The invention discloses an aerial display equipment fault detection method based on a convolutional neural network and a monitoring device, and the method comprises the following steps: 1, collectingan interface of aerial display equipment, and carrying out the pre-classification of a display interface; 2, designing a semantic comprehension model, constructing a convolutional neural network, andcarrying out semantic comprehension on a display interface of the aerial display equipment; 3, classifying the interfaces under the abnormal condition; 4, performing semantic understanding on the normally displayed aerial display interface, and detecting whether a fault occurs or not; 5, analyzing symbols and icons which cannot be understood; 6, sending the fault detection picture in combination with semantic comprehension information to a convolutional neural network for feature extraction and training; 7, optimizing and compressing the CNN fault classifier, and implanting the CNN fault classifier into SOM-RK3399 embedded equipment to obtain an aerial display monitoring device; and 8, enabling the monitoring device to collect an output signal of the aerial display device, form a display picture, send the display picture to the CNN processing module, and detect whether the aerial display device is in a normal working state.