Image Search Apparatus Reducing Communication Load via Region Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image search systems face high communication loads and processing burdens when comparing large numbers of images, especially when images are transmitted over networks, due to the need to compare entire images rather than focusing on specific similarities.
Innovation Solution
An image search apparatus and system that extracts common areas from stored images, determines non-common search areas, and transmits partial image information for comparison, reducing the need to transmit and compare entire images by focusing on specific search areas and requesting document images only when necessary.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If entire images are compared for search, then search accuracy is improved, but communication load and processing burden increase
Solution Approach 1:
The patent divides images into multiple regions (e.g., header region, body region, footer region) and compares only the relevant regions instead of entire images. This segmentation reduces the data volume for comparison while maintaining search accuracy by focusing on discriminative areas.
Solution Approach 2:
The patent extracts and transmits only the necessary image regions (search regions) to the search apparatus, rather than transmitting entire images. This extraction principle reduces communication load by sending only the relevant portions needed for accurate comparison.
2Adaptability or versatility
If entire images are transmitted over network, then complete image search is enabled, but network communication efficiency deteriorates
Solution Approach 1:
The patent extracts only the necessary search regions from images and transmits these extracted portions over the network, rather than transmitting entire images. This significantly reduces network communication load while preserving the ability to perform accurate image search by comparing only the relevant transmitted regions.
3Measurement precision
If all image areas are compared, then comprehensive search is achieved, but processing time increases
Solution Approach 1:
The patent segments images into multiple regions and performs comparison operations only on the segmented search regions rather than entire images. This reduces processing time by limiting comparison scope to relevant areas while maintaining search comprehensiveness through strategic region selection.
Solution Approach 2:
The patent applies partial action by comparing only the necessary search regions (header, body, footer) rather than performing exhaustive comparison on entire images. This partial comparison approach achieves sufficient search comprehensiveness for practical purposes while dramatically reducing processing time.
Data Source
AI summary
An image search apparatus capable of communicating with an image scanning apparatus includes: a common area extraction portion to compare a plurality of registration images stored in the storage portion and to extract a common area including an identical or similar image among a plurality of registration images; a search area determination portion to determine a non-common area excluding the common area, as a search area; a transmission request portion to request transmission of a document image if, as a result of comparison by the comparison portion, a registration image in which the reference image is identical or similar to the partial image does not exist in the plurality of registration images; and a registration portion to, in response to a request by the transmission request portion, store a document image received from the image scanning apparatus as a registration image into the storage portion.