Road Surface Crack Detection Through Oblique-Image Resolution Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Image data obtained by taking images from an angle on a road surface has varying resolution, leading to potential inaccuracies in crack detection.
Innovation Solution
A crack detection device that includes an image acquisition unit, an image classification unit to categorize data into acceptable and unacceptable ranges based on resolution, and a data output unit to utilize only the acceptable data for crack detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If image data is obtained by taking images from an oblique direction, then the device complexity is reduced and ease of operation is improved, but the measurement precision deteriorates due to varying resolution
Solution Approach 1:
The patent divides the image data into multiple regions based on resolution characteristics. The image classification unit segments the oblique image data into acceptable and unacceptable ranges, allowing the system to process only the high-resolution portions for crack detection while discarding or interpolating the low-resolution portions, thus maintaining measurement precision without requiring a dedicated ortho imaging device
Solution Approach 2:
The patent applies different processing qualities to different regions of the image data. High-resolution regions are used directly for crack detection, while low-resolution regions are either discarded or undergo interpolation treatment. This local differentiation ensures that crack detection precision is maintained in critical areas while accommodating the varying resolution characteristics of oblique imaging
2Device complexity
If image data with varying resolution is used for crack detection, then the device complexity is reduced, but the reliability deteriorates due to false negatives
Solution Approach 1:
The patent performs preliminary classification of image data into acceptable and unacceptable ranges before crack detection. By pre-filtering the data to identify regions with sufficient resolution, the system prevents unreliable low-resolution data from compromising crack detection accuracy, thereby maintaining high reliability while using simpler oblique imaging equipment
Solution Approach 2:
The patent introduces an image classification unit as an intermediary between the oblique image acquisition and crack detection processes. This intermediary component evaluates resolution characteristics and selectively passes only the acceptable high-resolution data to the crack detection algorithm, preventing false negatives while avoiding the need for complex ortho imaging hardware
Data Source
AI summary
In a crack detection device (10), an image acquisition unit (21) acquires image data acquired by taking an image of a road surface from an oblique direction with respect to the road surface. An image classification unit (22) classifies image data acquired into an acceptable range with a resolution higher than a standard value, and an unacceptable range with a resolution equal to or less than the standard value. A data output unit (23) outputs acceptable data being image data of a part classified into the acceptable range as data to detect a crack on the road surface. An image display unit (24) displays data output.