Software Modules

This is an important factor that contributes to image quality. Optimal illumination should reduce shadow, noise, and reflection and increase image contrast. The Acumen system can be easily configured to achieve optimal illumination which is saved to the AMM project for measurement accuracy and consistency.

This can be achieved with the Acumen system dynamic zoom capability to accommodate a range of part sizes.

without a jig can be easily achieved with image alignment. The AMM algorithm detects the positional relationship of the test object relative to the alignment reference features in the project file and corrects the positioning of the test object. Onscreen graticules can also be used to aid object placement

These are parts or patterns of an object in an image that is to be used for measurement. Features include properties like corners, edges, regions of interest, points etc. Feature types are defined using edge detection, feature detection or segmentation.

around the object feature defines the image pixels where image processing is to be performed.

Locates the sharp changes in pixel intensities in the image ROI to find the edge boundaries between the object feature and background. The direction of the feature edge can then be selected in the toolbar. A strong edge is important for repeatable and accurate robust measurements – thresholding and blur can be applied to improve the edge pixel boundary. Selecting the output requirement for each feature edge makes it available for measurement of distances, angles, areas, and offsets in the measurement mode.

This is used to detect an object feature that matches the template image in the project.  When a specific feature on an object may not have a significant edge, feature detection may be useful in detecting that feature for measurement. The feature must be unique within the specified ROI. Several images can be used to create a template feature image, where the necessary features have been extracted, grouped together to obtain a normalized feature image. The template matching then uses matching criterion to determine the similarity between the template image and test image.

This can be used to distinguish and detect features by colour. A clustering algorithm divides the image ROI into colour clusters. The number of clusters must be selected for the algorithm to cluster the image data accordingly. The colour selector allows the user to select which colour cluster is to be detected., creating a mask where that that cluster is located on the image.   Edges of this mask can then be used as measurement features.

These can be created, enabling measurements from lines, circles, or points.