Search

The algorithm based on KPDA and SVM is proposed. Firstly, gait energy image (GEI) and moment gait energy images (MGEI) are combined for expressing objects and features reduction. Then the low-dimensional gait characteristic is extracted by KFDA, which can obtain the best projection direction and enhance the capacity of data classification. Then the support vector machine (SVM) models are trained by the decomposed feature vectors. The gaits are classified by the trained SVM models. This algorithm is applied to a data-set including thirty individuals. Extensive experimental results demonstrate that the proposed algorithm performs at an encouraging recognition rate of 91% and at a relatively lower computational cost.

The algorithm based on KFDA and SVM is proposed. The information of width and angle is extracted from the human motion image sequences. The features of width and angle is merged and reduced by the KPCA. The Low-dimensional gait characteristic is extracted by modified KFDA, which can obtain the best projection direction and enhance the capacity of data classification. Then the support vector machine (SVM) models are trained by the decomposed feature vectors. The gaits are classified by the trained SVM models. This algorithm is applied to a data-set including thirty individuals. Extensive experimental results demonstrate that the proposed algorithm performs at an encouraging recognition rate of 92% and at a relatively lower computational cost.

The gait recognition algorithm adopt support vector machine based on hybrid kernel function and Parameter Optimization. Partial kernel function and overall kernel function are fitted to compose super-kernel function, so that the SVM obtain better generalization ability and generalization ability. In terms of parameter selection, the text uses the objective function and combine OPS algorithm to select the best kernel parameter. This method makes use of the distance of training samples of different classes to find the optimal (or effective) nuclear parameters instead of the standard SVM training samples. It avoids strong empirical and large amount of calculation of the traditional SVM on model selection. Then the gaits are classified by the support vector machine models. This algorithm is applied to a data-set including thirty individuals. Experimental results demonstrate that the algorithm performs at an encouraging recognition rate and at a relatively lower computational cost.

For batch processes, a complete batch trajectory can be obtained until the end of its operation. Besides, there are not enough fault data used to build models in historical dataset. Those make it very difficult to diagnose faults for batch processes based on Fisher discriminant analysis (FDA). In order to solve those problems a multi-model Fisher discriminant analysis with moving window is represented. Using moving window, the date used to build FDA model are made of the data of the current time and past time. Therefore, the proposed method not only sufficiently utilizes the finite fault data, but also overcomes the need of estimated or filled up the future unmeasured values in online fault diagnosis. An industrial typical streptomycin fermentation process is used to test the performance of fault diagnosis of the proposed method