Image Processing Apparatus for Cartilage Region Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing methods struggle to precisely recognize tissues with large density value fluctuations, such as cartilage, and require complex operator input for region extraction in images like those from the head, due to non-continuous and dispersed nature of cartilage regions.
Innovation Solution
An image processing apparatus and method that applies predetermined threshold conditions to a determination range including a target pixel and surrounding pixels, repeatedly moving the range to identify recognized pixels, allowing for precise recognition of tissues with density fluctuations without manual starting point specification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If the region growing method is used to extract cartilage regions, then the extraction process can be automated, but it requires specifying a starting point for each process which makes the operation very complex
Solution Approach 1:
The system automatically performs threshold determination by scanning the entire image and applying threshold conditions to determination ranges including target pixels and surrounding pixels. The process execution unit executes threshold determination repeatedly after moving the determination range successively, enabling the system to self-identify cartilage regions without requiring operator intervention to specify starting points.
2Ease of operation
If conventional threshold methods are used for region extraction, then the process is simple to operate, but it is difficult to precisely distinguish tissues with large fluctuation of density values such as cartilage
Solution Approach 1:
The threshold determination unit specifies the target pixel as a recognized pixel based on threshold conditions applied to a determination range including the target pixel and surrounding multiple pixels. This local evaluation approach allows precise identification of cartilage pixels by considering neighborhood context, improving measurement precision while maintaining operational simplicity through automated processing.
Solution Approach 2:
The invention transitions from single-pixel threshold evaluation to multi-pixel determination range evaluation. By applying threshold conditions to determination ranges that include target pixels and surrounding pixels, and by repeatedly executing threshold determination after moving the determination range successively, the system achieves precise cartilage recognition in three-dimensional space.
3Extent of automation
If the determination range is moved successively to scan the entire image, then all cartilage regions can be recognized without manual starting point specification, but the processing time increases
Solution Approach 1:
The process execution unit executes threshold determination repeatedly after moving the determination range successively throughout the image. This continuous scanning approach ensures that all cartilage regions are recognized automatically without manual intervention, achieving complete automation despite the increased processing time required to examine the entire image.
Data Source
AI summary
In order to provide an image processing apparatus etc. that can precisely recognize multiple regions with density value fluctuation in an image with a simple operation in a process of recognizing a specific region from the image, a CPU performs threshold determination for a target pixel and the multiple pixels surrounding the target pixel (determination range) included in a three-dimensional original image by applying predetermined threshold conditions and specifies the target pixel as a recognized pixel when the threshold conditions are satisfied. The threshold conditions preferably apply different thresholds between a pixel on the same flat surface with the target pixel and a pixel on the other flat surface. By successively moving the target pixel (determination range) and repeating the above-mentioned threshold determination, it is performed also for the entire three-dimensional original image. This allows an operator to more precisely recognize a tissue with density value fluctuation such as a cartilage automatically without setting a starting point.