Image-Based Product Comparison via OCR Attribute Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Users face difficulties in searching for products due to insufficient knowledge, leading to poor search results when inputting keywords in keyword-based image search technologies.
Innovation Solution
A method and apparatus for comparing data of the same attribute between at least two products recognized in an image, displaying the comparison result through labeling on the corresponding image, using a user interface to select and highlight common attributes, and providing the comparison result in a graphical form based on field values.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If keyword-based image search technology is used, then product search functionality is provided, but users with insufficient product knowledge cannot input proper keywords leading to poor search results
Solution Approach 1:
The patent introduces an image as an intermediary medium between the user and the product database. Instead of requiring users to directly input keywords, they upload product images which are then processed by OCR technology to extract text information and identify products, serving as a bridge that eliminates the need for keyword knowledge
Solution Approach 2:
The patent replaces the manual keyword input mechanism with automated optical character recognition (OCR) technology. The system captures text from uploaded images, processes it automatically, and retrieves products based on extracted information, substituting the user's manual text input action with an automated optical recognition system
2Loss of information
If all product categories are displayed for user verification, then complete product information is provided, but users must directly verify products in all categories which increases complexity
Solution Approach 1:
The patent extracts only the relevant product information from the uploaded image through OCR technology and presents it to the user for verification. Instead of displaying all product categories and information, the system selectively extracts and displays only the text and product details that are actually present in the uploaded image, reducing unnecessary complexity
Solution Approach 2:
The patent segments the product comparison process into distinct stages: image upload, OCR text extraction, product identification, information verification, and comparison. By dividing the complex task of product search and comparison into these manageable segments, the system reduces the cognitive load on users while maintaining information completeness
3Productivity
If product comparison is performed without visual feedback, then comparison data is processed, but users cannot easily understand comparison results
Solution Approach 1:
The patent uses color-coded visual indicators to represent different comparison results. Products are displayed with color markers indicating their relative positions in comparisons (e.g., superior, inferior, equal), allowing users to quickly grasp comparison outcomes through color differentiation without needing to interpret numerical data
Solution Approach 2:
The patent transforms one-dimensional comparison data (textual or numerical attributes) into two-dimensional visual representations by overlaying comparison results directly on product images. This spatial dimensionality change allows users to perceive comparison information in the context of the actual product appearance, enhancing clarity and understanding
Data Source
AI summary
Provided are a method, a system, and a non-transitory computer-readable record medium for comparing common features of products and providing a comparison result. A product comparison method includes recognizing at least two comparable products from at least one image; displaying at least one common attribute of the at least two comparable products through a user interface; and based on the user interface receiving a user input that selects one of the at least one common attribute, as a selected attribute, providing a result of comparison between the at least two comparable products with regard to the selected attribute.