基于CNN-FOA-SVM算法的地铁站人脸识别研究

Research on subway station face recognition based on CNN-FOA-SVM algorithm

地铁安全事故社会影响大,单纯依赖摄像头快速、准确搜索个人信息难度大,这使得人脸识别在地铁事故处理中具有广泛的应用。论文采用Viola-Jones算法检测人脸并进行人脸剪裁,并将剪裁后的人脸图像输入到CNN中对人脸表情特征提取。提取的人脸特征作为样本对FOA-SVM模型进行训练,获得用于人脸识别的FOA-SVM模型。将提出的FOA-SVM模型应用于现成的人脸数据库,结果表明该模型对人脸识别的准确率均在92%以上,同时人脸识别稳定性强,这对人脸识别在地铁运营中的应用具有一定的参考价值。

Subway safety accidents have a great social impact,and it is difficult to search personal information quickly and accurately by relying solely on cameras,which makes face recognition widely used in subway accident processing.In this paper,the Viola Jones algorithm is used to detect the face and cut the face,and the cut face image is input into CNN to extract the facial expression features.The extracted face features are used as samples to train the FOA-SVM model and obtain the FOA-SVM model for face recognition.The proposed FOA-SVM model is applied to the ready-made face database.The results show that the accuracy of the model for face recognition is more than 92%,and the stability of face recognition is strong,which has a certain reference value for the application of face recognition in subway operation.