Due to the large number of sensitive points of the flexible force sensitive sensor array, the output value of each sensitive point varies greatly. In order to solve the complex problem of sensor calibration, a fast calibration method based on BP neural network to establish a calibration classification model is given. The classification of sensitive points on the sensor array was calibrated. The BP neural network was used to train the classification model, and then the least squares method was used to fit each type of calibration curve. Through the validation sample, the classification accuracy of the classification model is over 98%. The experiment verifies that the calibration result of the algorithm is accurate. Calibration time is less than 1 s. The experimental results show that this method is ideal for calibration of largearea flexible force sensor, and it can be used to calibrate the flexible force sensor.