Fuente: WIPO "processing tomato" The invention belongs to the technical field of computer vision and pattern recognition, deep learning and string type tomato maturity detection. Aiming at solve the problems that a traditional image processing algorithm is relatively low in tomato detection accuracy, relatively slow in reasoning speed and relatively poor in adaptability and accuracy to complex scenes, and current deep learning lacks in a string-type tomato detection method and image data set processing, the invention discloses an improved YOLOv5-based string-type tomato real-time detection method, based on a YOLOv5 thought, an attention mechanism is introduced to mine deeper and better image feature information, practical requirements can be met in the aspects of detection precision and reasoning speed, and application to artificial intelligence embedded equipment is achieved.