In the strapdown inertial navigation system,the attitude information is obtained through an inertial measurement unit(IMU)device,which mainly includes a triaxial gyroscope,a triaxial accelerometer and a triaxial magne...In the strapdown inertial navigation system,the attitude information is obtained through an inertial measurement unit(IMU)device,which mainly includes a triaxial gyroscope,a triaxial accelerometer and a triaxial magnetometer.However,IMU sensors have system noise and drift errors,and these errors can accumulate over time,which makes it difficult to control the attitude accuracy.In order to solve the problems of gyro drift over time and random errors generated by the surrounding environment,this paper presents an attitude calculation algorithm based on wavelet neural network-extended Kalman filter(WNN-EKF).The wavelet neural network(WNN)is used to optimize the model and compensate the extended Kalman filter’s own model error.Through the semi-physical simulation experiment,the results show that the algorithm improves the accuracy of attitude calculation and enhances the self-adaptability to the environment.展开更多
Head pose estimation has been considered an important and challenging task in computer vision. In this paper we propose a novel method to estimate head pose based on a deep convolutional neural network （DCNN） for 2D...Head pose estimation has been considered an important and challenging task in computer vision. In this paper we propose a novel method to estimate head pose based on a deep convolutional neural network （DCNN） for 2D face images. We design an effective and simple method to roughly crop the face from the input image, maintaining the individual-relative facial features ratio. The method can be used in various poses. Then two convolutional neural networks are set up to train the head pose classifier and then compared with each other. The simpler one has six layers. It performs well on seven yaw poses but is somewhat unsatisfactory when mixed in two pitch poses. The other has eight layers and more pixels in input layers. It has better performance on more poses and more training samples. Before training the network, two reasonable strategies including shift and zoom are executed to prepare training samples. Finally, feature extraction filters are optimized together with the weight of the classification component through training, to minimize the classification error. Our method has been evaluated on the CAS-PEAL-R1, CMU PIE, and CUBIC FacePix databases. It has better performance than state-of-the-art methods for head pose estimation.展开更多
基金National Natural Science Foundation of China(No.61863024)Basic Research Innovation Group Program of Gansu Province(No.1606RJIA327)+2 种基金Higher Education Research Project Funding of Gansu Province(No.2018C-11)Natural Foundation of Gansu Province(No.18JR3RA107)Science and Technology Program Funding of Gansu Province(No.18CX3ZA004)。
文摘In the strapdown inertial navigation system,the attitude information is obtained through an inertial measurement unit(IMU)device,which mainly includes a triaxial gyroscope,a triaxial accelerometer and a triaxial magnetometer.However,IMU sensors have system noise and drift errors,and these errors can accumulate over time,which makes it difficult to control the attitude accuracy.In order to solve the problems of gyro drift over time and random errors generated by the surrounding environment,this paper presents an attitude calculation algorithm based on wavelet neural network-extended Kalman filter(WNN-EKF).The wavelet neural network(WNN)is used to optimize the model and compensate the extended Kalman filter’s own model error.Through the semi-physical simulation experiment,the results show that the algorithm improves the accuracy of attitude calculation and enhances the self-adaptability to the environment.
基金Project supported by the National Key Scientific Instrument and Equipment Development Project of China(No.2013YQ49087903)the National Natural Science Foundation of China(No.61402307)the Educational Commission of Sichuan Province,China(No.15ZA0007)
文摘Head pose estimation has been considered an important and challenging task in computer vision. In this paper we propose a novel method to estimate head pose based on a deep convolutional neural network （DCNN） for 2D face images. We design an effective and simple method to roughly crop the face from the input image, maintaining the individual-relative facial features ratio. The method can be used in various poses. Then two convolutional neural networks are set up to train the head pose classifier and then compared with each other. The simpler one has six layers. It performs well on seven yaw poses but is somewhat unsatisfactory when mixed in two pitch poses. The other has eight layers and more pixels in input layers. It has better performance on more poses and more training samples. Before training the network, two reasonable strategies including shift and zoom are executed to prepare training samples. Finally, feature extraction filters are optimized together with the weight of the classification component through training, to minimize the classification error. Our method has been evaluated on the CAS-PEAL-R1, CMU PIE, and CUBIC FacePix databases. It has better performance than state-of-the-art methods for head pose estimation.
