Back to results
US 11,600,025 B2Granted
Image processing method, image processing apparatus, image processing system, and learnt model manufacturing method
AssigneeCanon Kabushiki Kaisha
InventorNorihito Hiasa
Life of this patent
Estimating the term…
Abstract
An image processing method comprising steps of obtaining a first map representing a region outside a dynamic range of an input image based on a signal value in the input image and a threshold of the signal value, and inputting input data including the input image and the first map and executing a recognition task or a regression task.
Claims18
An image processing method comprising steps of:
- obtaining an input image that is at least part of a captured image obtained by an image sensor;
- obtaining a first map representing a region outside a dynamic range of the input image based on a signal value in the input image and a threshold of the signal value;
- and
- inputting input data including the input image and the first map to a neural network and executing a recognition task or a regression task,
- wherein the threshold of the signal value is set based on at least one of a luminance saturation value and a black level in the input image that is the at least part of the captured image obtained by the image sensor, and
- wherein the input data is obtained by concatenating the input image and the first map in a channel direction.
3 more elements
An image processing apparatus comprising:
- an obtaining unit configured to obtain (i) an input image that is at least part of a captured image obtained by an image sensor and (ii) a first map representing a region outside a dynamic range of the input image based on a signal value in the input image and a threshold of the signal value;
- and
- a processing unit configured to input data including the input image and the first map to a neural network, and to execute a recognition task or a regression task,
- wherein the threshold of the signal value is set based on at least one of a luminance saturation value and a black level in the input image that is the at least part of the captured image obtained by the image sensor,
- wherein the input data is obtained by concatenating the input image and the first map in a channel direction, and
- wherein at least one processor or circuit is configured to perform a function of at least one of the units.
3 more elements
An image processing system comprising:
- a first apparatus;
- and
- a second apparatus communicable with the first apparatus,
- wherein the first apparatus includes a transmitter configured to transmit a request to make the second apparatus execute processing on a captured image obtained by an image sensor,
- wherein the second apparatus includes:
- a receiver configured to receive the request transmitted by the transmitter;
- an obtainer configured to obtain (i) at least part of the captured image obtained by the image sensor and (ii) a first map representing a region outside a dynamic range of the captured image based on a signal value in the captured image and a threshold of the signal value;
- a processor configured to input data including the captured image and the first map to a neural network and to execute a recognition task or a regression task;
- and
- a transmitter configured to transmit a result of the executed task,
- wherein the threshold of the signal value is set based on at least one of a luminance saturation value and a black level in the at least part of the captured image obtained by the image sensor, and
- wherein the input data is obtained by concatenating the captured image and the first map in a channel direction.
9 more elements
An image processing method comprising steps of:
- obtaining a training image that is at least part of a captured image obtained by an image sensor, a first map representing a region outside a dynamic range of the training image based on a signal value in the training image and a threshold of the signal value, and ground truth data;
- and
- making a neural network learn for executing a recognition task or a regression task, using input data including the training image, the first map, and the ground truth data,
- wherein the threshold of the signal value is set based on at least one of a luminance saturation value and a black level in the training image that is the at least part of the captured image obtained by the image sensor, and
- wherein the input data is obtained by concatenating the training image and the first map in a channel direction.
2 more elements
A learnt model manufacturing method comprising steps of:
- obtaining a training image that is at least part of a captured image obtained by an image sensor, a first map representing a region outside a dynamic range of the training image based on a signal value in the training image and a threshold of the signal value, and ground truth data;
- and
- making a neural network learn for executing a recognition task or a regression task using input data including the training image, the first map, and the ground truth data,
- wherein the threshold of the signal value is set based on at least one of a luminance saturation value and a black level in the training image that is the at least part of the captured image obtained by the image sensor, and
- wherein the input data is obtained by concatenating the training image and the first map in a channel direction.
2 more elements
An image processing apparatus comprising:
- an obtaining unit configured to obtain a training image that is at least part of a captured image obtained by an image sensor, a first map representing a region outside a dynamic range of the training image based on a signal value in the training image and a threshold of the signal value, and ground truth data;
- and
- a learning unit configured to make a neural network learn for executing a recognition task or a regression task using input data including the training image, the first map, and the ground truth data,
- wherein the threshold of the signal value is set based on at least one of a luminance saturation value and a black level in the training image that is the at least part of the captured image obtained by the image sensor,
- wherein the input data is obtained by concatenating the training image and the first map in a channel direction, and
- wherein at least one processor or circuit is configured to perform a function of at least one of the units.
3 more elements
Dependencies are inferred from the claim text.