As the population aging is aggravated, the detection and alarm of falls in elderly people are becoming more and more important. In order to improve the accuracy of fall detection, an intelligent fall detection method based on discrete feature is proposed. Through analyzing the human motion data, seven kinds of discrete features of human motion are proposed. A fall detection model based on BP neural network is established. The extracted discrete features are used as the inputs of the fall detection model, and the output of the model is the result of fall detection. Through the learning and training of the model, fall detection is achieved. The method verification and product application results indicate that the intelligent fall detection method based on discrete features can effectively distinguish fall and nonfall, improve the fall detection accuracy and reduce the false alarm rate and missing alarm rate.