Abstract
This system used STM32H750XBH6 as the main control, and ART -Pi multimedia expansion board equipped with ILI9488 capacitive touch screen, WM8988 audio chip and GC0328C camera. It realizes recording, recognition, display, imposter capture and telescopic rod connection for door lock control. Based on convolutional neural network (CNN) and STM32 Cube. AI toolkit, voiceprint model was built and deployed to STM32H750XBH6. Combining Mel-Frequency Cepstral Coefficients (MFCC) and DTW algorithm, voice recognition function was realized. Through dual authentication of voiceprint verification and voice recognition, the system guaranteed high security. What is more, the system was equipped with a WIFI module, and the administrator can log in to the Onenet website to view the access control information. If there are three consecutive recognition errors, the system automatically capture the person and save the fake authentication certificate. After several improvements, the system can operate normally. The accuracy rate of voiceprint recognition and voice recognition were 97.83% and 96.00% respectively and the overall accuracy rate was 93.50%. Compared with traditional password access and card access, the system did not have problems of leak and lose. It provided users with high-security, true-intention, low-cost, and weak-privacy authentication services.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
