Imaging apparatus, car, and variation detection method
AssigneeKYOCERA Corporation
InventorYasushi Koshiba
Life of this patent
Estimating the term…
Abstract
Information is acquired to contribute to determining the existence of at least one of an attachment anomaly in attachment of an imaging unit to a car and an attachment anomaly in attachment of a predetermined fixed object to the car. An imaging apparatus includes an imaging unit and an image processor. The imaging unit captures an image of the surroundings of the car. The image processor detects continuous variation of a predetermined subject appearing in at least one predetermined region that is a portion of the entire region of the image captured by the imaging unit. The image processor makes a determination of whether an attachment anomaly exists based on the continuous variation.
Claims5
An imaging apparatus comprising:
- an imaging unit configured to capture an image of surroundings of a car;
- and
- an image processor configured to detect information related to location of a predetermined subject appearing in at least one predetermined region that is a portion of an entire region of each of the images of different frames captured by the imaging unit, and to make a determination of whether an attachment anomaly exists in at least one of attachment of the imaging unit to the car and attachment of the predetermined subject to the car based on Fourier transform of change over time of the information related to location, wherein
- the at least one predetermined region comprises a plurality of predetermined regions, and
- the image processor makes a determination of whether the attachment anomaly has occurred in attachment of the imaging unit to the car or in attachment of the predetermined subject to the car based on continuous variation of the predetermined subject appearing in each of the plurality of predetermined regions.
A car comprising:
- an imaging apparatus comprising an imaging unit configured to capture an image of surroundings of a car;
- and
- an image processor configured to detect information related to location of a predetermined subject appearing in at least one predetermined region that is a portion of an entire region of each of the images of different frames captured by the imaging unit, and to make a determination of whether an attachment anomaly exists in at least one of attachment of the imaging unit to the car and attachment of the predetermined subject to the car based on Fourier transform of change over time of the information related to location, wherein
- the at least one predetermined region comprises a plurality of predetermined regions, and
- the image processor makes a determination of whether the attachment anomaly has occurred in attachment of the imaging unit to the car or in attachment of the predetermined subject to the car based on continuous variation of the predetermined subject appearing in each of the plurality of predetermined regions.
A variation detection method comprising:
- capturing an image of surroundings of a car;
- detecting, by an image processor, information related to location of a predetermined subject appearing in at least one predetermined region that is a portion of an entire region of each of the images of different frames that are captured;
- and
- making, by the image processor, a determination of whether an attachment anomaly exists in at least one of attachment of an imaging unit to the car and attachment of the predetermined subject to the car based on Fourier transform of change over time of the information related to location, the imaging unit capturing an image of surroundings of the car, wherein
- the at least one predetermined region comprises a plurality of predetermined regions, and
- the image processor makes a determination of whether the attachment anomaly has occurred in attachment of the imaging unit to the car or in attachment of the predetermined subject to the car based on continuous variation of the predetermined subject appearing in each of the plurality of predetermined regions.
Dependencies are inferred from the claim text.