KEYWORDS: Facial recognition systems, Education and training, Clouds, Feature extraction, Control systems, Design and modelling, Cameras, Control systems design, Databases, LCDs
In order to improve the intelligent and secure management of community access control, a design scheme of face access control system based on SSD and OneNET cloud platform is proposed. The hardware system is mainly composed of the main control module of Raspberry Pi, human infrared detection sensor, RFID card reader, Web server and OneNET cloud platform. Support multiple door opening methods: face door opening, RFID card door opening and web remote door opening. Provide a Web background management system and APP mobile terminal user system for property management and residents. Face detection adopts a simple and convenient SSD framework for training. Select MobileNet-V3 as the backbone network and train a lightweight model suitable for running on mobile devices. All access control devices are connected to the OneNET cloud platform through the Raspberry wireless WiFi module, which is used for device access, device control and unified management of devices.
At present, in smart city and other applications, message middleware Kafka is usually used to distribute massive image terminal data to business components such as data analysis. However, in practical business applications, it is found that the traditional Kafka message distribution mode is usually difficult to apply to high real-time massive data distribution scenarios. To solve this problem, the message distribution mode of the system is optimized, and a mode of separating normal real-time distribution from abnormal reissue is proposed. In this distribution mode, when the subscriber is normal, the message is directly distributed to the subscriber in real time. When the subscriber is abnormal, Kafka is responsible for reissuing the message after fault recovery. The experimental results show that the delay and resource consumption of the distribution mode are significantly reduced, the delay is reduced by 60% ~ 70%, and the CPU and memory utilization are reduced by 45% and 43% respectively.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.