Sub-pixel Offset Image Combination for Machine Vision Resolution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Precision machine vision inspection systems face challenges in achieving high resolution and accuracy using lower magnification levels, which are less expensive but result in lower image sampling density, making it difficult for unskilled users to obtain desired levels of resolution and accuracy without complex and mechanically unstable dithering motions.
Innovation Solution
The method involves acquiring a plurality of images with sub-pixel offsets relative to each other, combining them to increase image sampling density, and using a fast multiple image acquisition mode with controlled camera and stage motion to achieve higher resolution than the native camera resolution, allowing for accurate edge detection and measurement.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If lower magnification levels are used, then cost is reduced, but image sampling density and resolution decrease
Solution Approach 1:
The inspection process is segmented into multiple image acquisitions at different sub-pixel offsets. Instead of relying on a single high-magnification image, the system divides the sampling task across multiple images taken at lower magnification but with offset positions, thereby achieving high effective sampling density without requiring expensive high-magnification optics.
Solution Approach 2:
The system adds the dimension of spatial offset between multiple images. By acquiring images at different sub-pixel offsets (shifting the imaging area relative to the pixel array) and combining them, the system effectively increases sampling density in the spatial domain without changing the optical magnification, thus resolving the contradiction between cost and resolution.
2Measurement precision
If dithering motion is used to increase sampling density, then resolution is improved, but mechanical complexity and vibrations increase
Solution Approach 1:
The system replaces complex mechanical dithering motion with a computational approach. Instead of physically moving the optical system or stage in a dithering pattern, the system acquires multiple images at fixed positions with different sub-pixel offsets and uses software algorithms to combine them, thereby achieving high resolution without mechanical complexity or vibrations.
Solution Approach 2:
The patent introduces a computational intermediary (image processing algorithm) that combines multiple images with sub-pixel offsets. This intermediary process acts as a mediator between the simple mechanical positioning and the desired high-resolution output, achieving resolution enhancement without requiring complex mechanical dithering systems.
3Measurement precision
If multiple images with sub-pixel offsets are acquired and combined, then effective sampling density is increased, but image acquisition time increases
Solution Approach 1:
The system maintains continuous useful action by acquiring multiple images in rapid succession without interrupting the inspection process. The images are captured sequentially with minimal delay, and the combination process occurs concurrently with other system operations, thereby minimizing the time penalty while achieving increased sampling density.
Data Source
AI summary
A method in a machine vision inspection system for obtaining two images of a workpiece with a desired sub-pixel offset between the images. The images are acquired using a fast multiple image acquisition mode of operation of a camera in the machine vision inspection system. In various implementations, the method includes removing the offset between the images such that the workpiece is congruent in the images and combining the congruent image data. The combined image data has a resolution better than that allowed by the native resolution of a camera that acquires images in the machine vision inspection system. The method may be implemented in an edge feature video tool for measuring edge features on the workpiece. The motion direction utilized for obtaining the two images may be made to be transverse to the edge direction of an edge that is being measured.