Image Correction for Natural Images with Backgrounds
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In the printing industry, particularly with on-demand printers, manually correcting the color of multiple photographed images for business documents is time-consuming and labor-intensive, and the visual appearance of natural images can vary significantly based on the background's lightness, hue, and saturation, affecting the image quality.
Innovation Solution
A data correction apparatus and method that recognizes natural images with added backgrounds, calculates correction conditions based on lightness, hue, and saturation, and applies image corrections to ensure the natural images appear uniform regardless of the background, using a system comprising an image recognizing unit, natural image determining unit, image analyzing unit, and correction condition calculating units within a RIP (Raster Image Processor) environment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If manual color correction is applied to each photographed image using retouching software, then color accuracy of individual images is improved, but time consumption and labor intensity increase significantly
Solution Approach 1:
The system automatically performs color correction on photographed images without requiring manual operation. The correction unit autonomously analyzes image characteristics and applies appropriate color corrections, eliminating the need for operators to manually adjust each image while maintaining color accuracy.
Solution Approach 2:
The system changes color parameters (hue, saturation, lightness) automatically based on image analysis. By adjusting these parameters programmatically rather than manually, the system achieves accurate color correction while significantly reducing time consumption and labor requirements.
2Ease of operation
If background illustration images are added around natural images, then visual appeal of pages is improved, but color appearance of natural images changes due to color contrast phenomenon
Solution Approach 1:
The system preemptively compensates for the color contrast effect by applying inverse color corrections to natural images before background images are added. For example, if a light background will make the natural image appear darker, the system pre-lightens the natural image to counteract this effect, thereby maintaining color appearance consistency while still allowing background images to enhance visual appeal.
Solution Approach 2:
The system analyzes the color characteristics of both natural images and background images, then uses this feedback information to calculate appropriate correction amounts. This feedback mechanism ensures that the color correction applied to natural images accounts for the specific background that will be added, maintaining color consistency across different background conditions.
3Productivity
If automatic color correction is applied to all images in pages, then processing efficiency is improved, but natural images with backgrounds may appear unnatural due to over-correction
Solution Approach 1:
The system applies different correction strategies to different types of images based on local characteristics. Natural images are analyzed to determine whether they have backgrounds, and correction amounts are adjusted accordingly. Images without backgrounds receive standard automatic correction, while natural images with backgrounds receive modified correction that preserves their natural appearance.
Solution Approach 2:
The correction amount is dynamically adjusted based on image characteristics and background conditions. Rather than applying a fixed correction to all images, the system varies the correction parameters according to the specific natural image and its background, maintaining naturalness while processing efficiently.
Data Source
Figure 1
Figure 2
Figure 3~4
AI summary
Natural appearance to the eye of a natural image with an added background in a page represented by page description data is provided, regardless of lightness, hue and/or saturation of the background. An image recognizing unit (10) recognizes images in a page represented by page description data, and a natural image determining unit (11) determines whether each recognized image is a natural image. An image analyzing unit (12) calculates a setup condition for image correction for the natural image. A second correction condition calculating unit (15) calculates, for each natural image with a background illustration image among the natural images, a correction condition depending on lightness, hue and/or saturation of the background illustration image. An image correcting unit (16) applies image correction based on the setup condition and the correction condition to the natural image with the background corresponding to the setup condition and the correction condition.