Back to results
US 2022/0101493 A1Published application
Image processing method, storage medium, image processing apparatus, trained model producing method, learning method, learning apparatus, and image processing system
AssigneeCanon Kabushiki Kaisha
InventorYoshinori Kimura
Life of this patent
Estimating the term…
Abstract
An image processing method includes generating a second image by applying noise to a first image, as a first process, and generating a third image by upsampling the second image through a neural network, as a second process. The second image is generated by applying noise to the first image expressed in a second bit depth higher than a first bit depth of the first image and then restoring the bit depth to the first bit depth.
Claims13
An image processing method comprising:
- generating a second image by applying noise to a first image, as a first process;
- and
- generating a third image by upsampling the second image through a neural network, as a second process,
- wherein the second image is generated by applying noise to the first image expressed in a second bit depth higher than a first bit depth of the first image and then restoring the bit depth to the first bit depth.
1 more element
An image processing apparatus comprising:
- a generation unit configured to generate a second image by applying noise to a first image;
- and
- a processing unit configured to generate a third image by upsampling the second image through a neural network,
- wherein the second image is generated by applying noise to the first image expressed in a second bit depth higher than a first bit depth of the first image and then restoring the bit depth to the first bit depth.
1 more element
A method of producing a trained model, the method comprising:
- generating a second training image by applying noise to a first training image, as a first process;
- and
- training a neural network for converting the second training image into a high-resolution training image by using a third training image having a higher resolution than the first training image and including a subject identical to a subject on the first training image, as a second process,
- wherein the second training image is generated by applying noise to the first training image expressed in a second bit depth higher than a first bit depth of the first training image and then restoring the bit depth to the first bit depth.
1 more element
A learning apparatus comprising:
- a generation unit configured to generate a second training image by applying noise to a first training image;
- and
- a processing unit configured to train a neural network for converting the second training image into a high-resolution training image by using a third training image having a higher resolution than the first training image and including a subject identical to a subject on the first training image,
- wherein the second training image is generated by applying noise to the first training image expressed in a second bit depth higher than a first bit depth of the first training image and then restoring the bit depth to the first bit depth.
1 more element
An image processing system comprising:
- a first apparatus;
- and
- a second apparatus to communicate with the first apparatus,
- wherein the first apparatus comprises a transmission unit configured to transmit a request for causing the second apparatus to perform processing on a first image,
- wherein the second apparatus comprises:
- a generation unit configured to generate a second image by applying noise to the first image;
- and
- a processing unit configured to generate a third image by upsampling the second image through a neural network, and
- wherein the second image is generated by applying noise to the first image expressed in a second bit depth higher than a first bit depth of the first image and then restoring the bit depth to the first bit depth.
6 more elements
Dependencies are inferred from the claim text.