28 n MACHINE VISION March 2024 www.drivesncontrols.com 3D cameras will drive the global machine vision market over the coming ve years, fuelled by strong growth in mobile robot and robotic picking applications. The predicted CAGR for 3D cameras of 13% in the period to 2028 is much higher than the 6.4% CAGR expected for the global machine vision market as a whole (see Fig. 1). Revenue for 3D machine vision cameras is forecast to grow from $767m in 2022 to almost $1.6bn by 2028, with particularly strong growth expected for time-of-ight and stereo-vision cameras. Interact Analysis recently published the rst edition of a report* on the global machine vision market, which nds that it generated revenues of $6.2bn in 2023 – a 2.8% decline from 2022. Despite this slight contraction, we expect a steady growth rate of 6.4% over the period to 2028, starting with a modest growth of 1.4% this year. Four technologies 3D cameras can be divided into four main product types, each of which has key features and advantages for dierent applications. Structured light 3D cameras project a known pattern or sequence of light onto a surface and analyse the deformation or distortion of this pattern when it interacts with an object. The camera observes how the structured light is deformed and, from this, can calculate the depth and shape of objects in the scene. These cameras are most commonly used when precise measurements and image acquisition are required – such as in bin-picking applications. Structured light 3D cameras are often more expensive than other types of 3D camera. One example of a camera of this type is the Norwegian Zivid 2+ device. Stereo-vision cameras are equipped with a pair of cameras that perceive depth through binocular disparity. The cameras capture two slightly oset images of the same scene. The disparity between corresponding points in the images is used to calculate depth information for objects in the scene. These cameras are most typically used in robotics and are particularly useful for autonomous driving – an application which oers signi cant growth potential. One example is Basler’s stereo cameras. Time-of-ight 3D cameras are imaging devices that determine the distance to objects in a scene by measuring the time it takes for light to travel from the camera to the object and back again. These cameras are typically used when high speed, but lower quality, image acquisition is needed. They are also a cheaper option for mobile robots, enabling them to avoid obstacles and to navigate around other robots. One example of a camera of this type is Lucid’s Helios 2. Laser triangulation 3D cameras use lasers to measure distances and create threedimensional representations of objects or scenes. The lasers project a line or pattern onto the target surface, and the camera observes the deformation or displacement of this line/pattern as it interacts with the object. The information captured is then processed to determine the depth or threedimensional structure of the object. These cameras oer high accuracy and resolution, and are typically used for quality inspection, although they also can also be used to guide mobile robots. An example is LMI’s ChromaScan system. Fig. 2 show expected changes in market share for each type of product in the 3D camera market. What is particularly interesting is the projected growth for stereo-vision and time-of-flight cameras. Although they currently hold just 3% and 2% market shares respectively, this is considerable because these systems are The global market for 3D machine vision systems is growing more than twice as fast as the total vision market. Jonathan Sparkes, a research analyst with Interact Analysis who specialises in machine vision, examines the reasons and the technologies involved. Fig.1: Revenues and growth of global 3D camera market 2022-2028 Source: Interact Analysis Fig. 2: Market shares for 3D camera types in 2023 (left) and the predicted shares for 2028 (right) Source: Interact Analysis 3D cameras are driving machine vision growth
RkJQdWJsaXNoZXIy MjQ0NzM=