Image Search Extractor Using Color Attribute Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image search devices face inefficiencies in searching for area images of unspecified target objects from entire images, especially when the type of target object varies greatly, requiring complex detection algorithms and increased calculation, and struggle to associate attribute information effectively.
Innovation Solution
An image search system that includes an extractor to acquire and extract area images and attribute information based on color, an information storage unit to search for area images with designated attribute information, and an interface to output search results, allowing for efficient extraction and identification of area images without needing pre-defined algorithms for each type of target object.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a detection algorithm is created for each target object type to handle great variation in photographed target objects, then the ability to search for diverse target objects is improved, but the amount of calculation required for detection processing increases
Solution Approach 1:
The patent creates a single detection algorithm that can detect multiple types of target objects (vehicles, pedestrians, animals, etc.) by using common attributes such as color, shape, and size ranges. Instead of developing separate algorithms for each object type, the system uses one universal algorithm that processes all objects through the same detection pipeline, thereby maintaining versatility while reducing computational overhead.
2Adaptability or versatility
If attribute information is not associated with target objects in advance, then the system can handle unspecified target objects, but the time required to search for area images increases
Solution Approach 1:
The patent performs preliminary extraction of attribute information (color, shape, size) for all detected target objects and stores this information in advance in a database. When a search is initiated, the system quickly retrieves pre-extracted attribute information and compares it with the search conditions, rather than performing detection and attribute extraction during the search process. This preliminary action significantly reduces search time while maintaining the ability to handle any target object type.
3Measurement precision
If the amount of attribute information of target object is increased to improve search accuracy, then the precision of searching is improved, but the device complexity increases
Solution Approach 1:
The patent extracts only the most essential and discriminative attribute information (color, shape, size) that is sufficient for effective image search. Rather than extracting and storing all possible attributes of target objects, the system selectively extracts only those attributes that are most useful for differentiation and search purposes. This extraction of key attributes improves search precision while avoiding the complexity that would arise from processing comprehensive attribute sets.
Data Source
AI summary
An image search device includes an extractor configured to acquire an entire image including an area image that is an image of a target object and extract one or more combinations of the area image and attribute information from the entire image on the basis of color, an information storage unit configured to search for the area image combined with the designated attribute information from the extracted one or more combinations of the area image and the attribute information, and an interface configured to output a search result.