This study introduces a form of vision-based machinery surveillance for abnormal behavior detection in high-speed periodic operation. This new surveillance method includes high-frame-rate (HFR) video logging with a phase encoding algorithm (for use with periodic operation) to efficiently match input images with pre-stored reference images. To allow periodic machinery operations to be monitored, the surveillance algorithm can estimate the phase of a periodic operation by inspecting temporal changes in the brightness at several significant pixels in an input image from a single camera; abnormal behavior in periodic operation can be intelligently detected at a crucial moment by comparing the input image with the reference image synchronized with its encoded phase without the heavy computation required to search all the reference images for the matching process. This algorithm was implemented by software on a high-speed vision platform, IDP Express, which can record input images of 512 × 512 pixels and processed results at 1000 fps for video logging. We verified the effectiveness of the HFR-video-based machinery surveillance by performing two on-line experiments for high-speed periodic operation at dozens of hertz using a sewing machine.