Advantages and limitations of 2D and 3D machine vision detection technology

Machine vision is generally composed of industrial light sources, image acquisition units, image processing units, image processing software and network communication devices. Both 2D and 3D technologies play an important role in the field of quality control and on-line inspection in the automation industry. How to combine the two to create a more reliable and efficient machine vision detection system, we must first understand the respective advantages and limitations of the two.

Start with 2D vision technology

2D technology started early and is relatively mature, and has proven to be very effective in a wide range of automation and product quality control processes over the past 30 years.

2D technology provides results based on the characteristics of contrast in grayscaness or color images. 2D is suitable for missing/presence detection, discrete object analysis, pattern alignment, barcode and optical character recognition (OCR), and a variety of two-dimensional geometric analysis based on edge detection for fitting lines, arcs, circles and their relationships (distance, angle, intersection, etc.).

Pattern matching: The key to handling part changes

2D vision technology is largely driven by contour-based pattern matching to identify the position, size and orientation of components. Technicians can use 2D to identify parts and create dynamically adapted part position, angle and size detection tools for robust measurement of part movement. Today's pattern matching handles masking, clutter, distortion, contrast reversal, overlapping parts and uneven lighting.

The shape could not be measured

The 2D sensor does not support shape-related measurements. For example, a 2D sensor is not capable of measuring features such as object plane, surface angle, partial volume, or distinguishing between objects of the same color, or between positions of objects with contact sides.

Vulnerable to variable lighting conditions

Because lighting determines edge position and measurement accuracy, lighting changes in the sensor's field of view can sometimes result in edge measurement errors unless specific techniques are used to compensate for this effect. Under factory conditions, lighting changes are a common challenge for unexpected environments or site sources on the production floor.

Limited contrast compensation

2D sensors rely on measuring the object's contrast (edge data), which means, for example, that they cannot measure black objects on a black background, or distinguish between partial features without specific lighting to expose the presence and definition of edges.

Sensitive to the motion of objects

Due to errors caused by component movement, 2D sensors need to be precisely fixed along the optical axis (Z-axis), detected using constant-scale features, or eliminated by large far-center optics that must match the FOV size.