Exposure to respirable coal mine dust(RCMD)can cause chronic and debilitating lung diseases.Real-time monitoring capabilities are sought which can enable a better understanding of dust components and sources.In many u...Exposure to respirable coal mine dust(RCMD)can cause chronic and debilitating lung diseases.Real-time monitoring capabilities are sought which can enable a better understanding of dust components and sources.In many underground mines,RCMD includes three primary components which can be loosely associated with three major dust sources:coal dust from the coal seam itself,silicates from the surrounding rock strata,and carbonates from the inert‘rock dust’products that are applied to mitigate explosion hazards.A monitor which can reliably partition RCMD between these three components could thus allow source apportionment.And tracking silicates,specifically,could be valuable since the most serious health risks are typically associated with this component-particularly if abundant in crystalline silica.Envisioning a monitoring concept based on field microscopy,and following up on prior research using polarized light,the aim of the current study was to build and test a model to classify respirable-sized particles as either coal,silicates,or carbonates.For model development,composite dust samples were generated in the laboratory by successively depositing dust from high-purity materials onto a sticky transparent substrate,and imaging after each deposition event such that the identity of each particle was known a priori.Model testing followed a similar approach,except that real geologic materials were used as the source for each dust component.Results showed that the model had an overall accuracy of 86.5%,indicating that a field-microscopy based moni-tor could support RCMD source apportionment and silicates tracking in some coal mines.展开更多
A mathematical model of quantum noise having much effect on the low light imaging system is set up. To simulate the quantum noise, the random numbers obeying noise distribution must be formed and are weighted on the...A mathematical model of quantum noise having much effect on the low light imaging system is set up. To simulate the quantum noise, the random numbers obeying noise distribution must be formed and are weighted on the basis of the model created. Three uniform random sequences are built by the linear congruential method, of which two are used to form integer number and decimal fraction parts of the new random sequence respectively and the third to shuffle the new sequence. And then a Gauss sequence is formed out of uniform distribution by a function transforming method. It actualizes the simulation in real time of quantum noise in the low light imaging system, where video flow is extracted in real time, the noise summed up and played back side by side with the original video signs by a simulation software.展开更多
A method based on multiple images captured under different light sources at different incident angles was developed to recognize the coal density range in this study.The innovation is that two new images were construc...A method based on multiple images captured under different light sources at different incident angles was developed to recognize the coal density range in this study.The innovation is that two new images were constructed based on images captured under four single light sources.Reconstruction image 1 was constructed by fusing greyscale versions of the original images into one image,and Reconstruction image2 was constructed based on the differences between the images captured under the different light sources.Subsequently,the four original images and two reconstructed images were input into the convolutional neural network AlexNet to recognize the density range in three cases:-1.5(clean coal) and+1.5 g/cm^(3)(non-clean coal);-1.8(non-gangue) and+1.8 g/cm^(3)(gangue);-1.5(clean coal),1.5-1.8(middlings),and+1.8 g/cm^(3)(gangue).The results show the following:(1) The reconstructed images,especially Reconstruction image 2,can effectively improve the recognition accuracy for the coal density range compared with images captured under single light source.(2) The recognition accuracies for gangue and non-gangue,clean coal and non-clean coal,and clean coal,middlings,and gangue reached88.44%,86.72% and 77.08%,respectively.(3) The recognition accuracy increases as the density moves further away from the boundary density.展开更多
A novel frame shift and integral technique for the enhancement of low light level moving image sequence is introduced. According to the technique, motion parameters of target are measured by algorithm based on differe...A novel frame shift and integral technique for the enhancement of low light level moving image sequence is introduced. According to the technique, motion parameters of target are measured by algorithm based on difference processing. To obtain spatial relativity, images are shifted according to the motion parameters. As a result, the processing of integral and average can be applied to images that have been shifted. The technique of frame shift and integral that includes the algorithm of motion parameter determination is discussed, experiments with low light level moving image sequences are also described. The experiment results show the effectiveness and the robustness of the parameter determination algorithm, and the improvement in the signal-to-noise ratio （SNR） of low light level moving images.展开更多
基金supported by the Alpha Foundation for the Improvement of Mine Safety and Health,grant number AFC316FO-84.
文摘Exposure to respirable coal mine dust(RCMD)can cause chronic and debilitating lung diseases.Real-time monitoring capabilities are sought which can enable a better understanding of dust components and sources.In many underground mines,RCMD includes three primary components which can be loosely associated with three major dust sources:coal dust from the coal seam itself,silicates from the surrounding rock strata,and carbonates from the inert‘rock dust’products that are applied to mitigate explosion hazards.A monitor which can reliably partition RCMD between these three components could thus allow source apportionment.And tracking silicates,specifically,could be valuable since the most serious health risks are typically associated with this component-particularly if abundant in crystalline silica.Envisioning a monitoring concept based on field microscopy,and following up on prior research using polarized light,the aim of the current study was to build and test a model to classify respirable-sized particles as either coal,silicates,or carbonates.For model development,composite dust samples were generated in the laboratory by successively depositing dust from high-purity materials onto a sticky transparent substrate,and imaging after each deposition event such that the identity of each particle was known a priori.Model testing followed a similar approach,except that real geologic materials were used as the source for each dust component.Results showed that the model had an overall accuracy of 86.5%,indicating that a field-microscopy based moni-tor could support RCMD source apportionment and silicates tracking in some coal mines.
文摘A mathematical model of quantum noise having much effect on the low light imaging system is set up. To simulate the quantum noise, the random numbers obeying noise distribution must be formed and are weighted on the basis of the model created. Three uniform random sequences are built by the linear congruential method, of which two are used to form integer number and decimal fraction parts of the new random sequence respectively and the third to shuffle the new sequence. And then a Gauss sequence is formed out of uniform distribution by a function transforming method. It actualizes the simulation in real time of quantum noise in the low light imaging system, where video flow is extracted in real time, the noise summed up and played back side by side with the original video signs by a simulation software.
文摘A method based on multiple images captured under different light sources at different incident angles was developed to recognize the coal density range in this study.The innovation is that two new images were constructed based on images captured under four single light sources.Reconstruction image 1 was constructed by fusing greyscale versions of the original images into one image,and Reconstruction image2 was constructed based on the differences between the images captured under the different light sources.Subsequently,the four original images and two reconstructed images were input into the convolutional neural network AlexNet to recognize the density range in three cases:-1.5(clean coal) and+1.5 g/cm^(3)(non-clean coal);-1.8(non-gangue) and+1.8 g/cm^(3)(gangue);-1.5(clean coal),1.5-1.8(middlings),and+1.8 g/cm^(3)(gangue).The results show the following:(1) The reconstructed images,especially Reconstruction image 2,can effectively improve the recognition accuracy for the coal density range compared with images captured under single light source.(2) The recognition accuracies for gangue and non-gangue,clean coal and non-clean coal,and clean coal,middlings,and gangue reached88.44%,86.72% and 77.08%,respectively.(3) The recognition accuracy increases as the density moves further away from the boundary density.
文摘A novel frame shift and integral technique for the enhancement of low light level moving image sequence is introduced. According to the technique, motion parameters of target are measured by algorithm based on difference processing. To obtain spatial relativity, images are shifted according to the motion parameters. As a result, the processing of integral and average can be applied to images that have been shifted. The technique of frame shift and integral that includes the algorithm of motion parameter determination is discussed, experiments with low light level moving image sequences are also described. The experiment results show the effectiveness and the robustness of the parameter determination algorithm, and the improvement in the signal-to-noise ratio （SNR） of low light level moving images.
文摘由于低照度图像具有对比度低、细节丢失严重、噪声大等缺点,现有的目标检测算法对低照度图像的检测效果不理想.为此,本文提出一种结合空间感知注意力机制和多尺度特征融合(Spatial-aware Attention Mechanism and Multi-Scale Feature Fusion,SAM-MSFF)的低照度目标检测方法 .该方法首先通过多尺度交互内存金字塔融合多尺度特征,增强低照度图像特征中的有效信息,并设置内存向量存储样本的特征,捕获样本之间的潜在关联性;然后,引入空间感知注意力机制获取特征在空间域的长距离上下文信息和局部信息,从而增强低照度图像中的目标特征,抑制背景信息和噪声的干扰;最后,利用多感受野增强模块扩张特征的感受野,对具有不同感受野的特征进行分组重加权计算,使检测网络根据输入的多尺度信息自适应地调整感受野的大小.在ExDark数据集上进行实验,本文方法的平均精度(mean Average Precision,mAP)达到77.04%,比现有的主流目标检测方法提高2.6%~14.34%.
