Surveillance camera and information processing apparatus
A surveillance camera includes an imaging unit configured to output a video frame, and a control unit configured to set a search range in the video frame and to extract a tracking-target image by using a feature quantity of an image in the search range. In a case where an object image other than the tracking target is contained in the search range, the control unit masks a feature quantity of the object image and extracts the tracking-target image.
1. A surveillance camera comprising:
a camera configured to output a video frame; and
a processor configured to:
set a search range in the video frame,
determine feature quantities of an image in the search range,
perform similarity calculations between the feature quantities of the image in the search range and feature quantities of a tracking-target image in a previous video frame, and
extract a tracking-target image in the video frame by using the similarity calculations,
wherein, in a case where a non-tracking target image is included in the search range, the processor masks feature quantities of the non-tracking target image, and extracts the tracking-target image in the video frame with the feature quantities of the non-tracking target image masked,
wherein the processor is configured to:
utilize a first masking intensity to mask the feature quantities of the non-tracking target image, in a case where a shape of an object in the non-tracking target image does not change, and
utilize a second masking intensity, which is larger than the first masking intensity, to mask the feature quantities of the non-tracking target image, in a case where the shape of the object changes.
2. The surveillance camera according to claim 1 ,
wherein the processor detects the non-tracking target image by using artificial intelligence.
3. The surveillance camera according to claim 1 ,
wherein an area of the non-tracking target image is designated by an information processing apparatus.
4. The surveillance camera according to claim 1 ,
wherein the processor detects an area of the non-tracking target image in the video frame in a case where an angle of view of the camera is changed.
5. The surveillance camera according to claim 4 ,
wherein the processor outputs an alarm signal in a case where a change amount of the area of the non-tracking target image exceeds a threshold.
6. The surveillance camera according to claim 2 ,
wherein the processor regularly detects the non-tracking target image.
7. An information processing apparatus comprising:
a receiver configured to receive a video frame from a surveillance camera; and
a processor configured to:
set a search range in the video frame,
determine feature quantities of an image in the search range,
perform similarity calculations between the feature quantities of the image in the search range and feature quantities of a tracking-target image in a previous video frame, and
extract a tracking-target image in the video frame by using the similarity calculations,
wherein, in a case where an non-tracking target image is included contained in the search range, the processor masks feature quantities of the non-tracking target image, and extracts the tracking-target image in the video frame with the feature quantities of the non-tracking target image masked,
wherein the processor is configured to:
utilize a first masking intensity to mask the feature quantities of the non-tracking target image, in a case where a shape of an object in the non-tracking target image does not change, and
utilize a second masking intensity, which is larger than the first masking intensity, to mask the feature quantities of the non-tracking target image, in a case where the shape of the object changes.
8. A method for an information processing apparatus, the method comprising:
receiving a video frame from a surveillance camera;
setting a search range in the video frame;
determining feature quantities of an image in the search range,
performing similarity calculations between the feature quantities of the image in the search range and feature quantities of a tracking-target image in a previous video frame, and
extracting a tracking-target image in the video frame by using the similarity calculations,
wherein, in a case where a non-tracking target image is included in the search range, feature quantities of the non-tracking target image are masked, and the tracking-target image in the video frame is extracted with the feature quantities of the non-tracking target image masked,
wherein a first masking intensity is utilized to mask the feature quantities of the non-tracking target image, in a case where a shape of an object in the non-tracking target image does not change,
wherein a second masking intensity, which is larger than the first masking intensity, is utilized to mask the feature quantities of the non-tracking target image, in a case where the shape of the object changes.
9. The surveillance camera according to claim 1 ,
wherein the processor masks feature quantities of the non-tracking target image by applying weights to the feature quantities of the non-tracking target image.