Back to results
US 2017/0374272 A1Published application
Image processing apparatus, imaging apparatus, and image processing method
AssigneeCanon Kabushiki Kaisha
InventorRyosuke Tsuji
Life of this patent
Estimating the term…
Abstract
One aspect of the present invention provides an image processing apparatus capable of tracking an image region to be tracked with high accuracy. A distance distribution generation unit generates distance information indicating a distance of each of regions in first image data. An object likelihood distribution generation unit generates a likelihood distribution of an object in the first image data based on information indicating a position specified as the object in the first image data, the distance information, and reliability of the distance information. A feature amount extraction unit calculates a feature amount to be used to detect the object from second image data based on the likelihood distribution.
Claims15
An image processing apparatus comprising:
- at least one processor;
- at least one nonvolatile memory device;
- an acquisition unit configured to acquire distance information indicating a distance of each of regions in first image data;
- a generation unit configured to generate a likelihood distribution of an object in the first image data based on information indicating a position specified as the object in the first image data, the distance information, and reliability of the distance information;
- and
- a calculation unit configured to calculate a feature amount to be used to detect the object from second image data based on the likelihood distribution,
- wherein at least one of the acquisition unit, the generation unit, and the calculation unit is realized by execution of a program stored in the at least one nonvolatile memory device by the at least one processor.
4 more elements
An image processing apparatus comprising:
- at least one processor;
- at least one nonvolatile memory device;
- an acquisition unit configured to calculate a defocus amount of each of regions in first image data from a plurality of pieces of image data corresponding to the first image data and having a parallax, and acquire distance information from the defocus amount of each of the regions;
- a generation unit configured to generate a likelihood distribution of an object in the first image data based on information indicating a position specified as the object in the first image data and the distance information;
- and
- a calculation unit configured to calculate a feature amount to be used to detect the object from second image data based on the likelihood distribution,
- wherein at least one of the acquisition unit, the generation unit, and the calculation unit is realized by execution of a program stored in the at least one nonvolatile memory device by the at least one processor.
4 more elements
An imaging apparatus comprising:
- at least one processor;
- at least one nonvolatile memory device;
- an image sensor configured to generate image data;
- an acquisition unit configured to acquire distance information indicating a distance of each of regions in first image data;
- a generation unit configured to generate a likelihood distribution of an object in the first image data based on information indicating a position specified as the object in the first image data, the distance information, and reliability of the distance information;
- a calculation unit configured to calculate a feature amount to be used to detect the object from second image data based on the likelihood distribution;
- a detection unit configured to detect the object from the second image data with use of the feature amount calculated by the calculation unit;
- and
- a control unit configured to control an imaging condition when the image senor captures an image according to information about an image region corresponding to the object detected by the detection unit,
- wherein at least one of the acquisition unit, the generation unit, the calculation unit, the detection unit, and the control unit is realized by execution of a program stored in the at least one nonvolatile memory device by the at least one processor.
7 more elements
An imaging apparatus comprising:
- at least one processor;
- at least one nonvolatile memory device;
- an image sensor configured to generate image data;
- an acquisition unit configured to calculate a defocus amount of each of regions in first image data from a plurality of pieces of image data corresponding to the first image data and having a parallax, and acquire distance information from the defocus amount of each of the regions;
- a generation unit configured to generate a likelihood distribution of an object in the first image data based on information indicating a position specified as the object in the first image data and the distance information;
- a calculation unit configured to calculate a feature amount to be used to detect the object from second image data based on the likelihood distribution;
- a detection unit configured to detect the object from the second image data with use of the feature amount calculated by the calculation unit;
- and
- a control unit configured to control an imaging condition when the image senor captures an image according to information about an image region corresponding to the object detected by the detection unit,
- wherein at least one of the acquisition unit, the generation unit, the calculation unit, the detection unit, and the control unit is realized by execution of a program stored in the at least one nonvolatile memory device by the at least one processor.
7 more elements
An image processing method comprising:
- acquiring distance information indicating a distance of each of regions in first image data;
- generating a likelihood distribution of an object in the first image data based on information indicating a position specified as the object in the first image data, the distance information, and reliability of the distance information;
- and
- calculating a feature amount to be used to detect the object from second image data based on the likelihood distribution.
1 more element
An image processing method comprising:
- calculating a defocus amount of each of regions in first image data from a plurality of pieces of image data corresponding to the first image data and having a parallax, and acquiring distance information from the defocus amount of each of the regions;
- generating a likelihood distribution of an object in the first image data based on information indicating a position specified as the object in the first image data and the distance information;
- and
- calculating a feature amount to be used to detect the object from second image data based on the likelihood distribution.
1 more element
A computer-readable nonvolatile storage medium storing a program for controlling an image processing apparatus, the program including codes for causing execution of:
- acquiring distance information indicating a distance of each of regions in first image data;
- generating a likelihood distribution of an object in the first image data based on information indicating a position specified as the object in the first image data, the distance information, and reliability of the distance information;
- and
- calculating a feature amount to be used to detect the object from second image data based on the likelihood distribution.
1 more element
A computer-readable nonvolatile storage medium storing a program for controlling an image processing apparatus, the program including codes for causing execution of:
- calculating a defocus amount of each of regions in first image data from a plurality of pieces of image data corresponding to the first image data and having a parallax, and acquiring distance information from the defocus amount of each of the regions;
- generating a likelihood distribution of an object in the first image data based on information indicating a position specified as the object in the first image data and the distance information;
- and
- calculating a feature amount to be used to detect the object from second image data based on the likelihood distribution.
1 more element
Dependencies are inferred from the claim text.