Image Measuring Apparatus Noise Reduction Edge Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Image measuring apparatuses face challenges in performing dimensional and form measurements due to noise in images, leading to inaccurate edge point detection and measurement precision issues.
Innovation Solution
The apparatus employs an imaging device and a processing device that generates a lower-resolution image pyramid, segments the image into regions, performs edge detection, and removes outliers to enhance measurement accuracy by reducing noise, with a control program recorded on a non-temporary medium to control these processes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If image processing is performed on high-resolution images to improve measurement precision, then measurement precision improves, but processing time and computational load increase
Solution Approach 1:
The image processing is divided into multiple resolution levels (pyramid structure). Coarse processing is performed on down-sampled lower-resolution images to identify candidate regions, while fine processing is applied only to specific regions of interest in the original high-resolution image. This segmentation of processing across different scales reduces overall computational load while maintaining measurement precision for critical features.
2Measurement precision
If noise filtering is applied to reduce measurement errors, then measurement precision improves, but edge point detection accuracy may deteriorate due to over-smoothing
Solution Approach 1:
Different processing strategies are applied to different regions of the image based on their characteristics. In regions with strong edges, minimal filtering is applied to preserve edge accuracy. In homogeneous regions with noise, more aggressive filtering is applied. This local differentiation allows the system to reduce measurement errors from noise while avoiding over-smoothing that would degrade edge point detection accuracy.
Solution Approach 2:
The problem is solved by transitioning between different resolution dimensions. Noise filtering and candidate region identification are performed on lower-resolution images where noise has less impact. Once candidate regions are identified, precise edge detection is performed on the original high-resolution image, leveraging the multi-dimensional (multi-resolution) approach to balance noise reduction with edge accuracy.
3Measurement precision
If full-resolution image processing is performed to maintain detail accuracy, then measurement precision is maintained, but processing speed decreases
Solution Approach 1:
Preliminary processing steps (down-sampling, coarse filtering, candidate region identification) are performed on lower-resolution images before final high-resolution processing. This preliminary action at reduced resolution quickly eliminates non-candidate regions, so that full-resolution processing is only applied to a small subset of interesting regions, thereby maintaining measurement precision while significantly improving processing speed.
Data Source
AI summary
An image measuring apparatus according to an embodiment of the present invention comprises: an imaging device that images a workpiece to acquire an image of this workpiece; and a processing device that performs measurement of the workpiece based on this image and outputs a measurement result. Moreover, the processing device, based on the above-described image, generates another image whose number-of-pixels is smaller than that of the image, sets a plurality of regions based on this another image, and calculates the above-described measurement result based on these plurality of regions.