Macc Series Iris Recognition Accelerator Cards – Multi-Model High-Performance

Product Specification

Product Overview

      ● Macc.p & Macc.m series: Equipped with Xilinx A7 chips, they are suitable for scenarios like database acceleration, machine learning, data analysis, and video processing.

      ● Macc.c series: Uses Xilinx K7 chips with 6 built-in AI cores, supporting fast comparison of data for up to 1 million people and providing a maximum comparison processing capacity of 800,000 templates per second.

Core Functions and Technical Features

1. Multi-Series for Diverse Scenarios

      ● Macc.p series (models: Macc.p-10K, Macc.p-50K, Macc.p-100K) caters to applications with 10,000–100,000 users. Leveraging Xilinx A7 chips, it excels in scenarios such as database acceleration and machine learning.

      ● Macc.m series (model: Macc.m-10K) is designed for 10,000-user scenarios. With a compact size of 40mm×22mm×3.85mm and Xilinx A7 chips, it’s ideal for embedded device integration.

      ● Macc.c series (models: Macc.c-500K, Macc.c-1000K) targets large-scale applications with 500,000–1,000,000 users. Featuring Xilinx K7 chips and 6 AI cores, it enables rapid comparison for massive datasets.

2. High-Speed Encoding & Comparison

      ● Encoding Speed: The Macc.c series achieves up to 50 original images per second, while the Macc.p and Macc.m series reach 20 original images per second — ensuring efficient iris feature extraction.

      ● Comparison Speed:

               Macc.p series: Supports 10,000–100,000 templates per second (scalable based on specific models).

               Macc.m series: Processes 10,000 templates per second.

               Macc.c series: Delivers up to 800,000 templates per second for large-scale real-time comparison.

      ● Accuracy: All series maintain a False Acceptance Rate (FAR) of 10⁻⁷ and a False Rejection Rate (FRR) of 10⁻³, guaranteeing high recognition precision.

3. Hardware Specifications & Interfaces

      ● Physical Interface:

               Macc.p series uses Mini PCIE, fitting embedded or small-form-factor devices.

               Macc.m series adopts an M.2 B&M key interface, optimized for compact embedded integration.

               Macc.c series utilizes a C-PCI interface, suited for industrial or large-scale device deployment.

      ● Communication Interface:

               Macc.p and Macc.m series rely on USB 2.0, enabling simple connections to X86 or ARM platforms.

               Macc.c series uses Ethernet, supporting cloud-based or networked large-scale deployments.

      ● Product Size & Power Consumption:

               Macc.p series: Measures 51mm×30mm×4mm, with power consumption ranging from 1W to 3W (varying by model).

               Macc.m series: Ultra-compact at 40mm×22mm×3.85mm, with a low power draw of 1.5W.

               Macc.c series: Larger form factor at 233.36mm×160mm×8mm, with power consumption between 18W and 25W to support large-scale computing.

4. Cross-Platform & Scalability

      ● Platform Compatibility: The cards support X86, ARM, Windows, and Android systems, enabling seamless integration into diverse device ecosystems.

      ● Scalable Deployment:

               Multi-card parallel processing is supported to scale performance (e.g., using multiple Macc.c cards to boost comparison speed).

               Deployment is interchangeable between cloud and local environments, allowing flexible expansion as application loads grow.

      ● SDK Support: Multi-platform SDK development kits are provided, enabling users to quickly conduct secondary development.

Typical Application Scenarios

      ● Embedded Devices: Macc.p and Macc.m series are used in iris access control terminals, smart locks, or portable iris collectors, benefiting from their small size and low power consumption.

      ● Large-Scale Systems: Macc.c series powers cloud-based iris recognition platforms (such as public security population management or enterprise-level access control for millions of users).

      ● Industry Integrators: Equipment manufacturers, system integrators, and solution providers can add iris recognition capabilities to existing embedded or PC-based systems via these accelerator cards.

Why Choose Macc Series?

      1. Algorithm Advantage: The Phaseiris^TM3.0 algorithm ensures high accuracy (FAR 10⁻⁷/FRR 10⁻³) and efficient processing.

      2. Multi-Model Scalability: The three series cover user scales from 10,000 to over 1,000,000, matching a wide range of project requirements.

      3. Cost-Effectiveness: Compared to CPU/GPU solutions, these cards offer faster throughput, lower latency, and more economical deployment.

      4. Easy Integration: Cross-platform support plus standardized interfaces (Mini PCIE, M.2, C-PCI) simplify hardware integration.

      5. Future-Proof Design: Scalable capacity/performance and interchangeable cloud/local deployment adapt to evolving application needs.

      As a professional enterprise in the biometric sector, Wuhan Homsh Technology drives secure authentication through technological innovation. Leveraging proprietary core algorithms, end-to-end technical capabilities, and high-precision hardware expertise, we provide professional support for our full range of biometric products, including iris recognition and vein recognition.

      We cater to the needs of global industries like finance and security, offering technical consultation, custom development, and round-the-clock after-sales support. We look forward to partnering with global counterparts to build a secure authentication ecosystem via biometric technology and empower industrial digital upgrading.

Specifications