An extraction unit for extracting a human body area from a first image and a second image. A determination unit for setting a plurality of first areas of interest used in determining whether or not a first behavior is performed to the first image with an equal space therebetween, and determining on the basis of the plurality of areas of interest and the human body area whether or not a monitored person has performed the first behavior. The determination unit sets a second area of interest used in determining whether or not a second behavior is performed to the second image, and determines on the basis of the second area of interest and the human body area whether or not the monitored person has performed the second behavior.