Color imaging has grown in popularity in recent years. The demand for machine vision technologies that can provide color information has risen exponentially with the growing need for automation and intensive exploration of new areas for robot adoption.
The ability to recognize color is crucial for many robotics applications, such as the inspection and classification of defects characterized by a specific color, sorting of products based on their color properties, or various color-detection applications.
Photoneo, a Slovakia-based company that develops innovative 3D machine vision technologies and smart automation, advanced robot vision to the next level when it introduced a technology that unlocks new application areas and improves existing ones. It credited an unparalleled combination of color information, 3D data, and the ability to capture images in motion.
Photoneo combines 3D, motion, and color
Each of these three properties plays a key role in machine vision applications. As already stated, color information is crucial for many inspection tasks or color-based recognition, decision-making, and sorting applications.
3D data is fundamental to any advanced automation system that needs to navigate a robot in a three-dimensional space defined by X, Y, and Z coordinates. It is mostly used in robotic object handling, bin picking, and pick-and-place applications, palletization and depalletization, packaging, inspection and quality control, robotic assembly, 3D model creation, or reverse engineering.
The element of motion has been a difficult challenge to tackle. Traditional 3D area-scanning technologies have struggled with image capture in motion, forcing customers to compromise between the quality of 3D data and the speed of scan acquisition.
Such a trade-off limits applications dependent on high-quality 3D data to static scenes. For instance, a robot with a 3D scanner attached to its arm needs to stop moving before the scanner makes a scan. This has a tremendous effect on cycle times.
In 2020, Photoneo introduced the MotionCam-3D camera using its Parallel Structured Light technology. This approach unlocked new applications by providing two features that had never been successfully combined before—the ability to provide high-quality 3D data and the ability to scan in motion.
But one last piece to the ultimate 3D vision puzzle was still missing—color.
3D data, color, and the ability to scan in motion can all be found in a number of technologies, but the had not been combined in a single device.
“We believe that a perfect vision system needs to have 3 core properties,” said Marcel Svec, vice president of strategy and product at Photoneo. “First, it must provide high-quality 3D data, which is important for the recognition of an object’s position and orientation in 3D space.”
“Second, the vision system should be able to work in a dynamic environment and provide a continuous stream of 3D point clouds without making any pauses for scan acquisition,” he added. “Third, the vision system should be able to capture color information and match it with the 3D output. This is a crucial feature for AI-based machine vision algorithms.”
Therefore, in September 2022, Photoneo introduced a new version of its black-and-white camera upgraded with color—MotionCam-3D Color. With the new camera, customers can improve their existing processes or explore entirely new opportunities. Below is just a fragment of them.
MotionCam-3D Color enables robot applications
The ability to capture dynamic scenes and create colorful 3D point clouds in real time and perfect quality is critical for artificial intelligence applications that are dependent on high-quality data.
An example is robotic decision-making where a robot recognizes, picks, and sorts products based on their 3D geometry and color features. Such a system can be deployed in the food and beverage industry to determine if a product is burnt or undercooked, or in agriculture to determine the ripeness of fruits and vegetables.
Other applications that can greatly benefit from color information and 3D data are inspection and quality control. The largest model of MotionCam-3D Color provides a scanning range of up to 4 m (13.1 ft.), so it can be used even for the inspection of very large objects such as cars, ships, or huge containers.
For instance, car rental companies thoroughly check each returned vehicle to make sure that no damage occurred to it during the rental period. Automated inspection is faster, more efficient, and more accurate and reliable than manual examination.
MotionCam-3D Color enables synchronized area scanning from multiple perspectives to examine the complete undercarriage through instant meshing. This approach does not require the use of marker patterns for scan alignment. The camera maps 3D data with color information and identifies surface defects such as scratches, chips, or paintwork damage.
Another type of application relying on 3D data and color of objects in motion is robotic painting or sandblasting of large objects such as transportation containers. A robot equipped with MotionCam-3D Color can inspect a container and recognize any 3D surface deformation and colored defects such as rust.
Based on the results, the robot can then move to the next steps. They may include rust and paint removal, blast cleaning (e.g. sandblasting or water-jet blasting), pressure washing, priming, or painting.
Thanks to Parallel Structured Light enabling 3D scanning in motion and the PhoXi 3D Instant Meshing for instant 3D model creation from all perspectives, the objects can be rotating or randomly moving in front of the robot without stopping and compromising the quality of 3D data. This can lead to significant reduction of cycle times, higher productivity, and improved efficiency.
Mobile robots show new capabilities
The recognition of color combined with 3D data in motion is also important forn many robotics applications in the logistics, pharmaceutical, and medical sectors. This level of perception opens up new areas for mobile robots.
Visitors to last month's VISION show in Stuttgart had a unique opportunity to see Boston Dynamics’ Spot equipped with MotionCam-3D Color.
The four-legged robot moved deliberately in the crowd, demonstrating the tremendous potential of these combined technologies.
Besides robotics, Photoneo’s technology also opens new applications for 3D model creation in the Metaverse, digital twins, or teleoperation of industrial applications.
About the author
Andrea Pufflerova is public relations specialist at Photoneo.
Article topics
Email Sign Up
Related Machine Vision News
Related Companies
