Linear Object Area Detection via Line-Segment Image Rotation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional image analysis methods for detecting target object regions in captured images, such as pipelines, suffer from non-detection or excessive detection due to the orientation of the target objects relative to the pixel arrangement, and manual affine transformation selection of points leads to inaccuracies.
Innovation Solution
A region detection device and method that includes a line segment detection unit, a main line segment extraction unit, and a correction unit to rotate the captured image such that the main line segment is perpendicular or parallel to the pixel arrangement, followed by region detection from the corrected image.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If conventional image analysis methods are used to detect target object regions, then detection can be performed automatically, but non-detection or excessive detection occurs due to target object orientation relative to pixel arrangement
Solution Approach 1:
The patent applies preliminary action by performing image rotation before region detection. The detection device rotates the captured image so that the target object's longitudinal direction becomes parallel to the pixel arrangement direction, then performs region detection on the rotated image. This preliminary orientation adjustment ensures that the detection algorithm works with optimally oriented data, eliminating the non-detection and excessive detection problems that occur when objects are at oblique angles to the pixel grid.
2Adaptability or versatility
If manual point selection for affine transformation is used, then tilt correction can be applied, but detection accuracy deteriorates due to manual selection errors
Solution Approach 1:
The patent applies self-service by enabling the system to automatically determine the rotation angle and perform image rotation without manual intervention. The detection device automatically detects line segments in the captured image, calculates the angle between these line segments and the pixel arrangement direction, and rotates the image by this calculated angle. This automated approach eliminates the manual point selection required in traditional affine transformation methods, removing the source of human error while maintaining tilt correction capability.
3Adaptability or versatility
If target objects extend obliquely to pixel arrangement direction, then imaging can capture the objects, but region detection accuracy decreases due to mismatch with pixel grid
Solution Approach 1:
The patent applies parameter changes by transforming the image orientation parameter. Instead of changing the target object's physical orientation or the detector's orientation, the system changes the parameter of image pixel arrangement through rotation. By rotating the image so that the target object's longitudinal direction becomes parallel to the pixel arrangement direction, the system optimizes the geometric parameter relationship between the object and the detection grid, thereby maximizing detection accuracy while maintaining the ability to image objects at various orientations.
Data Source
AI summary
A region detection device (1) according to the present disclosure includes a line segment detection unit (32) that detects a line segment from a captured image including an image of a target object having a linear band shape, a main line segment extraction unit (33) that extracts a main line segment (SL) that is a line segment having a length equal to or more than a predetermined value; a correction unit (34) that generates, as a corrected image, a rotated image obtained by rotating the captured image such that the main line segment (SL) is perpendicular or parallel to an arrangement direction of pixels in the captured image, and a region detection unit (35) that detects a target object region indicating the image of the target object from the corrected image.