Motion artifact is widely present in exercise electrocardiogram (ECG) signal, which is an important factor affecting the accuracy of QRS complex detection. A simple-fast QRS detection algorithm based on energy segmentation is proposed, which is suitable for exercise ECG measured by the wearable monitoring device. The proposed method mainly consists of four parts: the 15–25 Hz band-pass filter, the calculation of segmentation energy, the moving average filter and the detection of R-peaks. To enhance the information of the QRS complex, a band-pass filter is used to eliminate noises and suppress unwanted P/T waves. The information of R-R interval and QRS duration is used to energy segmentation, and moving average filter is used to obtain the adaptive energy threshold. Then, an algorithm combined adaptive energy threshold with amplitude threshold is used for QRS detection. The MIT-BIH arrhythmia database and the European ST-T database are chosen to value the correctness of the algorithm. The motion artifact contaminated ECG database is chosen to value the robustness. Exercise ECG signals obtained by lab are used to value the practicality of the algorithm. The simulation results show that the QRS detection algorithm has a sensitivity of 99.36%, a positive predictivity of 99.78%, an accuracy of 99.14%, and performs well even under ECG signal contaminated by strong motion artifacts.

运动伪影广泛存在于运动心电图（ECG）信号中，是影响QRS波群检测准确性的重要因素。提出了一种简单快速的基于能量分割的QRS检测算法，适用于可穿戴监测设备测量的运动心电图。该方法主要由四部分组成：15-25Hz带通滤波器、分割能量计算、移动平均滤波器和R峰检测。为了增强 QRS 复合波的信息，使用带通滤波器来消除噪声并抑制不需要的 P/T 波。利用RR间期和QRS时长信息进行能量分割，利用移动平均滤波器得到自适应能量阈值。然后，QRS检测采用自适应能量阈值与幅度阈值相结合的算法。选择MIT-BIH心律失常数据库和欧洲ST-T数据库来评价算法的正确性。选择运动伪影污染的 ECG 数据库来评估稳健性。实验室获得的运动心电信号用于评价算法的实用性。仿真结果表明，QRS检测算法的灵敏度为99.36%，阳性预测率为99.78%，准确率为99.14%，即使在受到强烈运动伪影污染的心电信号下也表现良好。实验室获得的运动心电信号用于评价算法的实用性。仿真结果表明，QRS检测算法的灵敏度为99.36%，阳性预测率为99.78%，准确率为99.14%，即使在受到强烈运动伪影污染的心电信号下也表现良好。实验室获得的运动心电信号用于评价算法的实用性。仿真结果表明，QRS检测算法的灵敏度为99.36%，阳性预测率为99.78%，准确率为99.14%，即使在受到强烈运动伪影污染的心电信号下也表现良好。