It is of great significance to detect and predict muscle fatigue for avoiding muscle injury and the negative effects on human-machine interface. This paper presents a novel approach to obtain reliable information about muscle fatigue by analyzing surface electromyography (sEMG) and near-infrared spectroscopy (NIRS) simultaneously. Muscle fatigue was induced via sustained isometric contraction at 50% of maximal voluntary contraction (MVC) force with six subjects. Four fatigue metrics, root mean square (RMS) and median frequency (MDF) of sEMG, blood volume (BV) and muscle oxygenation (ΔHbO2) extracted from NIRS, were proposed to measure muscle fatigue from the perspective of electrophysiology and hemodynamics respectively. The experimental results suggested that sEMG and NIRS applied together could provide more detailed and reliable information about muscle fatigue, gaining a better understanding of fatigue process.