Computer Vision

Trend Overview

Impact:

Significant

Adoption:

< 5 Years

Focus Areas:

Automation & Efficiency Health, Safety & Future of Work Physical & Digital Security Visibility & Resilience

Trend Clusters:

Artificial Intelligence Robotics

Sectors:

The trend of Computer Vision utilizes cameras to capture photos or videos and applies artificial intelligence (AI) algorithms to analyze data extracted from this digital imagery. Rudimentary visual AI systems are trained to simply differentiate objects from each other, while more advanced – increasingly AI-enabled – versions can track objects across viewpoints and learn on their own and, in recent developments, enable prediction through pattern recognition.

Enabled by AI, the trend of Computer Vision has developed in conjunction with advances in deep machine learning (ML), leveraging the rising quality and decreasing cost of camera devices.

Several things are driving the adoption of computer vision technology, including the growing need for workflow automation and optimization across many industries, from auto-mobility to healthcare and banking, financial services, and insurance to retail.

In 2023, the global computer vision market was valued at around US$ 17.7 billion and – with continual improvements in AI, vision systems, and computer processing – is expected to grow at a compound annual growth rate (CAGR) of 19.6% by 2026.

Today, advanced computer vision technology is perfecting depth perception, 3D reconstruction, and dark and blurred image interpretation, all of which will unlock more opportunities in supply chains. Future deployment will be driven by further implementation of AI, automated ML, edge computing, the Internet of Things (IoT), and more.

Computer vision will become commonly utilized in logistics operations within the next five years, and many new use cases are likely to emerge. This technology will underpin and drive future logistics success by enabling more automated and efficient processes as well as sustainable and safe operations.

However, to be fully realized, this trend will require additional investment. As experienced in the early days of sensor adoption, computer vision applications must be scalable for logistics organizations to maximize benefits.

Classification & Segmentation of Computer Vision

Source: Packt Publishing (2021): Computer Vision

Relevance to the Future of Logistics

People: Operations, Health & Safety

Computer vision and AI technologies offer effective solutions for enhancing workplace safety in logistics facilities by monitoring and analyzing movements of people and vehicles in real time. They enable proactive safety decisions, workflow improvements, and prompt corrective actions to mitigate risk and ensure a safer working environment.

In addition, computer vision used in human pose estimation (HPE) enables assessment of ergonomics in the workplace. TuMeke, a tech company, has developed an AI ergonomic risk assessment platform using computer vision that enables users to record and analyze smartphone videos of tasks like lifting boxes in warehouses to identify unsafe postures and enhance workplace safety.

The workforce should not only use appropriate personal protective equipment (PPE) but also wear it correctly. Using computer vision technology, AI systems can track and ensure compliance with PPE safety-at-work protocols, reducing incident risk.

Bottlenecks, inefficiencies, and activity hotspots can be revealed by computer vision-based heatmaps in warehouses and yards, informed by surveillance camera feeds. With this information, managers can optimize inventory placement and increase operational efficiency. Computer vision systems also offer a solution for efficiently counting people and vehicles in logistics facilities, helping managers optimize staffing levels and ensure compliance with safe occupancy requirements. Startups like AIVID provide automated detection and counting capabilities along with demographic insights.

To optimize the pick path, computer vision can be used alongside ML algorithms. Managers can analyze camera feeds to identify patterns, trends, and potential workflow adjustments to increase efficiency, save costs, and enhance overall warehouse operations.

Additionally, intelligent surveillance systems equipped with computer vision technology can analyze real-time video footage using advanced algorithms to enhance safety and security. They can detect unauthorized entry or suspicious behavior, generate alerts for prompt response and proactive intervention, and ultimately reduce theft and increase overall safety.

Next use case

Shipments: Measurement, identification & inspection

Accurate measurement and efficient processing of packages are essential to storing, handling, and transporting goods. The task of dimensioning can be automated using computer vision technology – examples include solutions from MetriXFreight and Qboid.

Computer vision systems can also make valuable contributions to quality inspection, especially as early detection of damage during shipment protects brand image and ensures consumer satisfaction. Items must be correctly and legibly labeled, and computer vision technology helps ensure label accuracy and integrity, identifying errors and anomalies to facilitate efficient resolution and compliance with legal requirements. Startups like Visionify offer solutions for label checking in various shipping scenarios. To further streamline product identification, companies like Banner and Zetes offer innovative solutions for decoding challenging barcodes and ensuring shipping and loading accuracy.

In addition, automated sorting systems utilizing AI and neural networks streamline the visual classification of shipments, enhancing efficiency and accuracy. Good examples are Photoneo singulation and sorting solutions which can recognize parcels with 95% accuracy and high throughput rates using 3D scanners and AI algorithms.

Previous use case Next use case

Assets

To protect logistics assets, computer vision technology can be used for predictive maintenance, issuing alerts so that the maintenance team can intervene before any issue arises. This also allows managers to schedule repairs and upkeep to prolong asset life and prevent failure. Combining computer vision with human observation helps capture the full picture, as asset performance may be impacted by qualitative factors, such as the way in which a worker interacts with a device.

When planning capacity to optimize asset utilization, computer vision can be implemented in asset management to deliver quicker insights than the human eye and those from human experience. For example, this AI-based technology can assess the overall space inside trucks and containers to calculate available volume prior to loading – information that helps determine the optimal arrangement of items to maximize loads and minimize wasted space. In the warehouse, computer vision can be used to analyze the dimensions and orientation of pallets and roller cages, helping ensure these assets are positioned for optimal load distribution and peak efficiency. Meanwhile, assets outside the warehouse can be monitored 24/7 by an integrated system combining computer vision with surveillance. To restrict yard access to registered vehicles only, cameras can identify each truck and log its entry time, exit time, and number of daily trips. The system can also measure asset usage patterns, including idle time, and use this data to help optimize fleet operations.

Previous use case

Challenges

If workplace surveillance methods are perceived as invasive, this may lower employee morale, increase work-related stress, and cause counterproductive work behaviors.

Cameras may not see everything; for example, if the field of view is obstructed during capacity assessment, perhaps by a stack of goods waiting to be packed, a workflow change may be needed.

Environmental factors such as poor lighting, shadows, and reflection may challenge the capabilities of computer vision, in which case a manual process could also be needed.

If workplace surveillance methods are perceived as invasive, this may lower employee morale, increase work-related stress, and cause counterproductive work behaviors.

Cameras may not see everything; for example, if the field of view is obstructed during capacity assessment, perhaps by a stack of goods waiting to be packed, a workflow change may be needed.

Environmental factors such as poor lighting, shadows, and reflection may challenge the capabilities of computer vision, in which case a manual process could also be needed.

Challenge 1 Challenge 2 Challenge 3

Outlook

Computer vision technology has matured significantly in the last two years, leading to remarkable growth in implementation and expanding applications in logistics operations. Increased investment in AI is anticipated to further advance computer vision solutions, with AI serving as the backbone for image recognition.

While computer vision has proved its value in dimensioning, health and safety, and other applications, challenges remain. These include technology acceptance, data privacy compliance, cybersecurity, hardware upgrades, and material identification. Despite these hurdles, computer vision remains integral to the automation and digitalization of logistics for more efficient and sustainable operations.

This trend should be ACTIVELY monitored,with implementations available for many use cases today.

Resources

Need support prioritizing high impact trends for your business?

Request a complimentary Trend Radar Mapping session at your regional DHL Innovation Center and prepare for the future today.

Request a Session

Stay Connected

Yes, sign me up!

Sources

DHL (2023): DHL – Computer Vision Health & Safety Proof-of-Concept with Protex AI
Global Data (2023): Computer Vision Market Size, Share, Trends and Analysis by Region, Industry Vertical and Segment Forecasts to 2026

Trends Filter

Focus Areas

Trend Clusters

Sectors

Trend Name