Abstract: The structure examination support system 100 in the present example is provided with a GNSS receiver 120 that measures the current location, an image pick-up device 110 that photographs a target, and a calculating device 130 that controls the image pick-up device 110, where these are equipped in a moving body. The computing device 130 controls the image pick-up device 110 on the basis of the installation location of the target that has been set in advance, the current location that has been measured by the GNSS receiver 120, and the moving speed of the vehicle so as to repeat photography a number of times, and carries out an image recognition process on each of the plurality of images outputted from the image pick-up device 110, and thus selects an image where the target is photographed and that is appropriate for the examination.

Abstract: The structure examination support system 100 in the present example is provided with a GNSS receiver 120 that measures the current location, an image pick-up device 110 that photographs a target, and a calculating device 130 that controls the image pick-up device 110, where these are equipped in a moving body. The computing device 130 controls the image pick-up device 110 on the basis of the installation location of the target that has been set in advance, the current location that has been measured by the GNSS receiver 120, and the moving speed of the vehicle so as to repeat photography a number of times, and carries out an image recognition process on each of the plurality of images outputted from the image pick-up device 110, and thus selects an image where the target is photographed and that is appropriate for the examination.

Abstract: The purpose of the present invention is to provide a technology which adaptively switches video analysis methods according to a congestion situation within a video, and which carries out a suitable video analysis by a limited computational volume. Provided is an image processing device 100, wherein a congestion estimation unit 102 estimates a congestion situation within a video which is acquired from an image input unit 101. A control unit 103 switches methods for video analysis according to the congestion situation. In a situation of low congestion, an image analysis unit for use in normal situations 105 carries out the video analysis by a basic method. In a situation of high congestion, an image analysis unit for use in congested situations 106 carries out a high-precision video analysis, such as an analysis which spans a plurality of frames.

Abstract: The present invention provides a system for visualizing the congestion degree of an entire facility using a monitoring camera. The system detects a crowd from a monitoring camera video and derives the magnitude (number of people) of the crowd from the edge amount thereof, and thereby estimates a congestion state in a range of camera video. The system also calculates a human flow (amount of movement of people) on the basis of a number of passing people counter and the magnitude and movement speed of the crowd, etc., and estimates the number of people in an area other than the range of camera video on the basis of a modeled crowd behavior pattern. The model is constructed by selecting a distribution function and a parameter thereof by a statistical model selection technique such as a Bayesian Information Criteria (BIC). The congestion state is plotted on a map, allowing the congestion state of an area of interest to be grasped at once.

Abstract: The purpose of the present invention is to provide a technology which adaptively switches video analysis methods according to a congestion situation within a video, and which carries out a suitable video analysis by a limited computational volume. Provided is an image processing device 100, wherein a congestion estimation unit 102 estimates a congestion situation within a video which is acquired from an image input unit 101. A control unit 103 switches methods for video analysis according to the congestion situation. In a situation of low congestion, an image analysis unit for use in normal situations 105 carries out the video analysis by a basic method. In a situation of high congestion, an image analysis unit for use in congested situations 106 carries out a high-precision video analysis, such as an analysis which spans a plurality of frames.

Abstract: The present invention provides a system for visualizing the congestion degree of an entire facility using a monitoring camera. The system detects a crowd from a monitoring camera video and derives the magnitude (number of people) of the crowd from the edge amount thereof, and thereby estimates a congestion state in a range of camera video. The system also calculates a human flow (amount of movement of people) on the basis of a number of passing people counter and the magnitude and movement speed of the crowd, etc., and estimates the number of people in an area other than the range of camera video on the basis of a modeled crowd behavior pattern. The model is constructed by selecting a distribution function and a parameter thereof by a statistical model selection technique such as a Bayesian Information Criteria (BIC). The congestion state is plotted on a map, allowing the congestion state of an area of interest to be grasped at once.

Abstract: Provided is a suspicious person detection method. First, a normal similar facial image search is carried out. Next, facial images, which are detected automatically from the input images and specified manually, are specified to be determined. Next, similar faces are searched for limited time on a time axis on the database. Next, the number of search results that distance between the features is lower than predetermined value is calculated and it is determined that the number of appearances is large and a possibility of a prowling person is high if the number of cases is large, and otherwise a possibility of prowling person is low. Last, a similarity between a facial image of a pre-registered residents and a facial image of a person whose number is large is calculated, and it is re-determined that the person is residents if the similarity is high, regardless of the determination.

Abstract: Provided is a suspicious person detection method. First, a normal similar facial image search is carried out. Next, facial images, which are detected automatically from the input images and specified manually, are specified to be determined. Next, similar faces are searched for limited time on a time axis on the database. Next, the number of search results that distance between the features is lower than predetermined value is calculated and it is determined that the number of appearances is large and a possibility of a prowling person is high if the number of cases is large, and otherwise a possibility of prowling person is low. Last, a similarity between a facial image of a pre-registered residents and a facial image of a person whose number is large is calculated, and it is re-determined that the person is residents if the similarity is high, regardless of the determination.