Focusing on the demand for teaching reform in digital image processing courses in universities, research on teaching methods for digital image processing aimed at enhancing innovation capabilities has been carried out. The existing problems in the teaching of digital image processing courses in universities have been analyzed, and specific plans for optimizing course content and improving teaching methods have been proposed. In response to the development of artificial intelligence, practical teaching cases for image pattern recognition have been designed. To cultivate students' practical and innovative abilities, a project-driven teaching method has been proposed, and corresponding comprehensive cases have been designed. The optimized course content and teaching methods have been tested in actual course teaching. In the future, the connection between industrial applications, scientific research, and teaching will be further strengthened to broaden students' horizons and cultivate their innovative capabilities.