Abstract: In response to the issues of waste equipment disposal and the cultivation of students’ practical abilities in colleges and universities, this paper proposes a teaching design based on “waste equipment disassembly, image recognition, and innovative reassembly”. This design aims to fully utilize the residual value of waste equipment and enhance engineering students’ practical operation, theoretical application, and problem-solving abilities. The teaching process is divided into three stages: “Extensive Disassembly”, “Extensive Observation”, and “Versatile Innovation”. The “Extensive Disassembly” stage allows students to repeatedly disassemble equipment to gain a deep understanding of its internal structure and principles. The “Extensive Observation” stage uses artificial intelligence technology to assist in fault identification, cultivating students’ ability to quickly and accurately locate faults and master diagnostic methods. The “Versatile Innovation” stage exercises students’ comprehensive application of knowledge, creative thinking, and practical operation through innovative design of disassembling original components and reassembling them, achieving the transformation from theory to application. Compared with traditional equipment disassembly and assembly teaching, this design constructs an iteratively updatable cloud-based image library on the basis of disassembly and assembly, and achieves fault categorization through image recognition technology. It further encourages cross-system reconfigurative innovation, thereby enhancing the efficiency of practical operations. Meanwhile, this approach effectively uses waste equipment, and improves students’ hands-on and innovative thinking abilities, laying a foundation for addressing future engineering challenges.