In today’s fast-paced world, rapid verification in access control is not just a convenience; it's a necessity. Delays in access verification can lead to bottlenecks in high-traffic environments, disrupt operations, and degrade the user experience. More critically, the speed at which individuals can be verified and granted access can be a matter of life and death.
Low speed of processing and available performance headroom of the equipment used may increase the risk of missing detection of people or objects due to an inability to use more advanced and more reliable image processing, such as using the latest neural networks, such as YOLOv8, picking the best picture from several, alignment, and real-time matching.
AI TECH INSIGHT
Access control is a fundamental element in safeguarding both physical and digital environments. Integrating vision AI has significantly advanced access control systems, offering a level of automation and intelligence previously unattainable. Especially for biometric access control systems. Yet, the challenge remains: How can we speed up verification without compromising accuracy? More specifically, how do we reduce false positives and negatives?
This blog explores the current and future state of AI access control, the pivotal role of verification speed, and a method to increase verification speed without increasing false positives in security.
The evolution of vision AI in access control
At present, vision AI applications in (biometric) access control systems are primarily used for identification and verification, and sometimes for motion detection and behavior analysis. With technologies like facial recognition, object detection and anomaly detection, we've moved from reactive to proactive security measures. Looking ahead, we envision a more sophisticated integration of AI in access control, where adaptive learning algorithms can predict potential security breaches before they occur, and personalized access protocols cater to the unique security requirements of individual users or entities.
The critical importance of speed in verification
In today’s fast-paced world, rapid verification in access control is not just a convenience; it's a necessity. Delays in access verification can lead to bottlenecks in high-traffic environments, disrupt operations, and degrade the user experience. More critically, the speed at which individuals can be verified and granted access can be a matter of life and death.
Low speed of processing and available performance headroom of the equipment used may increase the risk of missing detection of people or objects due to an inability to use more advanced and more reliable image processing, such as using the latest neural networks, such as YOLOv8, picking the best picture from several, alignment, and real-time matching.
Why accuracy matters too
Every millisecond saved in the verification process enhances the user experience and operational efficiency. However, every incorrect decision made by the system — be it a false positive or a false negative — undermines trust in the security framework and can cause delays itself. High traffic environments, such as airports, commercial buildings, and public events, require a solution that combines high-speed, high-accuracy verification to maintain security without disrupting the flow of movement. The goal, therefore, is a verification process that is not only fast but also reduces false positives and false negatives in security to the absolute minimum.
The challenge with current AI accelerators
Current AI accelerators have made significant strides in improving the efficiency of running vision AI models. However, they often face a trade-off between speed and accuracy, as they commonly deploy 8-bit integer inference arithmetic instead of 32-bit floating-point full-precision. High verification speeds can sometimes result in increased false positives and negatives, as the security and surveillance systems may not spend enough time analyzing the data to make accurate decisions. This is particularly problematic in access control, where errors can either compromise security by allowing unauthorized access or hinder operations by denying access to legitimate users. Therefore, eliminating false negatives and false positives in machine learning used for automatic identification is important.
Fortunately, Axelera AI solved the challenge of reducing precision of the mathematical computations without any practical accuracy loss, eliminating the false positives in security processes produced by vision AI accelerators.
The exceptional performance and accuracy of the Axelera AI acceleration platform have significantly fueled our collaborative efforts. Its unmatched performance-to-price ratio, surpassing traditional GPUs and dedicated AI processing units, has been critical in our selection process. We are confident that leveraging their state-of-the-art YOLO performances will empower us to tackle new challenges in our current and future video analysis applications.
Alexandre Perez, R&D Director at XXII.
How we accelerated vision AI applications without accuracy loss
To address the challenges outlined above, our engineers took a radically different approach to data processing. By combining Axelera’s proprietary digital in-memory computing technology (D-IMC) and a unique post-quantization method, Axelera has created the Metis AIPU – the most powerful AI accelerator for the edge you can buy today. Its unmatched efficiency and accuracy redefine the standard for AI access control. The technology ensures that vision AI models run with the same accuracy as PCs or GPUs (FP32 equivalent), but at significantly lower cost and power consumption while delivering the highest level of accuracy to minimize false positives and negatives. It can make biometric access control systems not only efficient but also highly reliable.
D-IMC explained
In-memory computing addresses two main issues with today’s AI processing hardware - power efficiency and cost. It is a radically different approach to data processing where the compute-intensive matrix-vector multiplications of neural networks happen “in-place” without intermediate movement of data between an on-chip memory block and the computational unit.
Unlike examples of in-memory computing architecture that uses analog computing, Axelera’s proprietary digital in-memory (D-IMC) technology takes the concept further, using SRAM (Static Random-Access Memory) memory that is densely interleaved with digital computation, each memory cell effectively becoming a matrix-vector compute element. This greatly increases the number of operations per computer cycle (one multiplication and one accumulation per cycle per memory cell) possible on a small area of silicon without suffering from issues, such as noise and limited accuracy commonly associated with analog in-memory computing. To put it in perspective, the Axlera Metis AIPU delivers an industry-leading 214 Tera-Operations per Second (TOPS) of AI processing, while achieving a power efficiency of 50 TOPS per Watt at 50% input and data sparsity.
Thanks to Axelera’s D-IMC technology, the Metis Evaluation Kits deliver lightning-fast inference processing without compromising accuracy [1,2,3 ].
High-accuracy quantization
To reach high performance, AI accelerators often deploy 8-bit integer processing of the most compute-intensive parts of neural network calculations instead of using 32-bit floating-point arithmetic. To do so, a quantization of the data from 32-bit to 8-bit needs to be done.
Axelera has developed a sophisticated quantization flow methodology and toolkit that does not require time-consuming retraining of models and only requires a fast calibration using a small subset of the image dataset. In combination with the mixed–precision arithmetic of the Axelera Metis AIPU, Axelera can deliver an accuracy practically indistinguishable from a reference 32-bit floating point model. As an example, Metis AIPU can run the ResNet50v1 neural network processing, at a full processing speed of 3,200 frames per second, with a relative accuracy of 99.9%.
How OEMs can profit from Metis
The just-described hardware and software technologies are responsible for the excellent cost efficiency, power efficiency and accuracy of the Metis AIPU. It’s why OEMs installing an Axelera AI accelerator, can reduce operational costs without increasing false positives in security. Or upgrade their biometric access control systems with an increased number of access control points connected to the edge server without loss of speed. This means that OEMs can offer competitively priced systems for AI access control, while contributing to sustainability efforts by minimizing energy consumption.
How Metis accelerates the OEM innovation process
To assist OEMs in the Access Control and Surveillance industries, Axelera created 4 variants of the Metis Evaluation Kit consisting of a host system (x86 or ARM), a Metis PICe card, and the Voyager SDK. Providing an end-to-end system with sample code and ready-to-deploy AI processing, it helps OEMs simplify and accelerate the integration of advanced Vision AI technologies:
Experience the high efficiency and accuracy of Metis yourself
We invite Embedded Hardware Engineers and Machine Learning Specialists of original equipment manufacturers in the Surveillance & Security industry and independent software vendors to explore the potential of our Metis Evaluation Kit. Together, we can establish new standards for accuracy, cost efficiency, and energy efficiency of AI access control – and biometric access control systems in particular – creating safer and more efficient environments for everyone. We invite you to experience the remarkable capabilities of our Metis Evaluation Kit firsthand and join us in redefining the future of security and surveillance.
References
Address
Sign Up for Our Newsletter
Thank you for your newsletter subscription
