Adaptive Image Encoding for Fixed-Length Edge-Preserving Compression

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional image encoding methods face challenges in maintaining fixed encoding length while minimizing image degradation, especially at image edges, due to varying differential values, leading to increased bit requirements and reduced compression rates.

Innovation Solution

The image encoding device automatically selects between PCM and DPCM processing based on differential values, adjusting quantization steps according to threshold values, to encode pixels with large or small differential values effectively, ensuring fixed-length encoding and minimizing image degradation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If DPCM encoding is applied to image data with large differential values (e.g., at edges), then fixed-length encoding is achieved, but image degradation increases

Engineering Contradiction:
Improveencoding length consistencyVSAvoidimage quality
Core Design Contradiction:
Device complexityVSManufacturing precision

Solution Approach 1:

The patent applies different encoding methods to different regions of the image based on local characteristics. Specifically, it uses DPCM encoding for regions with small differential values and PCM encoding for regions with large differential values (such as edges), thereby adapting the encoding quality to the local image content and preventing degradation at critical areas

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The patent dynamically adjusts the encoding method for each pixel based on the differential value between adjacent pixels. By calculating the differential value and comparing it against a threshold, the system selectively applies DPCM or PCM encoding on a per-pixel basis, making the encoding process adaptive rather than static

Inventive Principle:
Principle #15Dynamics

2Manufacturing precision

If encoding bits are increased to prevent image degradation at edges, then image quality improves, but compression rate decreases

Engineering Contradiction:
Improveimage qualityVSAvoidcompression rate
Core Design Contradiction:
Manufacturing precisionVSProductivity

Solution Approach 1:

The patent applies different encoding methods to different regions of the image based on local characteristics. Specifically, it uses DPCM encoding for regions with small differential values and PCM encoding for regions with large differential values (such as edges), thereby adapting the encoding quality to the local image content and preventing degradation at critical areas

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The patent changes the encoding parameter (encoding method) based on the differential value parameter. By setting a threshold for the differential value and switching between DPCM and PCM methods accordingly, the system optimizes the balance between image quality and compression rate by using more bits only where necessary

Inventive Principle:
Principle #35Parameter changes

3Manufacturing precision

If PCM encoding is used for all pixels, then image quality is maintained, but data amount increases

Engineering Contradiction:
Improveimage qualityVSAvoiddata amount
Core Design Contradiction:
Manufacturing precisionVSQuantity of substance

Solution Approach 1:

The patent applies different encoding methods to different regions of the image based on local characteristics. Specifically, it uses DPCM encoding for regions with small differential values and PCM encoding for regions with large differential values (such as edges), thereby adapting the encoding quality to the local image content and preventing degradation at critical areas

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The patent changes the encoding parameter (encoding method) based on the differential value parameter. By setting a threshold for the differential value and switching between DPCM and PCM methods accordingly, the system optimizes the balance between image quality and compression rate by using more bits only where necessary

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP2202983B1Image encoding device, image encoding method, and image encoding system
Publication Date: 2019.06.05 SHARP KK
  • EP2202983B1 patent drawingFigure 1
  • EP2202983B1 patent drawingFigure 2
  • EP2202983B1 patent drawingFigure 3A~3D

AI summary

An image encoding device comprises: a predicted value generation means for predicting target pixel data which is the pixel data of an encoding target pixel to generate predicted value data; a subtractor for calculating the difference value between the target pixel data and the predicted value data to generate difference value data; an encoding mode decision means for determining an encoding mode based on the difference value data and temporarily holding it in a buffer; an encoding target decision means for determining which data of the target pixel or the difference value is used as encoding target data to be encoded based on the encoding mode; a quantization means for applying a quantization processing to the encoding target data based on the encoding mode to generate quantized data; and a fixed-length code generation means for adding the encoding mode to the quantized data to generate a fixed-length code.