A low-resolution/high-framerate camera performs steady-basis imaging, and a server detects an area of a person, based on frame data from the low-resolution/high-framerate camera, and generates and outputs to a high-resolution/low-framerate camera imaging condition data. The high-resolution/low-framerate camera starts imaging based on the imaging condition data and outputs frame data to the server. The server performs trimming processing to extract from the frame data output from the high-resolution/low-framerate camera an area of a face of the person, recording of image data of the facial area, and sending of the image data to an external recognition device.