Blind image deconvolution is an ill-posed problem, which is mainly addressed by the regularization methods. Wavelet transform is an effective denoising method related to regularized inversion. In this paper, wavelet transform is utilized to decompose and extract the low and high frequency information of the blurred image, which is taken as the first step of the presented deblurring methods in this paper. However, when the process highlights the approximate portion of the image feature, the blurry image will be smoothed so that the image is distorted. Simultaneously, the overall wavelet transform will result in excessive data redundancy for the image. Thus in the second step, based upon the recursive convolution neural network, a deep recursive convolutional neural network (R-DbCNN) is designed, which can eliminate or weaken the characteristics of high data redundancy and image smoothness caused by wavelet transform to remove the blur of corrupted image. Comparing with the traditional convolution neural network in image restoration, R-DbCNN has better performance in deblurring with fast training speed. Thereafter a novel loss function is built to attain the best deblurring effect of the proposed method based on L2 regularization term. The experiment results demonstrate that our method has practical applicability for image restoration with different blurs, such as the motion blur, Gaussian blur and out-of-focus by camera.

Non–intrusive load monitoring based on power measurements is a promising topic of appliance identification in the research of smart grid; where the key is to avoid the power sub-item measurement in load monitoring. In this paper; a three–step non–intrusive load monitoring system (TNILM) is proposed. Firstly; a one dimension convolution neural network (CNN) is constructed based on the structure of GoogLeNet with 2D convolution; which can zoom in on the differences in features between the different appliances; and then effectively extract various transient features of appliances. Secondly; comparing with various classifiers; the Linear Programming boosting with adaptive weights and thresholds (ALPBoost) is proposed and applied to recognize single–appliance and multiple–appliance. Thirdly; an update process is adopted to adjust and balance the parameters between the one dimension CNN and ALPBoost on–line. The TNILM is tested on a real–world power consumption dataset; which comprises single or multiple appliances potentially operated simultaneously. The experiment result shows the effectiveness of the proposed method in both identification rates.