Image Compression Quality Prediction With Block Confidence Maps

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing electronic devices face challenges in accurately determining and enhancing image quality due to compression artifacts, particularly when images are compressed and displayed on larger screens, as the original compression quality information is lost, leading to artifacts that were not visible on smaller screens becoming apparent.

Innovation Solution

An electronic device uses a deep neural network to analyze partial regions of an image, estimating confidence in units of blocks to predict compression quality, remove outliers, and apply a noise removal model to enhance image quality.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If image compression is applied to reduce file size and transmission time, then storage efficiency and transmission speed are improved, but compression artifacts are generated and image quality deteriorates

Engineering Contradiction:

Improvetransmission speedVSAvoidcompression artifacts

Core Design Contradiction:

ProductivityVSObject-affected harmful factors

Solution Approach 1:

The image is divided into multiple blocks, and confidence is estimated for each block individually. This segmentation allows the system to identify and handle different regions with varying compression qualities, improving the accuracy of artifact detection and quality assessment without requiring full-image processing.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

A deep neural network is introduced as an intermediary to predict compression quality and generate confidence maps. This intermediary component bridges the gap between compressed image data and quality assessment, enabling accurate artifact detection without requiring access to the original uncompressed image.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Object-affected harmful factors

If compression quality is increased to reduce artifacts, then image quality is improved, but file size increases and storage efficiency decreases

Engineering Contradiction:

Improvecompression artifactsVSAvoidfile size

Core Design Contradiction:

Object-affected harmful factorsVSQuantity of substance

Solution Approach 1:

The system performs self-assessment by using the compressed image itself to generate confidence maps and predict quality metrics. By analyzing the compressed image data directly without requiring the original image or additional resources, the system can identify artifact-prone regions and optimize quality enhancement strategies accordingly.

Inventive Principle:

Principle #25Self-service

3Measurement precision

If deep neural network analyzes entire image to predict compression quality, then prediction accuracy is improved, but processing time increases

Engineering Contradiction:

Improvequality prediction accuracyVSAvoidprocessing time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system performs confidence estimation on selected blocks rather than the entire image. By applying partial action to representative blocks and using deep learning to generalize results, the system achieves sufficient prediction accuracy with reduced computational overhead and faster processing times.

Inventive Principle:

Principle #16Partial or excessive action

4Illumination intensity

If image is enlarged for large screen display, then display quality is improved, but compression artifacts become more visible

Engineering Contradiction:

Improvedisplay brightnessVSAvoidvisible artifacts

Core Design Contradiction:

Illumination intensityVSObject-affected harmful factors

Solution Approach 1:

The system performs preliminary quality assessment and confidence map generation before image enlargement. By identifying artifact-prone regions in advance using block-based confidence estimation, the system can apply targeted quality enhancement algorithms to specific blocks before scaling, reducing the visibility of artifacts when the image is enlarged for large screen display.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS12462356B2Apparatus and method for predicting compression quality of image in electronic device

Publication Date: 2025.11.04 SAMSUNG ELECTRONICS CO LTD

AI summary

A method and/or device for predicting a compression quality of an image during image correction (e.g., image quality enhancement) in an electronic device, and/or processing the image, based on at least the prediction, may be provided. The electronic device may include a display module, a memory, and a processor, wherein the processor may operate to display an image through the display module, extract designated multiple blocks from the image in a designated scheme, estimate confidence for each of the multiple blocks, identify, based on the estimation of the confidence, a first block corresponding to an outlier to be excluded in quality prediction, and a second block for which quality prediction is possible, among the multiple blocks, exclude the first block among the multiple blocks from a subject of quality prediction, and classify a compression quality of the image by using at least the second block remaining after excluding the first block from among the multiple blocks.