Adaptive Image Tiling for Cross-Image Compression Redundancy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The high storage requirements for product photos in e-commerce, despite image compression, due to redundancy and varying image qualities, lead to significant storage costs and inefficiencies.
Innovation Solution
Adaptive tiling and clustering techniques are applied to split images into tiles based on RGB color intensity, remove duplicates, and compress tiles using the best-suited methods, reducing storage needs by leveraging commonality across images.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If image compression is applied to product photos, then storage space is reduced, but storage requirements remain excessively high due to redundancy across multiple images of the same product
Solution Approach 1:
The patent divides images into multiple tiles or blocks, allowing independent processing and comparison of image segments. This segmentation enables the system to identify and remove duplicate tiles across multiple product images, significantly reducing storage requirements while maintaining image quality.
Solution Approach 2:
The patent creates a master template from the first product image and generates difference images by comparing subsequent images against this template. This copying approach allows the system to store only the unique differences rather than redundant complete images, dramatically improving storage efficiency.
2Reliability
If multiple high-quality product photos are stored to provide comprehensive visual information, then image quality and variety are improved, but storage costs increase exponentially
Solution Approach 1:
The patent extracts and removes duplicate tile content from multiple product images by comparing RGB values and identifying identical regions. This extraction process eliminates redundant storage of identical visual information while preserving all unique product details and variations.
Solution Approach 2:
The patent transforms images into a different representation by converting them into tiles with specific RGB value parameters. This parameter-based approach allows for systematic comparison and identification of duplicate content across images, enabling efficient compression while maintaining visual fidelity.
3Productivity
If standard image compression techniques are applied uniformly to all product images, then processing simplicity is maintained, but compression efficiency is suboptimal due to varying image characteristics
Solution Approach 1:
The patent segments images into tiles and applies compression operations at the tile level rather than processing entire images uniformly. This segmentation allows different compression strategies to be applied to different image regions based on their characteristics, improving overall compression efficiency.
Solution Approach 2:
The patent applies compression operations selectively to only those tiles that contain unique content, rather than compressing all tiles in all images. This partial action approach maximizes compression efficiency by focusing computational resources on reducing redundant data while maintaining necessary image quality.
Data Source
AI summary
Techniques are provided herein for more efficiently storing images that have a common subject, such as product images that share the same product in the image. Each image undergoes an adaptive tiling procedure to split the image into a plurality of tiles, with each tile identifying a region of the image having pixels with the same content. The tiles across multiple images can then be clustered together and those tiles having identical content are removed. Once all duplicate tiles have been removed from the set of all tiles across the images, the tiles are once again clustered based on their encoding scheme and certain encoding parameters. Tiles within each cluster are compressed using the best compression technique for the tiles in each corresponding cluster. By removing duplicative tile content between numerous images of the same subject, the total amount of data that needs to be stored is reduced.