Label Inspection

NI Vision Assistant

Label Inspection

Application Type Inspection
Image Characteristics Monochrome; Good Contrast
Machine Vision Tool(s) Pattern Matching; Golden Template Comparison; Particle Analysis

The label inspection example demonstrates finding defects such as double-prints, scratches, and lack of ink on a printed label. The defects are found by defining a golden template of a perfect label and comparing it to the inspection images. The particle analysis step at the end of the inspection characterizes the defects.

Pattern Matching—Finds the printed label in the image. A region containing the printed label is used as the template. Because the label can rotate slightly, rotation-invariant pattern matching is used to locate the label. The Pattern Matching function finds the location of the label and returns the pixel coordinates of the location.

Golden Template Comparison—Uses the same template from the pattern matching step as a golden template. This template represents a "perfect" label in the first image. The golden template comparison uses the coordinates returned by the pattern matching step to position the golden template in the image and compare the golden template with the corresponding pixels of the image. The difference image highlights the defects between the golden template and the image under inspection.

Particle Analysis—Analyzes the properties of the defects in the image. The particle measurement function can analyze up to 50 different properties of a particle.