Aiming at the problems of low accuracy of coal gangue recognition and difficult recognition of mixed gangue rate, a coal rock recognition method based on modal fusion of RGB and infrared is proposed. A fully mechanized coal gangue transportation test bed is built, RGB images are obtained by camera, and infrared images are obtained by industrial microwave heating system and infrared thermal imager. the image data of the whole coal, whole gangue, and coal gangue with different gangue mixing as training and test samples, identify the released coal gangue and its mixing rate. The AlexNet, VGG-16, ResNet-18 classification networks and their convolutional neural networks with modal feature fusion are constructed. results: The classification accuracy of ResNet networks on RGB and infrared image data is higher than AlexNet and VGG-16 networks. The early convergence network performance of ResNet is verified through the convergence of different models. The recognition rate of the network is 97.92 the confusion matrix statistics, which verifies the feasibility of the application of modal fusion method in the field of coal gangue recognition. The fusion of modal features and early models of ResNet coal gangue, which is the basic premise for realizing intelligent coal caving.
The neural network denoising technique has achieved impressive results by being able to automatically learn the effective signal from the data without any assumptions. However, it has been found experimentally that the performance of the method using neural networks gradually decreases with increasing pollution levels when processing contaminated seismic data, and how to improve the performance will become the direction of further development of the method. As a traditional method widely used for tainted seismic data, the wavelet transform can effectively separate the signal from the noise. Thus, we propose a method combining wavelet transform and a residual neural network that achieves good results in suppressing random noise data.
Ground-penetrating radar (GPR) crosshole tomography is widely applied to subsurface media images. However, the inadequacies of ray methods may limit the resolution of crosshole radar images, since the ray method is a type of high-frequency approximation. To solve this problem, the full waveform method is introduced for GPR inversion. However, full waveform inversion is computationally expensive. In this paper, we introduce a trained neural network that can be evaluated very quickly to replace a computationally intensive forward model. Additionally, the forward error of the trained neural network can be statistically analyzed. We demonstrate a methodology for a full waveform inversion of crosshole ground-penetrating radar data using the Markov chain Monte Carlo (MCMC) method. An accurate forward model based on Maxwell’s equations is replaced by a quickly trained neural network. This method achieves a high computation efficiency, which is four orders of magnitude faster than the accurate forward model. The inversion result of the synthetic waveform data shows a good performance of the trained neural network, which greatly improves the calculation efficiency.
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.