Adaptive Encoder Block Segmentation for Mixed Content Compression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data encoding methods are not well-suited for a broad range of content types, such as still images, video, audio, and graphics, as they either produce undesirable artifacts or require significant processing power, and lack an efficient method for encoding and decoding data that is compact and adaptable across various content types.
Innovation Solution
An encoder that analyzes data portions and directs them to appropriate encoding units, employing different algorithms to compute aggregate values and spatial masks, which are included in the encoded data for efficient encoding, and a corresponding decoder that extracts these values to regenerate the original data, using techniques like DPCM, RLE, and VLC for further compression.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If JPEG or JPEG2000 encoding methods are used, then natural image content is well compressed, but images with few color values and high spatial frequency components produce undesirable artifacts
Solution Approach 1:
The encoder divides the image into multiple blocks and analyzes each block's characteristics (color variation, spatial frequency) to select the most appropriate encoding method for that specific block, rather than applying a single encoding method to the entire image
Solution Approach 2:
The encoding method dynamically adapts to different content types by analyzing block characteristics and selecting from multiple encoding algorithms (DCT-based, wavelet-based, palette-based, or hybrid methods) based on the specific properties of each block
2Productivity
If GIF encoding with palette-based compression is used, then images with few color values are compressed efficiently, but images with subtle spatially-gradual color variations produce noticeable artifacts
Solution Approach 1:
The image is divided into blocks that are independently analyzed and encoded. Blocks with gradual color variations can use DCT or wavelet methods to preserve smooth gradients, while blocks with limited colors can use palette-based methods for compression
Solution Approach 2:
Different encoding methods with different quality characteristics are applied to different regions of the image based on local content properties, allowing each region to be encoded with the method best suited to its characteristics
3Adaptability or versatility
If OCR-based text encoding is used, then text content can be encoded, but the method requires considerable processing power and is sensitive to text positioning, tilting, font, and object location
Solution Approach 1:
The image is divided into blocks that are analyzed for text content. Instead of applying OCR to the entire image, text detection and encoding is performed on individual blocks, reducing computational complexity and improving robustness to positioning variations
Solution Approach 2:
The patent replaces complex OCR processing with simpler block-based analysis and encoding methods that are less computationally intensive and more robust to variations in text positioning, tilting, and formatting
Data Source
AI summary
An encoder generating encoded data. The encoder comprising an analysis unit for analysing portions of data to be encoded, and for directing the portions to one or more encoding units, the encoding units are operable to encode the data portions to generate encoded data. The one or more encoding units are operable to employ mutually different encoding algorithms when encoding the one or more portions. At least one encoding unit of the one or more encoding units is operable to compute data values present in each portion received thereat, to sub-divide the data values into at least two sets, to compute at least one aggregate value for a given set derived from the data values present in the given set. A corresponding decoder for decoding data generated by the encoder executes an inverse of encoding steps employed in the encoder.