Vital signs monitoring for smart clothing can provide comprehensive and personalized data support for health management. A vital signs monitoring system for non-tight-fitting garments is present, which senses the weak cardiopulmonary vibration signals of the human body by four embedded high-sensitivity piezoelectric sensors, and extracts vital parameters such as heart rate, heart rate variability, and respiratory rate. The extraction of vital parameters from these weak cardiopulmonary vibration signals involves steps such as the removal of motion artifacts, feature peaks localization, and vital parameters extraction. The heartbeat signal and the respiratory signal are separated by the IIR filter, and motion artifacts are removed by the Shannon energy envelope method. Feature peaks are located by setting appropriate thresholds, which are then used to calculate vital parameters such as heart rate, heart rate variability, and respiratory rate. Taking the synchronously acquired ECG data as the reference standard, the error of heart rate is 1 BPM and the error of heart rate variability is 16 ms. The vital signs monitoring system operates without direct contact with the skin and is suitable for non-tight-fitting garments.