Grayscale Feature Extraction for Color Image Retrieval
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing techniques fail to accurately retrieve original color electronic data from paper documents, especially when the documents are printed in grayscale, leading to unsatisfactory precision and unnecessary changes to the original data.
Innovation Solution
An image processing apparatus and method that extracts grayscale and color feature amounts from a reference image, allowing for precise retrieval of target images by comparing these features, without altering the original data, using storage, input, and processing means to segment, extract, and compare image attributes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If original document data containing color images is retrieved using text data as retrieval condition, then retrieval process is simple, but retrieval precision deteriorates when text data are identical but images are different
Solution Approach 1:
The patent segments the retrieval process into multiple stages: first extracting text features for initial retrieval, then extracting image features (color, brightness, texture) for refinement. This segmentation allows the system to handle both text-based and image-based retrieval requirements separately, resolving the contradiction between simple text-based retrieval and precise image-based retrieval.
Solution Approach 2:
The patent adds image feature dimensions (color, brightness, texture) to the traditional text-based retrieval dimension. By transitioning from one-dimensional text retrieval to multi-dimensional retrieval incorporating image characteristics, the system achieves higher precision without completely abandoning the simplicity of text-based search.
2Reliability
If original document data is printed once together with identification code to enable retrieval, then retrieval capability is achieved, but the original document data is changed in an unintended way
Solution Approach 1:
The patent creates a copy of the original document data for retrieval purposes, storing image features and metadata separately from the original. This allows retrieval operations to be performed on the copy without modifying the original document data, maintaining its integrity while enabling reliable retrieval.
Solution Approach 2:
The patent introduces an intermediary retrieval system that uses extracted image features and metadata as mediators between the original document and the retrieval process. This intermediary layer enables retrieval operations without directly manipulating the original document data, preserving its stability.
3Ease of manufacture
If paper document is created by copying original using copying machine or printing by monochrome printer for reference/confirmation, then distribution is enabled, but color information is lost and retrieval precision deteriorates
Solution Approach 1:
The patent extracts color information and other image features from the grayscale or monochrome printed documents using image processing techniques. By taking out and preserving these color features in the digital domain, the system enables retrieval with high precision even when the physical document appears grayscale or monochrome.
Solution Approach 2:
The patent performs preliminary extraction and storage of color and image features when the document is initially scanned or processed. This preliminary action preserves color information before it is lost in subsequent printing or copying operations, enabling accurate retrieval later without requiring color reproduction at the retrieval stage.
Data Source
AI summary
A reference comparison image obtained from a printed material is input. The grayscale feature amount of the reference comparison image is extracted. Using the extracted grayscale feature amount, a target comparison image corresponding to the reference comparison image is retrieved from a plurality of electronic data stored as target comparison images in a storage medium. The retrieved image is extracted from the storage medium and processed.