3D VISION


gdepends on several parameters (such as distance and reflectivity range, field-of-


view, light-power budget and so forth) to fit perfectly with application requirements. ‘Because of this, a deep knowledge is


required to build an efficient 3D vision system, or the help of a 3D vision integrator is needed. Teledyne e2v is able to support at all these levels to shorten a customer’s time-to-market and to get the best system to fit the application requirements. For that, we provide both laser triangulation and time-of- flight solutions, ranging from CMOS image sensors and customised camera modules, right up to full system integration support. Tis includes hardware and firmware development, light and optics assessment, eye safety assessment, application-level simulations, algorithms and factory calibration.’


What are the pitfalls and how do you avoid them? Kohli: ‘Different 3D modalities have different pitfalls. For example, 3D laser


triangulation faces challenges due to occlusion. Since the laser and the image sensor are mounted at an angle it is natural for the 3D profiler to encounter occlusion and shadowing. Te simplest and proven technique to overcome the occlusion is to use multiple image sensors to look at the object from the other side. Of course, doing so creates the additional challenge of synchronising the image sensors and then combining the two resulting images to create a corrected image. ‘Te other common problem with 3D


profilers is reflections. Although there is no single proven method of eliminating reflection, various techniques are used by 3D profilers to mitigate the effects of specular reflections. Tese include: changing the angle of the incident light, specialised optics, laser intensity management and filters to remove unwanted peaks. Te choice of method depends on the type of object surface, operating and performance requirements. ‘Laser speckles are present in every laser


Commercial products


One of Sick’s latest 3D vision products is the Ruler3000. Sick’s Fredrik Nilsson: ‘The Ruler3000 series is well suited for applications with moving objects that require high accuracy, even at very high transportation speed. With the different camera variants, we cover small fields of view, for example, in electronics production and small part assembly; the mid- sized models are aimed at consumer goods packaging and tyre manufacturing, whereas the models with large fields of view are aimed mainly at log inspection and logistics systems. In particular, the Ruler3000 excels in applications that need coverage of a large height range at high speeds as the


camera has the ability to use the full sensor image at 7,000 3D profiles per second. If less height range is needed, the speed increases proportionally to the reduction of sensor image in use. ‘The Ruler3000 is based on laser triangulation, but for applications without linear movement, other technologies may be more relevant. The main deciding factors are the frame rate and the resolution needed, as there is often a trade-off to be made between them. For example, a time-of- flight system can run at high frame rates but has limitations in the height accuracy, whereas structured light sensors can achieve really good accuracy at the expense of frame rate. It should not come as a big surprise, but you need to know your application and its specific requirements really well before you decide what technology to choose – this is true regardless of 2D or 3D vision technology.’ Teledyne e2v supplies time-


Sick’s Ruler3000 and Ranger3 cameras


of-flight and laser triangulation sensors, while Teledyne Dalsa


Teledyne Dalsa’s Z-Trak2 laser profiler


offers its Z-Trak2 laser profiler. Teledyne Dalsa’s Inder Kohli: ‘Z-Trak2 comes in a wide variety of configurations and laser options to handle a range of parts, from small electronic components to large automobile engine parts, and door frames to entire chassis. Applications that require height, width, length and volume information of moving parts at a close range are ideally suitable for Z-Trak2. The exact model of Z-Trak2 depends on the size and surface properties of the target object, scanning speed, accuracy and precision of the measurements required. ‘Z-Trak2 laser profilers offer speed, accuracy, ease-of-use


and help reduce the total cost of ownership by ensuring that systems can be built and maintained with standard, off-the-shelf networking parts, making it the right choice for in-line 3D measurement, inspection, identification and guidance applications.’ Photoneo’s 3D camera,


MotionCam-3D, can provide a lateral resolution of 2 megapixels with a depth accuracy of 50 to 900μm across the different models, while being able to capture objects moving up to 144 km/ hour. Photoneo’s Svorad Stolc: ‘We reduce the complexity of 3D vision by improving hardware robustness and aim to provide flexible enough APIs and tools for our users. Our sensors feature a highly durable carbon body and they are IP65 rated.’


because the laser light is coherent. Tis interference phenomenon is caused by microscopic irregularities of the surface, creating an interference pattern. Tis pattern manifests itself by making parts of the laser line appear brighter or darker along its length. Such variations limit the uniformity of the laser line limiting the 3D sensor’s achievable accuracy. Several techniques, such as the use of a laser with a shorter wavelength, optics with a bigger aperture and profile averaging can alleviate the effects of laser speckles.’ Nilsson: ‘Other challenges are the


unwanted secondary reflections one may get on shiny metal surfaces. Here, a polarising filter in front of the lens can be used to filter out the second order of reflections.’ O


Inder Kohli is senior product manager, vision solutions at Teledyne Dalsa; Fredrik Nilsson is head of business unit machine vision at Sick IVP; Yoann Lochardet is marketing manager, 3D at Teledyne e2v; and Svorad Stolc is CTO of 3D sensing at Photoneo.


Photoneo’s MotionCam-3D


26 IMAGING AND MACHINE VISION EUROPE AUGUST/SEPTEMBER 2021


@imveurope | www.imveurope.com


Page 1  |  Page 2  |  Page 3  |  Page 4  |  Page 5  |  Page 6  |  Page 7  |  Page 8  |  Page 9  |  Page 10  |  Page 11  |  Page 12  |  Page 13  |  Page 14  |  Page 15  |  Page 16  |  Page 17  |  Page 18  |  Page 19  |  Page 20  |  Page 21  |  Page 22  |  Page 23  |  Page 24  |  Page 25  |  Page 26  |  Page 27  |  Page 28  |  Page 29  |  Page 30  |  Page 31  |  Page 32  |  Page 33  |  Page 34  |  Page 35  |  Page 36  |  Page 37  |  Page 38  |  Page 39  |  Page 40  |  Page 41  |  Page 42  |  Page 43  |  Page 44