The human hand shows complex motor skills and is widely used in activities in daily living. Thus, finger movement recognition has many potential applications in rehabilitation, tele-operation, and prosthetic hands. Several surface electromyography (sEMG) interfaces have been developed to recognize finger motion without impeding finger movement. However, conventional interfaces are impractical because they require a large skin contact area, which inconveniences users and restricts their clothing. In this paper, we propose a metal microneedle-based high-density (MNHD) sEMG interface for finger movement recognition, which is smaller than a traditional wet electrode. The highly selective extraction of the sEMG signal was demonstrated in static finger flexion tests and fingertip force estimation.