Computer Vision
Human sight is tuned to a lifetime of contextualized visual data that we continue to accumulate as we age. This allows us to tell images and videos of objects and people apart from each other, and how we detect movements or errors in the things we see.
With computer vision, engineers are trying to replicate the same ability of ‘sight’ in computers. It is a field of artificial intelligence (AI) that seeks to understand and automate tasks that the human visual system can do. By enabling computers and systems to derive meaningful data from visual inputs - such as digital images and videos- machines can take or recommend actions based on the information they are given.
To perform these functions, machines are fed vast quantities of data and trained to continuously analyze the data until they are able to discern and contextualize images.
Computer vision can surpass human capabilities as they are able to analyze thousands of products or processes a minute. More importantly, it can notice defects or issues imperceptible to the human eye.
Computer vision is growing popular in various industries, from automotive and manufacturing to energy, utilities, and logistics. By 2022, the computer vision market is expected to be worth USD 48.6 billion.
To improve operational efficiency, DHL partnered up with Metrilus, a software startup with deep expertise in 3D computer vision to develop a low-cost, automated dimensioning solution that maximizes quality and productivity in warehousing operations.
The DHL team also recently launched a drone dimension project that utilizes high-resolution 4D cameras and advanced highly accurate software algorithms tailored to the needs of each customer.
English