| abstract |
The invention discloses a method and a device for quickly recognizing the damage type of coal based on deep learning technology, including collecting images of the damage type of coal, image preprocessing. The network image recognition model includes five structures: input layer, convolution layer, pooling layer, full connection and output layer. It consists of 1 input layer, 49 convolution layers, 2 pooling layers, 1 fully connected layer and 1 output layer is composed; the image recognition model of coal damage type is trained, the coal damage type image is recognized, and the damage type to which the coal image belongs is obtained. The invention can realize quantitative, safe, rapid and accurate identification of the damage type of coal. |