Dec. 12, 2023
3D Imaging
A wide range of applications in industrial automation require and are enabled by 3D imaging. 3D imaging technologies are in demand over varied industries including aerospace, automotive, logistics and more. Components and software that deliver high quality 3D data have been undergoing an evolution of sorts in recent years and the result in many cases has been a noted advance in image quality, resolution and precision, ease of use, and greatly expanded reliable and robust use cases.
In this article we will discuss:
• the broad range of 3D imaging techniques and recent advancements in the technologies
• the growing trend towards application-specific solutions and the automation tasks they enable
• key considerations that may impact component specification
• some hints on best practices for 3D imaging application implementation
Key Technologies: 3d imaging
Warehouse/Logistics
3D cameras as a vision system are a fundamental part of pick-and-place robotic systems, equally important as the gripper and the robot itself. The cameras provide detailed images of the scene that is used to calculate object shapes and positions which allow for flexible automation of most type of objects, enabling this emerging market. Looking at the high-speed and demanding world of e-commerce, it has been at the vanguard of robotic automation for many years. Modern fulfillment centers are alive with automated transport of thousands, or millions of SKUs. While automation thrived in some areas, one stubborn problem has remained: reliable piece-picking of all these items, where especially shiny and transparent objects and wrappings have proved to be a major hurdle. Transparency has long been considered an impossible feat in 3D machine vision. Consequently, pick and place robot could only address ‘the easy to pick stuff’, a small portion of the items in a fulfilment center. But that was then.
Key Technologies: 3D cameras, logistics, bin picking
Hyperspectral Imaging
Looming European Union regulations mandate better sorting of waste, particularly plastics. Hyperspectral machine vision, aka industrial spectroscopy, makes it possible to sort plastics that appear indistinguishable to the eye. Another emerging application for hyperspectral technology involves food sorting and quality assessment. These and other uses are benefiting from falling prices, increasing performance, and improving technology. These trends, along with the push of regulations, are moving hyperspectral machine vision into the mainstream.
Key Technologies: hyperspectral imaging
Sub-pixel High Dynamic Range Imaging
High Dynamic Range (HDR) technology makes it possible to capture images with exceptional clarity, preserving details in bright and dark regions. However, the quest for advancements in this domain led to the emergence of sub-pixel HDR technology. This cutting-edge innovation takes HDR imaging to new heights, empowering sectors like Advanced Driver Assistance Systems (ADAS), autonomous mobile robots and autonomous vehicles. Sub-pixel HDR technology, also known as split-pixel HDR technology, plays a crucial role by allowing camera systems to capture and reproduce a broader range of brightness levels. Referred to as sub-pixel HDR, it involves dividing each pixel on the sensor into two sub-pixels to capture brighter and darker areas simultaneously. So, the highlighted pixel makes sure that details in the brighter areas of the scene are captured. The shadow pixel ensures that details in the darker areas are captured. This reduces motion blur and produces vibrant, lifelike images with accurate color representation.
In comparison, multi-exposure or traditional HDR leverages short and long exposures to capture a wide range of brightness levels (brightest highlights to darkest shadows, respectively) in a single image. Then, the multiple exposures are merged using algorithms, delivering a final HDR image. Hence, traditional HDR relies on multiple exposures and post-processing, making it more time-consuming and potentially introducing artifacts in high-motion scenarios.
How sub-pixel HDR technology works
Each pixel in the sensor is composed of two types of on-chip lenses (OCL): one large and one small. The smaller lens is situated within the space of the larger one. This design allows for one pixel that captures dark scenes with high sensitivity and another that resists oversaturation in bright scenes.
Key Technologies: HDR imaging
Download Media Planner