本发明涉及人脸识别领域，特别涉及一种适应变电站进出监控需求的人脸快速识别方法。
背景技术：
目前变电站的进出监控依赖于人为的核准与监督而实行。人工核验需要用户的配合，效率低下，其准确率受限于检验人员的工作效能。基于机器学习的人脸检测技术具有不需要用户的主动配合，具有准确率高以及检测效率不随着时间变化而降低的优点。现有的人脸检测技术在图片尺寸较小的时候能够达到实时检测，但由于变电站属于开阔空间地带，摄像头需要覆盖较大的视野范围，其图片尺寸也会相应的增加。而在变电站的监控视频图片尺寸较大的时候，则无法在较少的时间内得出检测结果，因为当图片尺寸增加，其所包含的像素点会呈平方次数增加，在人脸检测单位处理速度不变的情况下，处理的时间也会呈平方次数增加。
在变电站内，人脸快速识别具有极其重要的意义。只有对视频内出现的人员进行实时分析，对不符合要求的人员及时识别报警并提醒值班人员关注，才能够有效地防患事故于未然。因此，在视频监控系统中，快速人脸识别算法能够保证视频监控报警的有效性，能够全面地提升变电站的安全系数，对变电站的安全工作具有重要的意义。
技术实现要素：
本发明的目的在于变电站进出监控中如何对人脸进行快速识别的问题，旨在提供一种适应变电站进出监控需求的人脸快速识别方法，以保证视频监控报警的有效性，更全面地提升变电站的安全系数。
本发明提出一种适应变电站进出监控需求的人脸快速识别方法，包括以下步骤：
一种适应变电站进出监控需求的人脸快速识别方法，其特征在于包括以下步骤：
(1)将待识别图像iw*h*3输入参数已知的yolo神经网络并输出图像中每个人的边界框；
(2)根据所得到的边界框，将每个检测到的人物从图像iw*h*3中提取出来，得到子图片集合
(3)对所述子图片进行灰度化处理并计算得到待识别的人脸特征值；
(4)根据所述待识别的人脸特征值与人脸数据库中各人脸特征值的距离l以得出人脸识别结果。
上述的一种适应变电站进出监控需求的人脸快速识别方法，其特征在于，一个视频由许多帧的的图片所组成，可定义一帧图像iw*h*3，为一个三维张量(tensor)，其中w为图片宽，h为图片高，3为图片通道数，分别为rgb通道。
上述的一种适应变电站进出监控需求的人脸快速识别方法，其特征在于，神经网络可定义为一个函数y＝f(w,x)，w是神经网络的参数；神经网络的输入x与输出y是一个张量，其参数通过梯度下降法求解得出；所述梯度下降法是一阶优化算法，求解过程为通过多次迭代寻找能够使得函数取得局部极小值的参数；
所述yolo神经网络为众多神经网络的一种，可接受一张608*608像素的图像作为输入，图片像素与多个卷积池化层以及全连接层的参数进行运算，最后输出检测到的物体的边界框；yolo神经网络的参数可通过梯度下降法求解得到，yolo神经网络的输出为n个边界框{bboxi|i∈0,1,2,...,n}，其中第i个边界框可表示为bboxi＝(xi,yi,wi,hi)；边界框指示了检测到的人在输入图片的位置，其中xi,yi为边界框的中心坐标，wi,hi分别为边界框的宽和高。
上述的一种适应变电站进出监控需求的人脸快速识别方法，其特征在于，所述“对所述子图片进行灰度化处理”是指将所述子图片集合
式中，
上述的一种适应变电站进出监控需求的人脸快速识别方法，其特征在于，所述“计算得到待识别的人脸特征值”是指对于每一张人脸灰度图
上述的一种适应变电站进出监控需求的人脸快速识别方法，其特征在于，所述“根据所述待识别的人脸特征值与人脸数据库中各人脸特征值的距离l”是指通过公式(2)计算待识别的人脸特征值和每一张已知人脸的特征值的l距离，具体为：
上述的一种适应变电站进出监控需求的人脸快速识别方法，其特征在于，所述“得出人脸识别结果”是指当l≤0.6成立时，则表示人脸识别匹配成功，即当前待识别的人脸图片在人脸数据库中；否则表示匹配失败，即当前待识别的人脸图片不在人脸数据库中。
与现有技术相比，本发明的有益效果在于：
(1)基于机器学习的人脸检测技术具有不需要用户的主动配合，具有准确率高以及检测效率不随着时间变化而降低的优点；
(2)可对监控视频出现的人员进行实时分析，能够全面地提升变电站人员进出安全系数。
附图说明
图1是一种适应变电站进出监控需求的人脸快速识别方法的流程示意图。
图2是一张待识别的现场图片。
图3是根据图2提取得到每个人的边界框。
图4是根据图3检测到人脸的图片。
图5是人脸数据库中的两张人脸图片。
具体实施方式
以下结合附图和实例对本发明的具体实施做进一步说明。
图1反映了一种适应变电站进出监控需求的人脸快速识别方法的具体流程。一种适应变电站进出监控需求的人脸快速识别方法包括：
(1)将待识别图像iw*h*3输入参数已知的yolo神经网络并输出图像中每个人的边界框；
(2)根据所得到的边界框，将每个检测到的人物从图像iw*h*3中提取出来，得到子图片集合
(3)对所述子图片进行灰度化处理并计算得到待识别的人脸特征值；
(4)根据所述待识别的人脸特征值与人脸数据库中各人脸特征值的距离l以得出人脸识别结果。
以下是本发明方法的一个实际算例，以一张待识别的现场图片进行分析统计，图2为该待识别的现场图片。
(1)给定一张输入图片并使用yolo神经网络检测，得到图片中每个行人的边界框，即图2中的绿色框；
(2)提取行人图片，结果如图3所示；
(3)检测行人图片中是否存在人脸，其中图3(a)无法检测到人脸，而图3(b)检测的人脸进行灰度化处理后如图4所示；
计算检测到的人脸特征值为：
y＝[-0.0653979,0.0481431,0.0459506,-0.00297655,-0.0784676,-0.0707229,0.0017481,-0.129902,0.0899502,-0.0961162,0.18472,-0.04566,-0.177452,-0.119043,-0.00185613,0.123365,-0.101321,-0.0666383,-0.0822533,-0.0954023,0.0388882,0.0508717,0.0408848,0.0335565,-0.048666,-0.263044,-0.0828425,-0.0912373,0.0823404,-0.022558,0.0207027,0.0748649,-0.162295,-0.0786254,-0.00673984,0.034479,-0.0504447,-0.0587812,0.283734,-0.0457539,-0.132397,-0.0031759,0.0189574,0.271597,0.222758,-0.0507444,-0.00261238,-0.0372537,0.147651,-0.213665,0.0485419,0.125421,0.116833,0.0580825,0.0118643,-0.092228,0.0441321,0.137712,-0.203144,0.0675892,0.0768298,-0.168619,-0.037038,-0.0307111,0.145781,0.0454205,-0.0487946,-0.157138,0.140418,-0.18163,-0.0306611,0.0311677,-0.13422,-0.139509,-0.310455,0.0523049,0.299887,0.160369,-0.182182,0.0295619,-0.0594396,-0.00560254,0.131522,0.030557,-0.0181286,-0.0943769,-0.115649,0.0168677,0.160838,-0.0744993,-0.0537497,0.240962,0.0239171,-0.00718346,0.0277776,0.00476684,-0.0710697,0.0229204,-0.126025,0.0441402,0.0814756,-0.113622,0.0699524,0.109151,-0.174594,0.17626,-0.0370767,0.0387941,0.0300786,-0.0633792,-0.155115,-0.0208994,0.190634,-0.121869,0.197604,0.131385,0.0213304,0.151944,0.101643,0.104232,-0.0347045,-0.0272474,-0.172505,-0.0462738,0.0765506,0.0254679,0.0702121,0.0738493]。
(4)计算图5(a)和图5(b)中人脸图片的特征值，其中图5(a)的特征值为：
x1＝[-0.117174,0.0432565,0.0615167,-0.0455164,-0.0706958,-0.0455732,-0.0510033,-0.166344,0.0750411,-0.131235,0.231625,-0.0978745,-0.196297,-0.163042,-0.0387142,0.198247,-0.165635,-0.0940813,-0.0818975,0.00266437,0.097009,0.00571842,0.0382776,0.0467014,-0.0377104,-0.343167,-0.142302,-0.0462031,0.0399787,-0.0398614,-0.0901296,0.0863726,-0.138756,-0.102175,0.0238005,0.0919469,0.0509093,-0.101035,0.16137,-0.0239519,-0.19225,0.0776478,0.0727825,0.227224,0.25847,0.0510676,0.033865,-0.11599,0.153841,-0.180689,0.0471576,0.0887595,0.0761694,0.0240254,0.0222875,-0.123475,0.0528614,0.109238,-0.154681,0.0114299,0.109626,-0.131581,-0.0354409,-0.0367509,0.172551,0.117364,-0.0914756,-0.234913,0.116833,-0.154709,-0.0901379,0.00995727,-0.212836,-0.0959615,-0.36433,0.0252173,0.399847,0.113429,-0.17775,0.0973187,-0.0163856,0.0396124,0.172737,0.177377,-0.0107027,-0.0259293,-0.0973322,-0.00115101,0.166939,-0.0176149,-0.0846889,0.226914,-0.00471437,0.0738132,0.0667178,0.0158959,-0.0309987,0.0283487,-0.114391,0.0181091,0.0817547,-0.0177323,0.0326894,0.116774,-0.121201,0.112234,-0.0803487,0.10843,0.0436463,-0.0872214,-0.0972368,0.0326281,0.0974472,-0.205026,0.155965,0.162803,0.0626691,0.0859266,0.165243,0.100398,0.0339094,0.0320152,-0.219555,-0.0229654,0.125368,-0.0185362,0.124461,0.0339912]；
图5(b)的特征值为：
x2＝[-0.0886385,0.0549423,-0.00546047,-0.0652526,-0.0734701,-0.0891774,-0.0802805,-0.107656,0.0891029,-0.0775417,0.223884,-0.0995733,-0.182051,-0.155407,-0.0623404,0.169066,-0.209147,-0.105432,-0.0768067,0.0224611,0.0916727,0.0154249,0.0561431,0.00278886,-0.0536536,-0.389321,-0.0723899,-0.0686888,0.0249965,-0.0387781,0.0280211,0.0195251,-0.205907,-0.0690564,0.0598259,0.045773,-0.000860474,-0.047665,0.166677,-0.000445154,-0.222018,-0.000221765,0.0265393,0.212571,0.173407,0.0996724,0.0324839,-0.161677,0.152244,-0.151677,0.0278293,0.153962,0.0366221,0.0213522,0.013047,-0.0643532,0.0583508,0.143498,-0.139628,-0.0140229,0.11262,-0.0348965,-0.0218506,-0.100726,0.203176,-0.00134878,-0.11952,-0.173758,0.0945559,-0.13104,-0.0708779,0.0354774,-0.12939,-0.165783,-0.316124,-0.00536548,0.363126,0.0811748,-0.144772,0.0391074,-0.0603582,0.0201964,0.123615,0.112289,-0.0304388,0.01851,-0.0759722,-0.0475834,0.213184,-0.0673732,-0.0385347,0.189155,-0.0458143,0.0599411,0.0131644,-0.028812,-0.0700638,0.057142,-0.0723014,0.00542674,0.0530222,-0.050184,0.0077821,0.0820619,-0.12622,0.10774,-0.0368184,0.0628868,0.0506034,-0.0115336,-0.0934998,-0.0939013,0.0922058,-0.190224,0.232884,0.183856,0.110786,0.120159,0.135563,0.100329,-0.0339753,0.0397254,-0.180747,-0.0519826,0.084807,0.0213815,0.0910269,-0.0357912]；
分别计算y与x1,x2的距离l，可得：
‖x1-y‖2＝0.574211989425；‖x2-y‖2＝0.604223930903；
由于‖x1-y‖2小于0.6，说明当前人脸数据库中的人脸图片图5(a)与图3(b)匹配。
以上所述实施例仅表达了本发明的实施方式，其描述较为具体和详细，但并不能因此而理解为对发明专利范围的限制。应当指出的是，对于本领域的普通技术人员来说，在不脱离本发明构思的前提下，还可以做出若干变形和改进，这些都属于本发明的保护范围。因此，本发明专利的保护范围应以所附权利要求为准。