Image Edge Correction Using Directional Filters and Mean Pixel Values
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing technologies fail to effectively enhance image edges while maintaining image brightness, particularly in images captured using various modalities like visible light, infrared, radiation, ultrasonic waves, and Magnetic Resonance signals.
Innovation Solution
An image processing apparatus and method that employs directional filters, such as Gabor filters, to detect edges, calculate mean pixel values, and apply correction values inversely proportional to these values to enhance edges without increasing image brightness, while also incorporating noise processing to remove speckle noise.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional image processing methods are used to enhance edges, then edge definition is improved, but image brightness is altered or lost
Solution Approach 1:
The patent applies different processing strengths to different regions of the image by using edge detection to identify edge regions versus non-edge regions. Edge regions receive enhancement processing while non-edge regions maintain their original brightness characteristics, thus improving edge definition without altering overall image brightness.
Solution Approach 2:
Instead of directly enhancing edges through conventional methods that increase overall brightness, the patent inverts the approach by first detecting edges and then applying correction values that preserve brightness. The correction values are determined based on edge detection results but applied in a way that maintains original brightness levels.
2Reliability
If noise processing is applied to remove speckle noise, then image quality is improved, but edge sharpness may be reduced
Solution Approach 1:
The patent segments the image processing into distinct stages: edge detection, noise processing, and correction value application. By separating these functions and applying them in sequence with careful parameter control, the patent achieves noise removal while preserving edge sharpness through the subsequent correction step.
Solution Approach 2:
The patent uses edge detection results as feedback to guide the noise processing and correction value determination. The edge information obtained from detection is fed back into the correction value calculation to ensure that noise reduction does not compromise edge definition, maintaining both image quality and edge sharpness.
Data Source
AI summary
Disclosed herein are an image processing apparatus, an image capturing apparatus, an image processing method, and a method of controlling an image capturing apparatus. The image processing apparatus includes: an input unit through which an image is input, a correction value decider configured to detect an edge of the image by applying at least one directional filter that is alignable in a direction with respect to the image, to calculate a mean value of a plurality of pixel values of a plurality of pixels that correspond to the edge, and to determine a correction value that corresponds to each pixel of the image according to the mean value of the plurality of pixel values, and an image corrector configured to apply the correction value to the image in order to correct the image.