic image creation based on the classification effect of neural network and the characteristics of quantum particle swarm algorithm. And according to the principles, standards, characteristics of neural network classification and the characteristics and technology of particle swarm algorithm, the traditional LB G algorithm and an improved LB G algorithm are discussed, and simulation experiments are carried out. During the experiment, the staff analyzed and optimized the specific process of the quantum particle swarm algorithm through a large number of calculations and simulation experiments. And according to the classification of neural network and quantum particle swarm algorithm, the researchers also proposed a set of practical ceramic image design methods. Through the inspection and comparison of the design results, the researchers preliminarily judged that the design method is not only practical, but also high Many advantages such as recognition, high accuracy and good visual experience. At the same time, the staff also optimized the method based on the preliminary design results.