Image processing

Many simple image processing functions are often necessary for image pre-processing. Arithmetic and logical operators can be used to add, subtract, multiply and divide images and/or constants among each other or link them to each other using logical truth tables.

Filter functions or morphological transformations suppress unwanted image noise or blurs and amplify desired effects such as contrast.

Functions for shading correction reduce the effects of uneven lighting. Color space conversion and classification allows the operations to be used for color image processing as well. Segmentation—the separation of foreground and background—is the basis of particle analysis.

Image analysis

The analysis functions of nVision use images such as line profiles or histograms to determine statistical information. Blob analysis or particle analysis can be used to group related pixel areas into regions. After segmentation and separation into individual objects, several geometric or intensity-based features can be calculated. A filtering or classification function divides the particles into classes.

Pattern matching

The functions for pattern matching are well-suited for finding known objects in images. To do this, a pattern is trained and then searched for in other images. nVision supports various metrics and allows changes to be made to intensity and distortion.

Geometric matching can also tolerate masking and interference.

Optical character recognition

Many industrial applications require that labels be recognized. The character recognition of nVision offers two application modes: Optical character recognition (OCR) and optical character verification (OCV).

Character sets and parameter configuration can be trained in advance to ensure maximum recognition rates, even when image quality is low.

Barcode/matrix code decoding

nVision's barcode and matrix code decoder can locate and decode printed, etched, lasered or punched codes of a variety of one-dimensional or two-dimensional symbologies.

In addition to pure decoding, the print quality of the code can be checked.

Calibration

Tolerances for sensor assembly, lens distortions and freely positioning the camera in space result in a non-linear relationship between pixel coordinates and positions in the real world.

The camera calibration of nVision enables pixel coordinates to be transformed into real coordinates. The function supports correction of non-linear deformations using the lens and angles.

The calibration is enhanced by the addition of the functions for geometric pixel image correction.

Camera support

nVision supports many industrial cameras from various manufacturers and different interface standards (GigE Vision, USB3 Vision, GenlCam).

Measurement

The measurement functions can be used to determine dimensions from the image quantitatively such as distances, angles, areas, line fits and curve fits.

In this process, the positions are determined down to the subpixel level using the grayscale values of the images. Together with the functions for camera calibration, this forms the basis for aligning parts or taking high-precision measurements.

Interaction

Using nVision means that you can both control Optotune LiquidLens modules from the software and control Gardasoft flash controllers.

This means that additional peripheral equipment can be controlled directly from nVision without any additional steps.

Communication

nVision communicates with the outside world through conventional industry standards:

This means that communication over TCP/IP TwinCat is a problem-free option, as is Profinet (using an optional Hilscher module).

Surface inspection

Scratches, holes, breaks, cavities, dirt—there are many defects that impede the quality of surfaces. Different types of defects often require different lighting (incident light, transmitted light, bright field, dark field) and they can often be detected only when different types of lighting are combined.

nVision has tools for surface inspection, such as multi-channel blob analysis, that allow the software to detect defects and classify types of defects easily.

Grading

The barcode and matrix code tools in nVision can determine the print quality of codes in accordance with ISO/IEC 15415, 15416 or AIM DPM 1-2006.

This allows you to ensure that you codes can always be read accurately.

The OCV functions also allow you to check the quality of labels.

Completeness

In many production steps, it is important that parts be checked for completeness.

This can be done using various nVision tools, e.g. template matching, color matching, edge measuring and decoding codes or text.

Dimensional measurement

You can achieve high-precision results using our telecentric lens and telecentric lighting coupled with the precise measurement algorithms of nVision.

The desired measurement (position, length, angle, edge position, area, etc.) is defined within a search range and evaluated precisely. The measurement results can be combined and can be either exported as a report or processed to make a good/bad statement.

Deformation monitoring

The movement or deformation of large or small parts can be measured and checked by the image processing function in real time.

The real-time monitoring can be used for objects ranging all the way from small machine parts to large building structures.

Calibration

Even when low-distortion lenses are being used, camera calibration is absolutely necessary if you want to take measurements with maximum accuracy.

Calibration is a way of compensating for the distortion of the lens-camera system so that the sub-pixel algorithms can deliver results at the highest possible level of accuracy.

Density measurement

The optical density is a measurement of product quality in some production processes, for example, when applying coatings.

When certain lighting is used, the coating thickness is the result of different grayscale values or colors that can be measured by the image processing function.

You can use image processing to keep the tolerances of your production processes within strict limits.