Despite the rapid development of deep learning-based intelligent fault diagnosis methods on rotating machinery, the data-driven approach generally remains a "black box" to researchers, and its internal mechanism has not been sufficiently understood. The weak interpretability significantly impedes further development and applications of the effective deep neural network-based methods. This paper contributes efforts to understanding the mechanical signal processing of deep learning on the fault diagnosis problems. The diagnostic knowledge learned by the deep neural network is visualized using the neuron activation maximization and the saliency map methods. The discriminative features of different machine health conditions are intuitively observed. The relationship between the data-driven methods and the well-established conventional fault diagnosis knowledge is confirmed by the experimental investigations on two datasets. The results of this study can benefit researchers on understanding the complex neural networks, and increase the reliability of the data-driven fault diagnosis model in the real engineering cases.

In traditional image segmentation, the GrabCut image segmentation algorithm is a popular and effective method. The current GrabCut image segmentation algorithm is based on the Gaussian mixture model of the global foreground and background of the image. Still, it cannot achieve good results when the foreground and background are similar. In this paper, we propose a method for local sampling of the foreground and background. This method samples the foreground and background around unknown pixels based on the distance. It has an advantage when the foreground and background are similar. The experimental results show that the GrabCut method with local sampling can achieve good results when many colors appear in the foreground and background at the same time.

No abstract

No abstract