Image processing device
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.
1. An image processing device that extracts information from a video image shot by a video device, the image processing device comprising:
a controller that:
estimates a congestion degree in a shooting region in real time based on the video image;
extracts information on an object other than a background from the video image using a plurality of image analysis techniques including a first analysis technique for real-time processing and a second analysis technique for non-real-time processing, the information on the object including information relating to a region of a person and a region of a left object; and
selects one of the plurality of image analysis techniques to be used for image analysis according to the estimated congestion degree; and
an image recorder that records the video image,
wherein the controller acquires a video image for use in the image analysis from the image recorder according to historical information on a change in the congestion degree and performs image analysis using the second analysis technique.
2. The image processing device according to claim 1 ,
wherein a plurality of the video devices may be provided,
wherein respective congestion degrees in the video images are estimated for the plurality of video devices, and
wherein the controller selects one of the plurality of image analysis techniques to be used in the image analysis according to the respective congestion degrees in the video images from the plurality of video devices such that the second analysis technique is more prone to be selected as the congestion degrees are higher.
3. The image processing device according to claim 1 , wherein crowds in the video image are represented by a heat map based on the video image and measures sizes of the crowds and congestion degrees in the individual crowds.
4. The image processing device according to claim 1 ,
wherein the controller performs 1) a congestion-time video analysis that analyzes an input image with a high congestion degree and 2) a normal-time video analysis that analyzes an input image with a low congestion degree,
wherein each of the congestion-time video analysis and normal-time video analysis is a real-time image analysis,
wherein high computing performance is used to perform the congestion-time video analysis, and
wherein relatively low computing performance is used to perform the normal-time video analysis.