Digital Raster Pattern Reconstruction After Lossy Compression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional image processing systems struggle with accurately detecting and reconstructing patterns in digital raster images due to the lossy nature of formats like JPEG, leading to inefficiencies and limitations in flexibility and accuracy during vectorization, particularly in extracting high-level pattern data.
Innovation Solution
A pattern reconstruction system that utilizes neighborhood descriptors to identify structurally similar pixels, applies non-maximum suppression to reduce these pixels to collinear sets, and refines pattern cells using a sliding window process to center on human-perceived objects, thereby enhancing detection and reconstruction accuracy and efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional image processing systems process digital raster images, then they can perform basic image editing tasks, but they struggle with accurately detecting and reconstructing patterns due to lossy compression
Solution Approach 1:
The system segments the image into neighborhoods around each pixel and extracts descriptors from these segments. By comparing neighborhood descriptors rather than relying on individual pixel values, the system can detect patterns even when compression has degraded specific pixel data, thus improving pattern detection accuracy and reconstruction reliability in lossy formats.
Solution Approach 2:
The patent introduces neighborhood descriptors as an intermediary representation between the compressed image data and the pattern detection process. These descriptors capture structural information that is more robust to compression artifacts, serving as a mediator that enables accurate pattern detection without requiring perfect pixel data.
2Measurement precision
If the system detects all structurally similar pixels, then it captures comprehensive pattern information, but it increases processing complexity and time
Solution Approach 1:
The system extracts only the most significant structurally similar pixels by applying non-maximum suppression. This operation removes redundant pixels that would not contribute to the final pattern detection, reducing processing complexity while maintaining detection accuracy. The extraction focuses computational resources on the most informative pixels.
Solution Approach 2:
The patent applies partial action by using non-maximum suppression to select only a subset of structurally similar pixels rather than processing all possible matches. This partial processing approach achieves sufficient accuracy for pattern detection while significantly reducing computational complexity and processing time.
3Adaptability or versatility
If the system processes images with lossy compression, then it can handle common digital image formats, but it reduces the accuracy of pixel value detection
Solution Approach 1:
The system applies local quality by analyzing neighborhood descriptors that capture the structural context around each pixel. This local structural analysis is more robust to compression artifacts than individual pixel values, allowing the system to maintain pattern detection accuracy while supporting lossy formats like JPEG. The method adapts to the degraded quality of compressed images by focusing on structural relationships rather than precise pixel values.
Data Source
AI summary
Methods, systems, and non-transitory computer readable storage media are disclosed for automatically detecting and reconstructing patterns in digital images. The disclosed system determines structurally similar pixels of a digital image by comparing neighborhood descriptors that include the structural context for neighborhoods of the pixels. In response to identify structurally similar pixels of a digital image, the disclosed system utilizes non-maximum suppression to reduce the set of structurally similar pixels to collinear pixels within the digital image. Additionally, the disclosed system determines whether a group of structurally similar pixels define the boundaries of a pattern cell that forms a rectangular grid pattern within the digital image. The disclosed system also modifies a boundary of a detected pattern cell to include a human-perceived pattern object via a sliding window corresponding to the pattern cell.