Chapter 10: Evolution of Machine Identification Technologies |
Summary in Brief |
For decades, the barcode was the workhorse of identification---a simple, reliable, and inexpensive way to label products, assets, and documents. But the world has outgrown the barcode. Today, machine identification technologies are evolving into intelligent data networks that do far more than just identify an item. Radio Frequency Identification (RFID) tags, combined with AI-powered analytics, IoT connectivity, and edge computing, are transforming simple labels into smart nodes that communicate location, condition, and context in real time. This chapter explores the evolution from passive identification to active intelligence. We will examine how American companies like Zebra Technologies and Datalogic are embedding AI into scanners and RFID systems, and how Chinese companies like Weibiao Technology are solving real-world problems in airports, hospitals, and logistics. The focus is on practical applications: how these technologies are reducing lost baggage, speeding up hospital lab work, enabling self-checkout, and providing end-to-end visibility across supply chains. The central theme is clear: machine identification is no longer about reading a label; it is about creating a digital thread that connects physical objects to intelligent systems. |
|
Introduction: The Barcode's Legacy and Its Limits |
In 1974, the first barcode was scanned on a pack of Wrigley's chewing gum at a Marsh supermarket in Ohio. That simple act marked the beginning of a revolution in retail and logistics. The barcode---a pattern of black bars and white spaces---encoded a product's identity in a format that could be read quickly and reliably by a laser scanner. It was a brilliant solution to a pressing problem: how to track inventory without manual data entry. |
The barcode's success was built on simplicity. It was cheap to print, easy to read, and robust enough for most environments. It enabled the modern supply chain, transforming retail, manufacturing, and logistics. Today, billions of barcodes are scanned every day, from grocery checkout lines to warehouse receiving docks. |
But the barcode has fundamental limitations. As the table below illustrates, barcodes require a direct line of sight to be read, they can only be read one at a time, they cannot be read if dirty or damaged, and the information they contain is fixed and cannot be updated . A barcode tells you what type of item you have, but not which specific item it is. It is a one-way communication: you read it, and that is it. |
| Feature | Barcodes | RFID Tags | |
||-|--| |
| Line of sight required | Yes | No | |
| Read individually | Yes | Multiple simultaneously | |
| Read if dirty/damaged | No | Yes | |
| Read if concealed | No | Yes | |
| Identifies type of item | Yes | Yes, and specific item | |
| Information can be updated | No | Yes | |
| Tracking method | Manual | Automatic | |
The limitations become acute in modern, complex supply chains. A warehouse might need to scan thousands of items per hour; a single-item scan is a bottleneck. A barcode on a package can become smudged or torn, making it unreadable. And once the barcode is printed, the information is fixed---you cannot update it to reflect that the item was inspected or that it changed hands. |
This is where the next generation of machine identification technologies comes in. Automatic Identification and Data Capture (AIDC) is the broad category that encompasses barcodes, RFID, biometrics, optical character recognition, and near-field communication . The goal of AIDC is to identify objects, collect data about them, and enter that data directly into computer systems without human intervention . The evolution of AIDC has been from simple identification to intelligent data networks. |
The evolution has been driven by a convergence of technologies. The Internet of Things (IoT) has connected devices and enabled real-time data sharing. Artificial intelligence has enabled pattern recognition, anomaly detection, and predictive analytics. Edge computing has brought processing power closer to the point of data capture. Together, these technologies are transforming machine identification from a passive label-reading exercise into an active, intelligent data-gathering process. Think of RFID as the eyes and IoT as the brain: RFID tags identify an asset, while IoT devices process that information and share it through a cloud platform . |
This chapter explores how this evolution is unfolding. We will begin by tracing the history of machine identification, from the barcode to RFID and beyond. We will then look at how American and Chinese companies are deploying these technologies in real-world applications, from retail and logistics to healthcare and manufacturing. We will also examine the role of AI in enhancing identification, and consider the future of a world where every object has a digital identity. |
|
Part One: A Brief History of Machine Identification |
The history of machine identification is a story of solving one problem, then discovering the next. |
The Dawn of the Barcode |
The barcode was invented in the late 1940s by Bernard Silver and N. Joe Woodland, who were inspired by Morse code. They created a pattern of concentric circles that could be read from any angle. It took decades for the technology to become practical, requiring lasers and microprocessors. The first commercial barcode scanner was installed in 1974, and the Universal Product Code (UPC) was adopted as the standard for retail. The barcode quickly became ubiquitous, spreading from retail to manufacturing, logistics, and healthcare. |
The barcode is a 1D technology: it encodes data horizontally in the width of bars and spaces. It is limited to about 20 characters of information, which is sufficient for a product code but not much more . The 2D barcode, which encodes data both horizontally and vertically, was introduced later, enabling storage of up to 7,089 characters. Examples include QR codes and Data Matrix codes, which are used in applications ranging from advertising to industrial traceability . |
The Rise of RFID |
Radio Frequency Identification (RFID) was developed in the 1970s and 1980s. Early applications included electronic article surveillance (EAS) systems in retail stores to detect theft, and identification friend-or-foe (IFF) systems in aviation . But it was not until the 2000s that RFID became commercially viable for broader applications. This was driven by improvements in functionality, decreases in size and cost, and agreements on communication standards . |
An RFID system consists of three components: a tag (or transponder), a reader, and an antenna. The reader emits radio waves that activate the tag, which then transmits its stored information back to the reader . Unlike barcodes, RFID does not require line of sight. Tags can be read from several meters away, through packaging, and in harsh environments. Multiple tags can be read simultaneously, and the information on a tag can be overwritten repeatedly . |
RFID tags come in two main types. Passive tags are powered by the reader's electromagnetic field; they are cheap and have a shorter read range. Active tags have their own power source (a battery) and can transmit over longer distances . The choice depends on the application: passive tags are suitable for item-level tracking in retail, while active tags are used for tracking high-value assets over large areas. |
The Convergence with IoT and AI |
The next major shift came with the convergence of RFID with the Internet of Things and artificial intelligence. IoT provides the network that connects RFID readers to central systems, enabling real-time data sharing and remote monitoring . AI provides the intelligence to analyze the data, detect patterns, and generate insights. For example, an RFID system in a factory can track the location of parts, and an AI system can analyze the data to identify bottlenecks or predict when a machine will need maintenance . |
The convergence is creating what are sometimes called 'intelligent data networks.' These are not just passive identification systems; they are active, adaptive systems that provide real-time visibility, predictive analytics, and automated decision-making. The RFID tag is no longer just a label; it is a sensor node in a broader digital ecosystem. |
|
Part Two: American Companies Leading the Evolution |
American companies are at the forefront of the evolution from simple identification to intelligent data networks. They are embedding AI, IoT, and analytics into scanners, mobile computers, and RFID systems, transforming how businesses track assets, manage inventory, and serve customers. |
Zebra Technologies: From Scanning to Frontline Intelligence |
Zebra Technologies is a global leader in digitizing and automating workflows. The company provides a comprehensive portfolio of hardware, software, and services for asset visibility, data capture, and frontline automation. Zebra's footprint is vast: it operates in more than 100 countries, serves more than 80 percent of Fortune 500 companies, and has made strategic acquisitions to build a complete end-to-end solution . |
Zebra's evolution illustrates the broader shift in the industry. The company started with barcode printing and scanning, but has expanded through acquisitions into machine vision, AI-driven software, robotics, and workforce management. The vision is to provide not just tools for data capture, but integrated solutions that digitize and automate entire workflows . |
At the IOTE 2025 exhibition in China, Zebra showcased four core application scenarios: an AIoT smart factory, a smart cloud warehouse, smart retail, and a smart healthcare platform . The demonstrations highlighted Zebra's two key technologies: RFID and machine vision. The RFID technology enables item-level tracking, inventory management, and loss prevention. The machine vision technology, powered by AI, enables automated quality inspection, optical character recognition (OCR), and anomaly detection. |
Zebra's FS42 fixed industrial scanner is a prime example of how barcode technology is evolving. The scanner provides 'excellent barcode reading capabilities on high-speed production lines,' enabling information traceability and improving production efficiency . This is not a traditional barcode scanner; it is a high-speed, AI-enhanced device that can read codes on moving products in challenging industrial environments. |
Even more advanced is Zebra's NS42 smart vision sensor, which is driven by AI and designed for automated inspection in complex industrial scenarios. The NS42 supports deep learning-based OCR and anomaly detection tools, and it can achieve 'faster and more accurate detection' without requiring manual labeling of defect samples . This demonstrates the convergence of machine identification and machine vision: the scanner reads the barcode, but the vision system checks the product's quality. |
In the logistics domain, Zebra has developed a wireless forklift scanning solution that integrates fixed scanning and RFID technology to automate the collection of pallet data, reducing the risk of manual errors . The AI-driven RFID smart inventory cart is capable of batch identification of assets, with real-time visualization and integration with management systems . This is a far cry from the manual barcode scanning that was standard just a decade ago. |
Zebra's retail solutions are also integrating AI and RFID. The Modern Store management solution uses powerful RFID technology to enable multi-process tracking of goods, significantly improving the efficiency of receiving, inventory, and checkout processes, and optimizing the customer experience . The FXP20 POS RFID reader, designed for retail environments, can be deployed at checkout counters and other retail control points, supporting both assisted and self-checkout . |
Zebra's latest mobile computers, the TC501 and TC701, are described as 'AI-native' devices . They feature the industry's first AMOLED display and the industry's first scan engine with color imaging. They are equipped with Qualcomm processors and support Wi-Fi 7 and 5G connectivity. These devices also integrate UHF RFID and Impinj Gen2x extensions for 'advanced visibility and security of critical data' . |
The intelligence in these devices is powered by Zebra's Frontline AI Suite, which consists of three key components. First, Frontline AI Enablers are on-device AI models and APIs that allow developers to build enhanced applications. By processing data at the point of capture, these tools reduce latency, enhance accuracy, and support faster decision-making. Second, Frontline AI Blueprints combine enablers into purpose-built templates that automate multi-step workflows, such as shelf merchandising and proof of delivery. Third, Zebra Companion provides GenAI-driven agents that empower frontline associates with on-demand knowledge and sales coaching . |
Zebra's approach is to embed AI into the hardware itself, creating 'intelligent' devices that do not just capture data but also analyze it and act on it. This is a fundamental shift from the passive scanner to the active, intelligent node. |
Datalogic: Embedded AI for the Intelligent Store |
Datalogic is another global leader in automatic data capture and process automation, with a strong focus on retail. At the NRF 2026 and EuroShop 2026 trade shows, Datalogic showcased how its integrated portfolio of scanners, mobile computers, RFID, and IoT solutions is enabling 'the intelligent store' . |
Datalogic's embedded AI approach is a significant differentiator. The company's Magellan 9600i and 9900i fixed retail scanners are 'the world's first and only devices with embedded AI Smart Vision for Retail software solution' capable of real-time detection of self-checkout issues . These issues include mis-scans (scanning a cheaper item instead of a more expensive one), stacked items (two items scanned as one), label switching, and produce recognition. |
The key innovation is that the AI is embedded directly in the scanner, not in an external camera or cloud-based system. 'Unlike existing solutions that rely on external cameras and cloud or edge external processors and software, Datalogic's integrated design combines high-resolution computer vision and AI directly in the device' . This reduces costs, simplifies installation, and enables real-time, on-device decision-making. |
Datalogic also showcased the Joya Smart and Joya Smart+, which the company describes as 'the industry's first self-shopping devices with integrated AI technology' . These handheld devices allow shoppers to scan items as they move through the store, monitor their basket in real time, and complete payment quickly at self-service kiosks . By combining customer autonomy with intelligent loss prevention capabilities, the Joya Smart addresses retailers' most critical operational concerns . |
The integration of barcode and RFID on a single device is another important trend. Datalogic's PowerScan 9600 RFID solution captures data from both barcodes and RFID tags, providing real-time movement visibility that supports loss prevention and inventory accuracy . This is 'ideal for item-level tracking in sensitive areas such as checkout lanes, exits, and back rooms' . |
Datalogic's Gryphon 4600 handheld scanner demonstrates how AI is also improving basic scanning. It is 'powered by AI neural decoding,' delivering faster and more accurate reads while consuming less energy . The scanner is also designed with sustainability in mind, using recycled and recyclable materials. This shows that the evolution of machine identification is not just about intelligence; it is also about efficiency and environmental responsibility. |
|
Part Three: Chinese Companies Driving Local Innovation |
While American companies lead in global retail and logistics solutions, Chinese companies are applying similar technologies to solve specific, localized problems. The focus is often on practical applications in challenging environments. |
Weibiao Technology: Smart Labels for Real-World Challenges |
Weibiao Technology (Chongqing Weibiao Technology Co., Ltd.) is a Chinese company that has developed RFID solutions for a range of industries, including aviation, healthcare, and manufacturing. The company's approach is to identify 'pain points' in existing processes and apply RFID technology to solve them . |
The most visible application is in airport baggage handling. Traditional baggage tags use barcodes, which can become damaged or obscured during transit, requiring manual sorting. Weibiao Technology's solution embeds an RFID chip into the baggage tag. The tag is thin and flexible, and it can be read without line of sight, even if the tag is dirty, folded, or partially obscured . |
The impact is significant. According to Weibiao Technology, the RFID tags achieve a 99 percent correct identification rate, compared to about 90 percent for barcodes . For a busy airport like Chongqing Jiangbei International Airport, which handles millions of bags per year, this improvement translates into substantial cost savings. The company estimates that the RFID system can save the airport 'tens of millions of yuan' annually in labor costs and reduce lost baggage claims . |
Weibiao Technology's system was first deployed at Chongqing Jiangbei International Airport's Terminal 3A in 2017, and the company won the contract in competition with 12 other companies. In 2019, the technology was adopted by Beijing Daxing International Airport. In 2021, the Civil Aviation Administration of China promoted the technology to all major airports in the country. By 2023, more than 20 Chinese airports had adopted Weibiao's RFID baggage tracking technology . |
The company has also applied RFID to hospital laboratory workflows. In a typical hospital, blood samples and other specimens must be collected, labeled, transported, and tested. The process is labor-intensive and prone to errors. Weibiao Technology's solution automates the process from end to end. At the collection point, a device prints and applies a label with an RFID tag to each sample tube. The tube is then placed into a pneumatic tube system that transports it to the laboratory at high speed. At the laboratory, an RFID reader automatically identifies the tube and routes it to the appropriate testing station. The system tracks each sample throughout the process, ensuring chain of custody and providing real-time visibility . |
The company claims the system can improve testing efficiency by more than 50 percent. It has been deployed in several major hospitals in Chongqing, and the company plans to expand to other regions . |
Weibiao Technology has also developed a smart reagent management system for hospitals. Reagents are medical supplies used in diagnostics and treatment, and they must be stored under strict conditions. The system applies RFID tags to individual reagent packages, and tracks them from procurement through to usage. The system can automatically detect expired or soon-to-expire products, and it can trigger alerts when inventory is low. The company estimates that the system can save 'over one million yuan per year' for a large hospital, primarily by reducing waste and improving inventory management . |
Chinese Research and Applications in Specialized Domains |
Chinese research institutions are also contributing to the evolution of machine identification. A study on RFID-based coding for e-commerce material distribution addresses the challenge of high error rates of traditional barcodes in complex environments. The researchers propose a 'multi-band adaptive RFID coding solution' that improves identification accuracy. This work is funded by a state grid company and aims to support the 'smart logistics' strategy of China's 14th Five-Year Plan . |
Another research project, supported by the National Standardization Administration, is the revision of China's national standard for apparel coding and RFID tags. The standard specifies the encoding rules and RFID tag specifications for clothing commodities. It uses the GS1 EPC (Electronic Product Code) encoding scheme, which is a globally recognized standard, and it also incorporates the 'YiLian Eco-Alliance' clothing tag coding scheme . |
The standard is significant for two reasons. First, it provides a common framework for tagging apparel, enabling interoperability across the supply chain. Second, it extends the lifecycle of apparel management from the distribution chain to the consumer. The RFID tag can be used for 'automated identification and care, virtual fitting and outfit recommendations, and smart storage' in the home . This shows how machine identification is moving beyond the industrial supply chain into the consumer domain. |
A third research project, from the Wanfang Data knowledge platform, describes a system that integrates QR code and RFID technology for personnel and vehicle identification in challenging terrain . The system uses deep learning to optimize QR code recognition accuracy, and RFID for anti-counterfeiting verification. In field tests, the system achieved a 98.7 percent recognition accuracy and an anti-counterfeiting response time of less than 200 milliseconds. This demonstrates the value of combining multiple identification technologies and AI for enhanced performance . |
|
Part Four: The Role of AI in Enhancing Identification |
The evolution from simple identification to intelligent data networks is powered by AI. AI enhances machine identification in several ways. |
AI for Better Data Capture |
AI is being used to improve the accuracy and speed of barcode and RFID reading. Neural decoding, as used in Datalogic's Gryphon 4600 scanner, enables faster and more accurate reads even from damaged or poorly printed codes. Deep learning-based OCR, as used in Zebra's NS42 vision sensor, allows the system to read text that would be illegible to traditional OCR systems . |
AI for Data Interpretation and Analytics |
The data captured by RFID and barcode systems is more valuable when it is analyzed. AI can detect patterns, identify anomalies, and generate insights. For example, Zebra's Frontline AI Suite includes tools for analyzing data from multiple sources to provide real-time insights to frontline workers . |
AI for Loss Prevention and Security |
AI is a powerful tool for loss prevention. Datalogic's Magellan scanners use embedded AI to detect mis-scans, ticket switching, and item stacking during self-checkout . AI can also be used to identify unusual patterns of movement in a warehouse or retail store, flagging potential theft. |
AI for Predictive Maintenance |
By combining RFID tracking with IoT sensors and AI analytics, companies can monitor the condition of assets and predict when maintenance is needed. This is particularly valuable in manufacturing, where unexpected downtime is costly . |
AI for Workflow Optimization |
The data from identification systems can be used to optimize workflows. For example, RFID data can show where bottlenecks are occurring in a supply chain. AI can analyze that data and recommend changes to improve efficiency . |
|
Part Five: The Future of Machine Identification |
The evolution of machine identification is far from over. Several trends will shape the future. |
Ubiquitous RFID and Item-Level Tracking |
RFID tags are becoming cheaper and smaller, making item-level tracking feasible for a wider range of products. The apparel standard being developed in China is one example of how this is happening . In the future, every item in a retail store, every package in a warehouse, and every component in a factory might have an RFID tag. This will enable a level of visibility that is currently unimaginable. |
Integration of Multiple Technologies |
The future of machine identification is not just RFID or barcodes; it is a combination of multiple technologies. As the research on integrated QR code and RFID systems shows, different technologies have different strengths, and combining them can provide a more robust solution . We will likely see systems that combine RFID, machine vision, AI, and IoT in seamless ways. |
AI-Driven, Self-Learning Systems |
The systems of the future will not just be programmed; they will learn. They will analyze data to improve their own accuracy and efficiency. They will adapt to changing conditions without human intervention. This is the 'intelligent' in intelligent data networks. |
From Identification to Full-Spectrum Sensing |
The RFID tag of the future may do more than just identify an item. It may also sense temperature, humidity, shock, and other environmental factors. This would transform the tag from a passive identifier into an active sensor node. |
Consumer-Facing Applications |
Machine identification is moving beyond the supply chain into consumer applications. The apparel RFID standard mentioned above includes applications like automated washing, virtual fitting, and smart storage . Consumers are already familiar with QR codes for menus and payment; they may soon interact with RFID tags in their clothing, appliances, and personal items. |
Sustainability and Circular Economy |
Machine identification can support the circular economy by enabling tracking of products throughout their lifecycle. An RFID tag could track a product from manufacture to use to recycling, enabling better management of resources and reducing waste. |
|
Conclusion: A Detailed Summary |
Machine identification technologies have evolved dramatically from the simple barcode to intelligent data networks. The barcode, which was a revolutionary invention in the 1970s, has fundamental limitations: it requires line of sight, it cannot be read if damaged, it can only identify the type of item, and the information is fixed and cannot be updated . RFID overcomes these limitations: it does not require line of sight, it can be read in harsh environments, it can identify a specific item, and the information can be updated repeatedly . |
The evolution has been driven by the convergence of RFID with IoT, AI, and edge computing. IoT provides the connectivity to share data in real time. AI provides the intelligence to analyze the data and generate insights. Edge computing provides the processing power at the point of capture, reducing latency. The result is a system that is not just identifying items but providing a digital thread that connects physical objects to intelligent systems. |
American companies are leading the commercial deployment of these intelligent identification systems. Zebra Technologies has evolved from a barcode printer to a provider of 'frontline intelligence.' The company's portfolio includes AI-native mobile computers, AI-driven vision sensors, and a Frontline AI Suite that provides on-device AI models and GenAI-driven agents. Zebra's solutions are used in manufacturing, logistics, retail, and healthcare, enabling automated inventory tracking, quality inspection, and workflow optimization . |
Datalogic is embedding AI directly into its scanners and self-shopping devices. The Magellan scanners are 'the world's first and only devices with embedded AI Smart Vision for Retail,' enabling real-time detection of self-checkout issues. The Joya Smart devices combine scanning, RFID, and AI to enable a seamless self-shopping experience . |
Chinese companies are applying similar technologies to solve local challenges. Weibiao Technology has deployed RFID baggage tags at more than 20 Chinese airports, achieving a 99 percent correct identification rate and saving millions of yuan in labor costs. The company has also deployed RFID systems in hospitals to automate sample transport and reagent management, improving efficiency by more than 50 percent . Chinese research institutions are developing AI-enhanced systems for personnel identification in challenging terrain and contributing to national standards for apparel RFID tags that extend from the supply chain to the consumer . |
The role of AI in machine identification is multifaceted: it improves data capture accuracy, enables real-time analytics, supports loss prevention, and enables predictive maintenance. The future will see more ubiquitous RFID tagging, integration of multiple technologies, AI-driven self-learning systems, and consumer-facing applications. Machine identification will become a key enabler of the circular economy by tracking products throughout their lifecycle. |
|
In the end, the evolution of machine identification is about transforming simple labels into intelligent nodes in a broader digital ecosystem. The barcode was a one-way communication: you read it, and that was it. The intelligent data network is a two-way, adaptive system: it identifies, tracks, analyzes, and acts. This is not just an incremental improvement; it is a fundamental shift in how we connect the physical and digital worlds. The machine identification technologies of today are not just about knowing what something is; they are about knowing where it is, what condition it is in, and what it needs to be efficiently managed. |
cs@easiersoft.com