18150610. APPARATUS AND METHOD FOR IMAGE PROCESSING simplified abstract (CANON KABUSHIKI KAISHA)
Contents
APPARATUS AND METHOD FOR IMAGE PROCESSING
Organization Name
Inventor(s)
APPARATUS AND METHOD FOR IMAGE PROCESSING - A simplified explanation of the abstract
This abstract first appeared for US patent application 18150610 titled 'APPARATUS AND METHOD FOR IMAGE PROCESSING
Simplified Explanation
The abstract describes an apparatus that includes processors for acquiring training and correct answer images, generating an output image using a neural network, and calculating the error between the correct answer image and the output image. The apparatus also includes an update unit to adjust the parameters of the neural network based on the calculated error.
- The apparatus includes processors for image acquisition, generation, error acquisition, and updating.
- The image acquisition unit acquires a training image and a correct answer image.
- The generation unit inputs the training image to a neural network and generates an output image.
- The error acquisition unit adjusts the color signal value of the correct answer image and the output image, and calculates the error between them.
- The update unit updates the parameters of the neural network based on the calculated error.
Potential Applications
- Image recognition and classification systems.
- Computer vision applications.
- Artificial intelligence and machine learning algorithms.
Problems Solved
- Improves the accuracy of image recognition and classification systems.
- Enhances the performance of computer vision applications.
- Enables more efficient training and updating of neural networks.
Benefits
- Higher accuracy in image recognition and classification tasks.
- Improved performance and efficiency of computer vision systems.
- Enhanced capabilities of artificial intelligence and machine learning algorithms.
Original Abstract Submitted
An apparatus includes one or more processors that function as an image acquisition unit configured to acquire a training image and a correct answer image, a generation unit configured to input the training image to a neural network to generate an output image, an error acquisition unit configured to subject each of the correct answer image and the output image to processing for adjusting a color signal value, and acquire an error between the correct answer image and the output image that have been subjected to the processing, and an update unit configured to update parameters of the neural network based on the acquired error.