While 2D machine vision has contributed significantly to the evolution of the manufacturing sector, 3D 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 vision sensors.
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
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.
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.
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.
Our solutions have helped us overcome all the challenges discussed in this article and have firmly positioned us as industry leaders. Through our intelligent algorithm, our software takes advantage of the latest multi-core CPUs and GPUs to reduce processing time from minutes to less than three seconds.
Bluewrist industrial communications software comXtream also enables the seamless connection of hundreds of cameras, industrial robots, and other devices through a standard TCP/IP protocol, greatly simplifying the setup and configuration requirements for end-users.
3D vision can offer companies in the manufacturing sector a significant competitive and revenue advantage through productivity increase and defect detection and prevention.