Image Similarity Detection via Discrete Transform and Color Histogram Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Online marketplaces face challenges in detecting and reducing duplicate or near-duplicate images of merchandise listings, which can clutter search results and deter buyers, as existing methods struggle to identify images that have been digitally manipulated.
Innovation Solution
A system and method that determine image similarities by analyzing discrete transforms, intensity vectors, and color histograms to identify duplicate or near-duplicate images, even when they have been digitally manipulated, and adjust their presentation in search results accordingly.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If multiple listings are created for the same item using the same image, then the seller's opportunity to sell the item increases, but the buyer's experience deteriorates due to redundant search results
Solution Approach 1:
The system performs preliminary detection of duplicate images before listings are published or during indexing. By pre-identifying duplicate images through image recognition and hashing techniques, the system can prevent redundant listings from appearing in search results, thus maintaining buyer experience while allowing sellers to create multiple listings.
Solution Approach 2:
The patent introduces an intermediary image recognition system that acts as a mediator between seller listings and buyer search results. This intermediary layer analyzes images, identifies duplicates, and selectively filters or consolidates them, allowing the system to maintain both seller productivity and buyer experience by presenting only unique items to buyers.
2Ease of operation
If image similarity detection is performed to identify duplicate listings, then redundant listings are reduced, but the system complexity increases due to digital manipulation challenges
Solution Approach 1:
The image analysis system is segmented into multiple independent modules: image preprocessing module, feature extraction module (analyzing discrete transforms, intensity vectors, color histograms), duplicate detection module, and decision-making module. This segmentation allows each component to specialize in one aspect of image analysis, making the overall system more manageable and easier to optimize despite the complexity of detecting digitally manipulated images.
Solution Approach 2:
The system analyzes multiple parameters of images including discrete transforms, intensity vectors, and color histograms. By examining changes across these different parameter spaces, the system can detect duplicates even when images have been digitally manipulated. This multi-parameter approach breaks down the complex problem into several simpler comparisons.
3Measurement precision
If advanced image analysis techniques are used to detect digitally manipulated duplicate images, then detection accuracy improves, but computational resources increase
Solution Approach 1:
The system applies partial action by analyzing only the most discriminative features of images (discrete transforms, intensity vectors, color histograms) rather than processing every pixel in full resolution. This selective analysis maintains high detection accuracy for digitally manipulated images while significantly reducing computational resource consumption compared to exhaustive image processing.
Data Source
AI summary
A method to determine image similarities. The method may include obtaining a first image and a second image and determining a discrete transform difference between a first discrete transform of the first image and a second discrete transform of the second image. The method may also include determining multiple first intensity vectors for the first image and determining multiple second intensity vectors for the second image. The method may also include determining an intensity vector difference between the multiple first intensity vectors and the multiple second intensity vectors and determining a color difference between a first color histogram of the first image and a second color histogram of the second image. The method may also include determining a similarity between the first image and the second image based on the discrete transform difference, the intensity vector difference, and the color difference.