Abstract: The present disclosure provides a deep learning-based stop control method and system for a high-speed train, and relates to the technical field of rail transit management and control. The method includes: obtaining a training data set; establishing a convolutional neural network (CNN); training and optimizing the CNN by using the training data set, to obtain an optimized CNN; obtaining actual running data of a to-be-controlled train; inputting the actual running data into the optimized CNN to obtain a stop position of the to-be-controlled train; determining whether the stop position of the to-be-controlled train is 0; and if the stop position of the to-be-controlled train is 0, outputting a breaking command; or if the stop position of the to-be-controlled train is not 0, performing the step of “obtaining actual running data of a to-be-controlled train”. The present disclosure can ensure accurate stop of a high-speed train without high costs.