Search

In response to the problem of low recognition rate caused by low quality face image in face recognition, a face quality evaluation method based on deep convolutional neural network is proposed. Firstly, the depthwise separable convolution method is adopted to establish a depth-convolution model which contain 8 blocks to achieve face feature extraction. Secondly, the last layer of the convolution module connects 8 output branches, and uses regression and classification methods to predict the probability of 8 attributes, namely yaw, pitch, norm, opening mouth, occlusion, blur, dim, and closed eyes. The proposed method also proves that norm value can help analyze the quality of face image. Finally, the weight optimization is realized by taking the video pass rate as the optimal objective function. the quality score of face image can be calculated by weighting each branch The higher the score, the better the quality of the face image. In this paper, an end-to-end face quality evaluation model is implemented by using depthwise separable convolution method. The parameters of the model are less than 60000, the operation speed is fast, and the evaluation results are accurate; and it can filter out low-quality face image in real time, and recommend high-quality face image to face recognition model.

Fuzzy control is a human-imitating control technique which is independent of the mathematical model of plants. It utilizes priori knowledge to carry out approximate reasoning. However, it lacks the abilities of self-tuning or self-learning in industrial applications. The temperature control process of an oil-burning tunnel kiln is a multivariable and nonlinear dynamic system. This paper presents a fuzzy neural network control strategy which is able to enhance the capacity of self-learning of fuzzy control rules, based on the self-learning ability of neural networks. Simulation research and a physical analog experiment prove the feasibility of this control strategy.

A fuzzy-neuro network architecture is proposed. An initial fuzzy knowledge base can be built by means of self-adaptive learning from historical data, fuzzy rules and system parameters can be optimized by online learning of the FNN via real-time data, which make the system possess distinguished adaptive features and self-learning capability. Taking fire auto-detecting in a fire protection system as application background, a series of experimental research has been carried out. Experimental results demonstrate the feasibility of the neuro-fuzzy system.

A fuzzy neural network control strategy is presented. Fuzzy rules are optimized when the strategy is used to control side-wind for chemical fiber which require constant temperature, constant humidity and constant air velocity. The feature of the algorithm is introduced, and this method is proved feasible by simulation.