Binarization Unit Detects Closed Regions for Accurate Image Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing techniques struggle to accurately binarize multi-valued images, often merging objects like characters with their backgrounds due to inappropriate threshold values, especially when luminance differences between regions are subtle.
Innovation Solution
An image processing apparatus and method that detects closed regions within multi-valued images, adjusting binarization based on luminance inside and around these regions, and considering the shape and size characteristics of the regions to differentiate objects from their periphery, using a binarization unit to determine threshold values dynamically and binarize pixels accordingly.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a threshold value is determined based on a luminance histogram of a region to be binarized, then the binarization processing can be performed efficiently, but objects such as characters may be merged with the background
Solution Approach 1:
The image is divided into multiple regions of interest (ROIs) based on edge detection and contour analysis. Each ROI is processed independently with its own threshold value determined from its luminance histogram, allowing objects to be differentiated from background even when luminance differences are subtle.
Solution Approach 2:
Instead of applying a global threshold to the entire image, the patent applies local thresholding to each ROI. The threshold value is determined based on the luminance histogram of each specific region, enabling adaptive binarization that preserves object-background differentiation across different luminance conditions.
2Device complexity
If binarization is performed using a single threshold value for the entire image, then the processing is simple and fast, but objects with subtle luminance differences merge with their background
Solution Approach 1:
The image is segmented into multiple ROIs through edge detection and contour analysis. Each ROI is then binarized using its own locally-determined threshold value, achieving high binarization accuracy without requiring complex global thresholding algorithms.
Solution Approach 2:
The patent performs preliminary edge detection and ROI segmentation before binarization. This preliminary action identifies regions that require special attention, allowing the subsequent binarization process to focus computational resources on critical areas while maintaining overall processing efficiency.
3Manufacturing precision
If the luminance threshold is adjusted to separate objects from background, then object differentiation improves, but processing time and computational complexity increase
Solution Approach 1:
By segmenting the image into ROIs based on edge detection, the patent reduces the computational domain for threshold determination. Each ROI's threshold is calculated from its own luminance histogram, which is computationally more efficient than analyzing the entire image, thus maintaining accuracy while reducing processing time.
Solution Approach 2:
The patent applies complex thresholding only to specific ROIs that contain objects of interest, identified through edge detection. Other regions are processed more simply, reducing overall computational complexity while maintaining high accuracy for critical object regions.
Data Source
AI summary
Provided is an image processing apparatus including an acquisition unit configured to acquire a multi-valued image and a binarization unit configured to generate a binary image obtained by binarizing the multi-valued image, and the stated image processing apparatus is configured such that the binarization unit detects a closed region within the multi-valued image, and binarizes the inside of the closed region based on luminance inside the closed region and luminance around the closed region.