Kitchen garbage impurity identification method based on target detection technology

The invention discloses a kitchen garbage impurity identification method based on a target detection technology, belongs to the field of computer vision, wherein aims at kitchen garbage image data, modeling is carried out based on a convolutional neural network, effective features are extracted by using the convolutional neural network, and impurity classification and positioning are carried out by using a classification layer and a regression calculation layer. The method mainly comprises the steps that establishing a kitchen garbage image library and pre-processing, and expanding the data scale in a data enhancement mode; constructing a target detection model based on a convolutional neural network; training the target detection model by using the kitchen garbage image library; and inputting the target image into the trained target detection model for judgment, and finally outputting the position and category of the impurity in the target image, thereby realizing automatic identification of the impurity in