Image Processing Apparatus Pattern-Based Search Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing technologies face inefficiencies in creating an image corpus due to extensive computation time and inefficient similarity searches, particularly when searching for similar areas in large datasets, as they often require searching across the entire image and do not effectively utilize patterns of objects with assigned meta-data.
Innovation Solution
The proposed solution involves an image processing apparatus that predicts patterns of objects after meta-data is assigned, limits search ranges based on these patterns, and uses graph matching to efficiently find similar areas by converting distribution patterns into graphs and calculating priority levels for search results, thereby streamlining the search process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If similar area search is performed across the entire image to ensure comprehensive search coverage, then search completeness is improved, but computation time increases significantly
Solution Approach 1:
The image is divided into multiple blocks, and the search is performed only within a limited range around the specified area rather than across the entire image. This segmentation approach maintains search completeness for relevant areas while significantly reducing computation time by excluding unrelated regions from the search process
Solution Approach 2:
Pattern information is extracted and stored in advance from training images, creating a database of object patterns before the actual search is performed. This preliminary action enables the search to focus on matching against pre-processed patterns rather than analyzing all image data from scratch, reducing computation time while maintaining accuracy
2Measurement precision
If manual meta-data assignment is performed for each object to ensure accuracy, then assignment accuracy is improved, but operational efficiency deteriorates
Solution Approach 1:
The system automatically assigns meta-data to images by extracting pattern information from training images and matching it against target images. This self-service approach eliminates the need for manual meta-data assignment while maintaining high accuracy through pattern-based automatic recognition, thereby significantly improving operational efficiency
Solution Approach 2:
Pattern information is copied from training images with known meta-data and applied to target images through pattern matching. This copying approach allows accurate meta-data assignment to be automatically replicated across multiple images without requiring manual intervention for each image, improving both accuracy and efficiency
3Productivity
If pattern-based search limitation is applied to reduce search range, then search efficiency is improved, but search coverage may be reduced
Solution Approach 1:
The search limitation is applied selectively based on local pattern characteristics rather than uniformly across all search cases. The system determines the appropriate search range by analyzing the specific pattern being searched, allowing comprehensive coverage when needed while limiting search only where pattern-based constraints are valid, thus maintaining reliability while improving efficiency
Data Source
AI summary
An image processing apparatus includes: a storage configured to store, for a plurality of areas in a target image, attributes and patterns associated with the attributes, each the attributes being assigned to each the plurality of areas, and each the pattern being created from areas having an identical attribute among the attributes; and a processor configured to: obtain an area that is specified in the target image and an attribute that is assigned to the specified area, search the storage for a pattern based on similarity between the pattern and a pattern created from a certain number of areas having the assigned attribute; and search for an area in the target image based on similarity between the area and the certain number of areas in accordance with the searched pattern.