18337899. METHOD AND APPARATUS WITH IMAGE PROCESSING BASED ON A NEURAL NETWORK simplified abstract (SAMSUNG ELECTRONICS CO., LTD.)
METHOD AND APPARATUS WITH IMAGE PROCESSING BASED ON A NEURAL NETWORK
Organization Name
Inventor(s)
Jaehyoung Yoo of Suwon-si (KR)
Hyong Euk Lee of Suwon-si (KR)
METHOD AND APPARATUS WITH IMAGE PROCESSING BASED ON A NEURAL NETWORK - A simplified explanation of the abstract
This abstract first appeared for US patent application 18337899 titled 'METHOD AND APPARATUS WITH IMAGE PROCESSING BASED ON A NEURAL NETWORK
Simplified Explanation
The abstract describes a neural network-based image processing method and apparatus that can operate in either an independent processing mode or a cooperative processing mode. In the independent processing mode, only the main neural network is used to generate intermediate image data. In the cooperative processing mode, both the main neural network and an auxiliary neural network are used to generate intermediate image data based on an auxiliary parameter. The output image data is then generated using an image signal processing block based on the intermediate image data and the current processing mode.
- Neural network-based image processing method and apparatus
- Can operate in independent or cooperative processing modes
- Main neural network used in independent mode
- Main and auxiliary neural networks used in cooperative mode
- Output image data generated using image signal processing block
- Based on intermediate image data and processing mode
Potential Applications
- Image processing in various industries such as healthcare, automotive, and entertainment
- Enhancing image quality in cameras and smartphones
- Improving image recognition in security systems
Problems Solved
- Efficient image processing using neural networks
- Flexibility in choosing processing modes based on requirements
- Enhanced image processing capabilities with auxiliary neural network
Benefits
- Improved image processing accuracy and efficiency
- Enhanced image quality and recognition capabilities
- Flexibility in adapting to different processing requirements
Original Abstract Submitted
A neural network-based image processing method and apparatus are provided. A method includes receiving input image data comprising original image data, when a current image processing mode is an independent processing mode, generating first intermediate image data by executing a main neural network based on the original image data and by not executing an auxiliary neural network based on the original image data, when the current image processing mode is in a cooperative processing mode, generating second intermediate image data by determining an auxiliary parameter by executing the auxiliary neural network based on the original image data and by executing the main neural network based on the original image data and based on the auxiliary parameter, and generating output image data by operating an image signal processing (ISP) block based the first intermediate image data or the second intermediate image data according to the current image processing mode.
