The invention provides a rainy season land slide deformation monitoring method and a rainy season land slide deformation monitoring system. A data integration center is used to acquire a displacement value and a speed value of a monitoring point of a current sampling time point according to a displacement sensor and a speed sensor, and is used to acquire displacement values and speed values of the previous one or two time point/s, and two given error covariance matrixes, and is used to acquire a measurement noise covariance matrix estimation value and a process noise covariance matrix estimation value of the current sampling time point according to the ordinal number of the current sampling time point and the above mentioned parameters. Kalman filtering of the displacement value and the speed value of the current time point can be carried out by adopting the acquired matrix estimation values, and then the optimal displacement estimation value and the optimal speed estimation value of the current sampling time point can be acquired. The displacement value and the speed value can be the displacement weighted average and the speed weighted average. The two covariance matrixes can be estimated according to the actually acquired displacement value and the actually acquired speed value, and are not given, and therefore the matrix estimation values are real, and the filtering result is accurate. The invention also relates to the rainy season land slide deformation monitoring system.