17754192. IMAGE PROCESSING METHOD AND IMAGE PROCESSING APPARATUS simplified abstract (BOE TECHNOLOGY GROUP CO., LTD.)
Contents
IMAGE PROCESSING METHOD AND IMAGE PROCESSING APPARATUS
Organization Name
BOE TECHNOLOGY GROUP CO., LTD.
Inventor(s)
IMAGE PROCESSING METHOD AND IMAGE PROCESSING APPARATUS - A simplified explanation of the abstract
This abstract first appeared for US patent application 17754192 titled 'IMAGE PROCESSING METHOD AND IMAGE PROCESSING APPARATUS
Simplified Explanation
The abstract describes an image processing method that involves using convolutional neural networks to convert a standard dynamic range (SDR) image into a high dynamic range (HDR) image.
- Obtaining a to-be-converted SDR image
- Using a first convolutional network to analyze features of the SDR image and obtain weights
- Obtaining a first 3D lookup table for the SDR image based on the weights and preset 3D lookup tables
- Adjusting the color information of the SDR image using the first 3D lookup table to obtain an HDR image
- Using a second convolutional neural network to refine and correct the HDR image to obtain an output image
Potential applications of this technology:
- Enhancing image quality in photography and videography
- Improving visual effects in movies and video games
- Enhancing medical imaging for better diagnostics
Problems solved by this technology:
- Enhancing the dynamic range of images for better contrast and detail
- Improving color accuracy and fidelity in image processing
Benefits of this technology:
- Producing high-quality HDR images from SDR sources
- Enhancing overall image quality and visual appeal
- Providing more accurate and realistic color representation in images
Original Abstract Submitted
The present disclosure provides an image processing method and an image processing apparatus. The image processing method includes: obtaining a to-be-converted SDR image; using a first convolutional network to perform feature analysis on the SDR image, to obtain N weights of the SDR image; where the N weights are respectively configured to characterize proportions of color information of the SDR image to color information characterized in preset N 3D lookup tables, the N 3D lookup tables are configured to characterize color information of different types; obtaining a first 3D lookup table for the SDR image according to the N weights and the N 3D lookup tables; using the first 3D lookup table to adjust the color information of the SDR image to obtain an HDR image; and using a second convolutional neural network to perform refinement correction on the HDR image to obtain an output image.
