Detection device, detection system, detection method, and computer program

Abstract

Provided is a technology that reduces the effects of changes in the imaging state of an imaging device on a background image to make it possible to detect a mobile body from a captured image. A detection device according to the present invention acquires a training image background model from a background generation device that has used situation information that represents the situation in a monitoring area when images of the monitoring area were captured to generate the training image background model by means of artificial intelligence (AI), uses the background model and an image that has been captured of the monitoring area to detect a mobile body from the captured image, and outputs detection information that indicates that the mobile body was detected.

Claims

1. A detection device comprising: background acquisition means for acquiring a background model of a training image from a background generation device that generates a background model of a training image using AI (Artificial Intelligence) technology with respect to a situation information representing the state of the monitoring area when the monitoring area is photographed; detection means for detecting a moving object from the photographed image using the photographed image of the monitoring area and the background model; and output means for outputting detection information indicating that the moving object has been detected.

2. The detection device according to claim 1, wherein the detection means further comprises a tracking means for detecting the moving object from each of a plurality of captured images taken at different times, and for tracking the moving object in the plurality of captured images, and the output means outputs a predetermined set of captured images and the tracking results of the moving object as detection information to a display device.

3. The detection device according to claim 2, characterized in that the output means outputs detection information that causes the display device to display an image on which the tracking results are superimposed on the predetermined captured image.

4. The detection device according to claim 1, further comprising classification means for classifying whether the moving body is one of the predetermined detection targets or not, wherein the output means outputs detection information including the classification result of the moving body.

5. The detection device according to claim 1, further comprising information acquisition means for acquiring character information describing the detection information from a large-scale language model, wherein the output means outputs the character information.

6. The detection device according to claim 2, further comprising a function that monitors changes in the range of the moving object in a plurality of captured images taken at different times, and outputs warning information when it is determined that the moving object is approaching based on predetermined criteria.

7. The detection device according to claim 1, further comprising a receiving means for receiving an image of the moving object from another imaging device when the detection means detects the moving object from an image captured by one imaging device, wherein the output means further outputs the image received from the other imaging device.

8. A detection system comprising: a detection device according to claim 1; an imaging device for transmitting the captured image to the detection device; and a display device for outputting the detection information of the detection device.

9. A detection method comprising: obtaining a background model of a training image from a background generation device that generates a background model of a training image using AI (Artificial Intelligence) technology with contextual information representing the state of the monitoring area when the monitoring area is photographed by a computer; detecting a moving object from the photographed image using the photographed image of the monitoring area and the background model; and outputting detection information indicating that the moving object has been detected.

10. The detection method according to claim 9, further comprising the steps of detecting the moving object from each of a plurality of captured images taken at different times, and tracking the moving object in the plurality of captured images, wherein in the output step, a predetermined captured image and the tracking result of the moving object are output to a display device as detection information.

11. The detection method according to claim 10, characterized in that, in the output step, the detection information is output to display on the display device an image in which the tracking result is superimposed on the predetermined captured image.

12. The detection method according to claim 9, further comprising the step of classifying whether the moving object is one of the predetermined detection targets or not, wherein the output process outputs detection information including the classification result of the moving object.

13. The detection method according to claim 9, further comprising the step of obtaining character information describing the detection information from a large-scale language model, wherein the character information is output in the output step.

14. The detection method according to claim 10, further comprising a function that, in the detection step, monitors changes in the range of the moving object in a plurality of captured images taken at different times, and outputs warning information when it is determined that the moving object is approaching based on predetermined criteria.

15. A computer program for causing a computer to perform the following: background acquisition processing to acquire a background model of a training image from a background generation device that generates a background model of a training image using AI (Artificial Intelligence) technology with contextual information representing the state of the monitoring area when the monitoring area is photographed; detection processing to detect a moving object from the photographed image using the photographed image of the monitoring area and the background model; and output processing to output detection information indicating that the moving object has been detected.

16. The computer program according to claim 15, wherein the detection process further comprises detecting the moving object from each of a plurality of captured images taken at different times, tracking the moving object in the plurality of captured images, and the output process outputs predetermined captured images and the tracking results of the moving object as detection information to a display device.

17. The computer program according to 16, characterized in that the output processing outputs detection information that causes the display device to display an image on which the tracking results have been superimposed on the predetermined captured image.

18. The computer program according to claim 15, further comprising a classification process for classifying whether the moving object is one of the predetermined detection targets or not, wherein the output process outputs detection information including the classification result of the moving object.

19. The computer program according to claim 15, further comprising an information acquisition process for obtaining character information describing the detection information from a large-scale language model, wherein the output process outputs the character information.

20. The computer program according to claim 16, further comprising a function to monitor changes in the range of the moving object in multiple captured images taken at different times, and to output warning information when it is determined that the moving object is approaching based on predetermined criteria.

Images