Image Processing Apparatus Color Banding Suppression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional image processing methods, such as dithering and error diffusion, fail to effectively suppress color banding in high-definition displays due to noise reduction limitations and the inability to fully utilize higher bit-depth capabilities of newer display interfaces like HDMI 1.3, leading to stair-like boundaries in gradation images.
Innovation Solution
An image processing apparatus that generates a correction function from M-bit gradation image data to produce N-bit gradation display image data, utilizing a hierarchical structure and software decoding to convert 8-bit JPEG data into 12-bit DeepColor output, thereby reducing quantization step width and suppressing color banding.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If noise is added to image data to reduce color banding, then color banding is suppressed, but image compression ratio is reduced
Solution Approach 1:
The correction function is generated in advance from the original image data before display. This pre-computed function captures the necessary correction information without requiring real-time processing of noise addition, thereby maintaining compression efficiency while enabling color banding suppression during display
Solution Approach 2:
A correction function serves as an intermediary between the compressed image data and the final display output. This function mediates the transformation by providing correction values that suppress color banding without requiring direct manipulation of the original image data, thus preserving compression ratios
2Object-affected harmful factors
If noise is added to image data to reduce color banding, then color banding is suppressed, but subsequent image processing alters the added noise, preventing it from performing the intended function
Solution Approach 1:
The correction function is generated beforehand from the original image data, capturing the optimal noise pattern for color banding suppression. This pre-computed function remains stable and reliable even when the original image data undergoes subsequent processing, as the correction values are already optimized
Solution Approach 2:
The correction function is generated based on analysis of the original image data's characteristics. This feedback mechanism ensures the correction values are specifically tailored to the image content, making the noise addition more effective and less susceptible to degradation from subsequent processing
3Loss of energy
If 8-bit quantization is used in image formats like JPEG or MPEG, then image compression is achieved, but high-definition display interfaces cannot be fully utilized
Solution Approach 1:
The invention extends the display from 8-bit to 12-bit gradation by introducing a correction function that operates in the additional bit depth dimension. This allows the system to maintain 8-bit compressed storage while utilizing the full capability of 12-bit display interfaces, effectively adding a new dimension to the data representation
Solution Approach 2:
The correction function changes the gradation parameter from 8-bit to 12-bit by providing correction values that expand the effective bit depth. This parameter transformation enables high-definition display interfaces to be fully utilized while maintaining compatibility with standard compressed image formats
4Device complexity
If 8-bit gradation image data is displayed, then image data storage and processing is simplified, but stair-like boundaries appear due to color banding
Solution Approach 1:
A correction function acts as an intermediary layer between the simple 8-bit image data and the final display output. This function provides the necessary complexity to suppress color banding while keeping the original image data storage and processing simple, effectively isolating the complexity from the main data flow
Solution Approach 2:
The correction function is generated in advance from the 8-bit image data, capturing all necessary correction information beforehand. This preliminary action allows the simple 8-bit data to be enhanced without requiring complex real-time processing during display, maintaining simplicity while eliminating color banding
Data Source
AI summary
A main memory stores M-bit gradation image data (M is a natural number). A correction function generating unit generates a correction function in N-bit gradation (N is a natural number, N>M) from image data. A display image generating unit generates display image data using a correction function generated by the correction function generating unit. The display image generating unit uses a correction function so as to generate display image data resulting from correcting image data obtained by decoding compressed image data in M-bit gradation with the restriction of M-bit quantization.