The efficient separation of coal and gangue in the mining process is of great significance for improving coal mining efficiency and reducing environmental pollution. Automatic detection of coal and gangue is the key and foundation for the separation of coal and gangue. In this paper, we proposed a hierarchical framework for coal and gangue detection based on deep learning models. In this framework, the Gaussian pyramid principle is first used to construct multi-level training data, leading to the sets of coal and gangue image features with multiple scales. Then, the coal and gangue regional proposal networks (CG-RPN) are designed to determine the candidate regions of the target objects in the image. Next, convolution neural networks (CNNs) are constructed to recognize coal and gangue objects on the basis of extracted candidate regions. We performed our method on three different datasets. Experimental results showed that the proposed method improves the detection accuracy of coal and gangue objects by 0.8% compared with the previous methods, reaching up to 98.33%. In addition, our proposed method enables the detection of multiple coal and gangue objects in an individual image and solves the problem of queuing requirements in traditional methods.INDEX TERMS Coal and gangue, detection, deep learning, CNNs, multi-level training data.
Rolling bearings are important components of rotating machines. For their preventive maintenance, it is not enough to know whether there is any fault or the fault type. For an effective maintenance, a fault severity monitoring needs to be conducted. Currently, the bearing fault diagnosis method based on time–frequency image (TFI) recognition is attracting increasing attention. This paper contributes to the ongoing investigation by proposing a new approach for the fault severity monitoring of rolling bearings based on the texture feature extraction of sparse TFIs. The first and main step is to obtain accurate TFIs from the vibration signals of rolling bearings. Traditional time–frequency analysis methods have disadvantages such as low resolution and cross-term interference. Therefore, the TFIs obtained cannot satisfactorily express the time–frequency characteristics of bearing vibration signals. To solve this problem, a sparse time–frequency analysis method based on the first-order primal-dual algorithm (STFA-PD) was developed in this paper. Unlike traditional time–frequency analysis methods, the time–frequency analysis model of the STFA-PD method is based on the theory of sparse representation, and is solved using the first-order primal-dual algorithm. For employing the sparse constraint in the frequency domain, the STFA-PD obtains a higher time–frequency resolution and is free from cross-term interference, as the model is based on a linear time–frequency analysis method. The gray level co-occurrence matrix is then employed to extract texture features from the sparse TFIs as input features for classifiers. Vibration signals of rolling bearings with different fault severity degrees are used to validate the proposed approach. The experimental results show that the developed STFA-PD outperforms traditional time–frequency analysis methods in terms of the accuracy and effectiveness for the fault severity monitoring of rolling bearings.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.