中圖分類號(hào)： TN929.5；TP391.4 文獻(xiàn)標(biāo)識(shí)碼： A DOI：10.16157/j.issn.0258-7998.200861 中文引用格式： 周揚(yáng)，熊俊俏. 基于Wi-Fi和人臉識(shí)別技術(shù)的智能鎖系統(tǒng)設(shè)計(jì)[J].電子技術(shù)應(yīng)用，2021，47(6)：57-61. 英文引用格式： Zhou Yang，Xiong Junqiao. Design of intelligent lock system based on Wi-Fi and face recognition technology[J]. Application of Electronic Technique，2021，47(6)：57-61.
Design of intelligent lock system based on Wi-Fi and face recognition technology
Zhou Yang，Xiong Junqiao
School of Electrical and Information Engineering，Wuhan Institute of Technology，Wuhan 430205，China
Abstract： Aiming at the security problems existing in traditional door locks, an intelligent lock system combining Wi-Fi technology and face recognition technology is designed. The system adopts STM32F103 as the core control device and CC1101 as the communication and transmission module, which can effectively reduce power consumption and realize information interaction. MULBP(Multiple Uniform Local Binary Patterns，MULBP) face recognition algorithm is put forward at the same time, compared with before improvement to enhance the image grayscale average and each row, column and diagonal elements, the relationship between value of new LBP operator is more able to reflect the local texture feature, and combined with 2DLDA(Two-dimensional Linear Discriminant Analysis，2DLDA) algorithm find the projection matrix, to make the sample more easy to distinguish, easier for feature extraction. The results show that the fusion algorithm not only reduces the feature dimension but also improves the recognition rate, making the system achieve the purpose of intelligent anti-theft.
Key words : intelligent lock；wireless communication technology；local binary mode；two-dimensional linear discriminant analysis
0 引言
隨著無線通信技術(shù)和物聯(lián)網(wǎng)技術(shù)的不斷發(fā)展，人們的財(cái)產(chǎn)安全和信息安全意識(shí)不斷提升，智能家居的理念也已經(jīng)成為人們關(guān)注的熱點(diǎn)問題[1-2]。文獻(xiàn)[3]中介紹的智能鎖系統(tǒng)采用了一種藍(lán)牙與FPGA相結(jié)合的技術(shù)，經(jīng)驗(yàn)證表明該系統(tǒng)的保密性和安全性都有所增強(qiáng)，但存在功耗較大和成本較高的問題；文獻(xiàn)[4]中提出了一種基于LBP特征和改進(jìn)Fisher準(zhǔn)則的方法來提取直方圖的特征，得到了較好的效果，但忽略了人臉的局部細(xì)微特征。因此，本文采用傳輸速度快、接入方便的Wi-Fi技術(shù)來完成智能鎖的通信，選擇功耗較低且頻段豐富的CC1101芯片來完成數(shù)據(jù)的收發(fā)。并結(jié)合多重均勻局部二值模式(Multiple Uniform Local Binary Patterns，MULBP)與二維線性判別分析(Two-dimensional Linear Discriminant Analysis，2DLDA)融合人臉識(shí)別算法采集信息進(jìn)行身份識(shí)別，經(jīng)實(shí)測(cè)表明該系統(tǒng)實(shí)現(xiàn)數(shù)據(jù)穩(wěn)定傳輸?shù)耐瑫r(shí)略微提升了識(shí)別的準(zhǔn)確率。
