Multi-pass collimator blade edge detection in digital radiography
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing digital radiography image processing methods, such as the Hough transform, often fail to detect weak collimator blade edges accurately, leading to erroneous area selection and compromised diagnostic image quality.
Innovation Solution
A two-pass edge detection process is employed, using conventional edge detection algorithms like the Hough transform in the first pass and histogram matching for image enhancement in the second pass, with results combined to improve edge detection and define the shutter area.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the Hough transform is used to detect collimator blade edges, then the process is simple and fast, but weak edges are not detected accurately leading to erroneous area selection
Solution Approach 1:
The patent applies preliminary action by performing image enhancement through histogram matching before edge detection. This preprocessing step strengthens weak edges in the image, ensuring they are detectable in subsequent edge detection passes. The enhancement is performed on the original image to improve the visibility of collimator blade edges before any detection algorithms are applied.
Solution Approach 2:
The patent implements periodic action through multiple passes of edge detection. Instead of a single detection attempt, the system performs edge detection multiple times with different parameters and on enhanced images, combining results from each pass. This multi-pass approach ensures that weak edges detected in one pass are complemented by stronger edges detected in other passes, achieving both speed and accuracy.
2Measurement precision
If image enhancement is applied to strengthen weak edges, then edge detection accuracy improves, but processing time and complexity increase
Solution Approach 1:
The patent applies parameter changes by modifying image characteristics through histogram matching to enhance edge visibility. This transformation adjusts the distribution of pixel intensities to strengthen weak edges while preserving the overall image structure. The enhanced image parameters are then used in subsequent detection passes, improving accuracy without requiring fundamentally different detection algorithms.
Data Source
AI summary
A process for detecting the edges of collimator blades in digital radiography images in the first pass detects the edges of the collimator blades using original image, and the in the second pass repeats edge detection using an image enhanced by a histogram matching technique, for example. The edge detection using an enhanced image may also be repeated any number of times in cases of complex anatomy or when selected radiographic techniques does do not provide sufficient imaging data. The results of the second pass, or the collection of the results of multiple second passes, are then combined with the result from the first pass to form a list of the potential blade edge candidates. A desirable number of edges are then selected from the combined list to form a polygon which encloses the target area of the image, thereby providing the shutter area.