Xuanang Feng’s research while affiliated with Nagoya University and other places

Personal identification is the task of authenticating a person using individual biological features. Deep neural networks (DNNs) have demonstrated an impressive performance in this field. Since no general algorithm is available for the design of network structures and the parameters adopted in DNNs for every application problem, DNNs should be determined according to the programmers’ experiments and know-how. For a new application task, it is very time-consuming for non-experts to design network structure, hyperparameters and an ensemble of base models adequately and effectively. In this paper, we present a genetic algorithm (GA)-based approach to construct network structures, tune their hyperparameters, and generate base models for the ensemble algorithm. The ensemble is constructed from base models with different network structures according to the voting ensemble algorithm. Our original personal identification dataset is employed as the numerical example to illustrate the performance of the proposed method. The results show that the prediction accuracy of the ensemble model is better than that of the base models and that the prediction of walking behavior toward the Kinect at 90 degrees and 225 degrees is more difficult than other walking behaviors.

This article proposes an algorithm to recommend apposite ID photos for users by judging the photo of which the facial expression is apposite or not as the ID photo. Microsoft’s Kinect sensor is used for taking photos. Parts of the face, such as eyes, nose, and mouth, are analyzed as explanatory variables for judging face expression. Some body coordinate information such as head and shoulders is used to trim the photos. Neural networks and support vector machines are employed and compared to our proposed method. To achieve accurate results, ten examinees including specialized staff are selected for taking ID photo used for training models. A series of experiments are conducted to examine the validity. As a result, the accuracy of neural networks is better than that of the support vector machine. Furthermore, we analyze and discuss the difference between system results and specialized staffs’ opinions.

This article focuses on a new approach for personal identification by exploring the features of pedestrian behavior. The recent progress of a motion capture sensor system enables personal identification using human behavioral data observed from the sensor. Kinect is a motion sensing input device developed by Microsoft for Xbox 360 and Xbox One. Personal identification using the Microsoft Kinect sensor (hereafter referred to as Kinect) is presented in this study. Kinect is used to estimate body sizes and the walking behaviors of pedestrians. Body sizes such as height and width, and walking behavior such as joint angles and stride lengths, for example, are used as explanatory variables for personal identification. An algorithm for the personal identification of pedestrians is defined by a traditional neural network and by a support vector machine. In the numerical experiments, pictures of body sizes and the walking behaviors are captured from fifteen examinees through Kinect. The walking direction of pedestrians was specified as 0°, 90°, 180°, and 225°, and then the accuracies were compared. The results indicate that identification accuracy was best when the walking direction was 180°. In addition, the accuracy of the vector machine was better than that of the neural network.

This article proposes a new approach to personal authentication by exploring the features of a person’s face and voice. Microsoft’s Kinect sensor is used for facial and voice recognition. Parts of the face including the eyes, nose, and mouth, etc., are analyzed as position vectors. For voice recognition, a Kinect microphone array is adopted to record personal voices. Mel-frequency cepstrum coefficients, logarithmic power, and related values involved in the analysis of personal voice are also estimated from the voices. Neural networks,support vector machines and principal components analysis are employed and compared for personal authentication. To achieve accurate results, 20 examinees were selected for face and voice data used for training the authentication models. The experimental results show that the best accuracy is achieved when the model is trained by a support vector machine using both facial and voice features.