Abstract: In view of the wide application of strain sensors in human motion monitoring, health monitoring and other fields, it is important to design flexible strain sensors with high sensitivity and large strain range. In this paper, the quadrilateral and hexagonal mesh flexible strain sensors were prepared by template method based on Ecoflex-graphene composites. By comparing the strain range and tensile breaking limit of two different mesh sensors, it is found that the comprehensive performance of hexagonal mesh flexible strain sensor is more excellent, and under the condition of 80% strain, the tensile/releasing fatigue life detection is carried out. The sensor shows good reliability, and the sensor performs well in monitoring the elbow joint movement and different breathing conditions of the human body. A multi-channel detection system was constructed by combining hexagonal mesh flexible strain sensors to realize multiple gesture recognition, which has a broad market application prospect in the field of artificial intelligence and motion recognition.