Bluewrist Insight
3D vs 2D Inspection
3D automated inspection system
History of Vision Inspection in Manufacturing
The use of 2D machine vision to provide image-based automatic quality control in manufacturing process was first tested four decades ago. Since the mass adoption of the technology in the 1980s, it has played an increasingly important role in changing the way manufacturers make, inspect, and quality-check products. Overtime manufacturing has become increasingly accurate, and products are made with ever-tighter tolerances, so manufacturers have taken advantage of 2D machine vision to perform various tasks with a high level of accuracy, including positioning, inspection, measuring, and reading tasks until 3D automated inspection system offers better.
While 2D machine vision has contributed significantly to the evolution of the manufacturing sector, 3D machine vision represents the next step in what is possible. However, when it comes to implementing machine-vision solutions, there are important distinctions to make when considering whether to install 2D or 3D automated inspection system.
The Limitation of 2D Vision Automated Inspection
A pure 2D machine uses 2D cameras to provide traditional vision inspection and robot guidance. A 2D industrial camera is similar to the typical smartphone camera that outputs an image file in various common formats, such as JPEG, TIFF, or PNG. This method assumes that the target-object image acquired for processing is a flat, two-dimensional object. 2D machine vision is unable to provide any height information, so it is excellent for processing X-axis and Y-axis data but offers no depth-of-field (Z-axis) data.
Machine vision forms images of objects using the light reflected from the object. Therefore, even slight variations in lighting in the field of view due to changes in ambient conditions or artificial lighting can have an adverse impact on accuracy. Too much light, too little light, or shadowing in the factory environment can adversely affect the clarity of edges and features appearing in the 2D plane.
Without Z-axis data, the acquired image lacks the volumetric information that is critical in determining dimensional measurements of complex parts. Therefore, 2D machine vision systems are unable to cope with the intricacies of three-dimensional shapes or forms.
This issue is critical when it comes to complex parts and assemblies where dimensions need to be measured beyond the X and Y axes. Examples include components where volume needs to be determined and parts that need to be picked and positioned in a precise fashion. Simply put, 2D machine vision just isn’t up to the task because it is unable to recognize shapes and volume.
2D Vision has commonly been used for
- 2D surface-area checking
- Barcode reading
- Character recognition
- Label verification
- Basic quality inspection
Benefits of 3D Vision Inspection
By comparison, a 3D machine vision system simultaneously provides X-, Y-, and Z-axis data, along with the respective rotational information around each of the axes. This system understands each part as a three-dimensional point cloud of precise coordinates where the position of every pixel in space is known.
With the ability to capture third dimension data reliably, 3D machine vision systems are immune to the environmental factors that adversely affect 2D systems, such as lighting, contrast, and distance to the object. 3D cameras use either laser or structured light to capture and produce a 3D picture of the object in a point cloud format, offering an accurate 3D representation of the physical object with highly detailed image resolutions of up to 50 microns. This file size is much larger, which reflects the vast amount of data contained in a single image.
While manufacturers have avoided using 3D cameras in the past because of the hardware and software processing time involved (which may have slowed the overall production line), modern systems have generally overcome this issue.
3D systems are also more complex because when they are deployed to assess very large parts such as the entire frame of a car, it may be necessary to set up multiple cameras. A car body may need 60 separate 3D cameras or fewer robot mounted cameras to cover the inspection area. This need introduces several technical challenges, such as the industrial communication of robots, PLCs, cameras, and host systems, as well as overcoming the inaccuracies of the robot motion, which may distort the quality of the data collected.
Vision sensors are often paired with industrial robots on the plant floor to complete different tasks. These robots work in a three-dimensional world. While a “blind” robot is limited to performing repetitive and structured tasks, 3D machine vision systems allow robots to sense variations in the physical environment and adapt accordingly, increasing flexibility, utility, and velocity.
3D vision has pushed the envelope of what is possible. Pick-and-place applications have been improved utilizing 3D vision systems, and random bin picking has been made both possible and highly efficient.
How to Choose the Right 3D Automated Inspection System
Now that we have examined both 2D and 3D vision systems, let’s look at how to determine which solution would benefit you. Some of the factors to consider, all of which are job-specific, include the complexity of the part and the dimensions that need to be accounted for (X, Y, and Z).
The main difference between 2D and 3D systems from a technological perspective relates to the imaging chain that is used to capture images for analysis. While 3D systems were once too slow to keep up with production line speeds, that is not the case anymore.
Additionally, the pricing of 2D and 3D systems are much closer than they used to be, with the premium paid for the 3D vision system being amply justified by the capabilities of the technology and the benefits to the overall production process.
3D Vision Inspection System From Bluewrist is Highly Scalable
