Image Encoding Quantization Control via Section Complexity Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image encoding technologies face challenges in optimally controlling the quantization level to maintain a constant bitrate across images, leading to potential loss of quality due to varying redundancy and localized detail, especially when encoding images in real-time with constrained buffer sizes.
Innovation Solution
A method is introduced to determine a suitable quantization level for each image section by analyzing pixel complexity and using a relationship to estimate the number of bits generated at different quantization levels, allowing for selection of an optimal level based on a target bitrate, thereby improving encoding quality and maintaining consistent bitrate.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If the QP is increased to reduce bit rate, then compression ratio is improved, but image quality deteriorates
Solution Approach 1:
The patent applies local quality by determining different quantization parameters for different image sections based on their individual complexity characteristics. High-detail sections use lower QP values to preserve quality, while low-detail sections use higher QP values to reduce bit rate, thereby optimizing the overall compression ratio without uniformly sacrificing image quality.
Solution Approach 2:
The patent implements dynamics by making the quantization parameter adaptive rather than static. The QP is dynamically adjusted for each image section based on real-time complexity analysis, allowing the encoding system to respond to varying local requirements and achieve better quality-bit rate trade-offs compared to fixed QP approaches.
2Productivity
If the QP is varied significantly across image sections to optimize bitrate allocation, then compression efficiency is improved, but encoding complexity increases
Solution Approach 1:
The patent applies segmentation by dividing the image into multiple sections and analyzing the complexity of each section independently. This allows different quantization parameters to be assigned to different segments based on their specific characteristics, improving compression efficiency while keeping the complexity manageable through localized processing.
Solution Approach 2:
The patent implements preliminary action by performing complexity analysis and determining appropriate quantization parameters for each image section before the actual encoding process. This pre-planning allows the encoder to efficiently allocate bits across different sections without adding significant complexity during the encoding phase itself.
3Speed
If the buffer size is constrained to reduce storage requirements and latency, then real-time encoding capability is improved, but bitrate control precision deteriorates
Solution Approach 1:
The patent applies segmentation by processing the image in sections rather than as a whole, allowing the encoder to work with smaller data buffers while maintaining effective bitrate control. Each section can be encoded and buffered independently, reducing the total buffer size requirement while preserving precision through localized complexity-based QP adjustment.
Data Source
AI summary
An image processor comprises an image pre-processing block and an encoder processing block for processing and encoding an image. The image pre-processing block receives image data and processes it to provide an image comprising image sections which each comprise pixels. For each of the image sections, the pixels are analyzed to estimate an indication of the complexity of the image section, and metadata is determined based on the estimated complexity indications of the image sections. The metadata is passed to the encoder processing block which uses it to determine a quantization level for use in encoding the image. The encoder processing block can then encode the image using the determined quantization level. Conveniently, the image pre-processing block 106 processes the image data to provide the image, and therefore has access to the image which it can analyze to determine the metadata without requiring a separate read operation of the image.