Content area
In order to improve the security and practicality of the access control system, an online recognition access control system based on the Internet of Things and microcontroller is designed. Taking STM32FI03RCT6 microcontroller as the core control center, RFID technology is used for personnel information recognition, and convolutional neural networks are introduced for facial image processing. Meanwhile, Raspberry Pi 3B+ is used as an auxiliary controller to achieve liveness detection. The experiment was conducted under the Windows 10 operating system using Intel (r) Core TM i5-10400F processor and 8GB memory are used for face detection under different lighting conditions to evaluate the robustness of the system. The results showed that the proposed method detected the target for the first time with an average frame rate of 194, which had stronger performance compared with support vector machines and convolutional neural networks. In addition, the accuracy of the system was 98.3%, and the final loss value was 0.012%. The research shows that this online identification access control system can effectively meet the needs of modern households and businesses for fast and accurate identity verification, demonstrating good practical prospects.
Details
Operating systems;
Face recognition;
Accuracy;
Internet of Things;
Brain cancer;
Communication;
Microprocessors;
Artificial neural networks;
Windows (computer programs);
Radio frequency identification;
Data processing;
Control systems;
Image processing;
Access control;
Efficiency;
Authentication protocols;
Support vector machines;
Households;
Neural networks;
Control centres;
Classification;
Target detection;
Design;
Methods;
Algorithms;
Image processing systems