Color Search System Using Image Processing for Accurate Item Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing search by color techniques in online shopping are inaccurate due to reliance on textual descriptions, inconsistent color terminology, and lack of explicit color information, leading to unsatisfactory search results.
Innovation Solution
A method that processes item images to obtain color space parameter values and names, which are used to create a database of color attributes, allowing for accurate color-based searches by matching dominant colors rather than textual descriptions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If text-based search is used to find products by color, then the search process is simple to implement, but the search accuracy deteriorates due to inconsistent color terminology and brand names containing color words
Solution Approach 1:
The patent replaces the mechanical text-based search system with an optical image processing system. Instead of searching through text descriptions using keywords, the system captures images of products, extracts color information from pixel data, and performs color-based matching. This substitution of mechanical text processing with optical analysis resolves the contradiction by achieving both implementation feasibility and high search accuracy.
Solution Approach 2:
The patent changes the search parameter from text-based keywords to color space parameters (RGB, HSV values). By converting color information into standardized numerical parameters and creating tolerance ranges, the system enables precise color matching while maintaining simple search operations. This parameter transformation resolves the contradiction between ease of implementation and search accuracy.
2Adaptability or versatility
If different color names are used to refer to the same color across websites, then the system is flexible in terminology, but the color identification accuracy deteriorates
Solution Approach 1:
The patent applies homogeneity by standardizing color representation across all products regardless of the original terminology used. All color information is converted into a unified color space parameter system with defined tolerance ranges, ensuring consistent color identification. This homogenization of color data representation resolves the contradiction by maintaining terminology flexibility in data collection while achieving precision in color matching.
3Device complexity
If text descriptions are used to store color information, then the storage method is simple, but the completeness of color information deteriorates when color is not explicitly described
Solution Approach 1:
The patent creates a digital copy of the actual product color by capturing images and extracting color parameters from pixel data. This color copy stored in standardized parameters preserves the complete color information without relying on text descriptions. The copying of visual color information into structured data resolves the contradiction between storage simplicity and information completeness.
Data Source
AI summary
A system and method of facilitating search by color processes an image of an item to obtain its color space parameter values and/or color names based on the color space parameter values, and adds the color space parameter values and/or color names as color attribute of the item. When a user inputs a color as a search criterion, the method searches the color attributes of items, and returns a list of items matching the color selected by the user.