An intelligent lithology identification method is proposed based on deep learning of the rock microscopic images.Based on the characteristics of rock images in the dataset,we used Xception,MobileNet_v2,Inception_ResNe...An intelligent lithology identification method is proposed based on deep learning of the rock microscopic images.Based on the characteristics of rock images in the dataset,we used Xception,MobileNet_v2,Inception_ResNet_v2,Inception_v3,Densenet121,ResNet101_v2,and ResNet-101 to develop microscopic image classification models,and then the network structures of seven different convolutional neural networks(CNNs)were compared.It shows that the multi-layer representation of rock features can be represented through convolution structures,thus better feature robustness can be achieved.For the loss function,cross-entropy is used to back propagate the weight parameters layer by layer,and the accuracy of the network is improved by frequent iterative training.We expanded a self-built dataset by using transfer learning and data augmentation.Next,accuracy(acc)and frames per second(fps)were used as the evaluation indexes to assess the accuracy and speed of model identification.The results show that the Xception-based model has the optimum performance,with an accuracy of 97.66%in the training dataset and 98.65%in the testing dataset.Furthermore,the fps of the model is 50.76,and the model is feasible to deploy under different hardware conditions and meets the requirements of rapid lithology identification.This proposed method is proved to be robust and versatile in generalization performance,and it is suitable for both geologists and engineers to identify lithology quickly.展开更多
Optical vortices(OVs) with unique square symmetry are widely used in various applications including particle manipulation,microscopy, and image processing. However, the undesired higher-order foci introduced by the co...Optical vortices(OVs) with unique square symmetry are widely used in various applications including particle manipulation,microscopy, and image processing. However, the undesired higher-order foci introduced by the conventional vortex lens such as square spiral zone plates(SSZPs) may lead to additional artifacts and thus degrade contrast sensitivity. In this endeavor, herein,we propose a methodology to combine the merit of SSZPs and the advantage of Gabor zone plates(GZPs) in establishing a specific single optical element, termed binary single focused square spiral zone plates(BSSZPs). In contrast to the abrupt transitions of the SSZPs, our central idea aims to realize the sinusoidal transmittance along the radial direction of SSZPs by a series of randomly distributed annulus-quadrangle-shaped nanometer structure apertures. The innovative design can simultaneously generate OVs with unique square symmetry, and eliminate the interference of higher-order foci along the propagation direction. Guided by our theoretical predication, the focusing property of such optics was further experimentally demonstrated.These findings are expected to direct new avenue towards improving the performance of optical image processing and alignment system.展开更多
基金support from the National Natural Science Foundation of China(Grant Nos.52022053 and 52009073)the Natural Science Foundation of Shandong Province(Grant No.ZR201910270116).
文摘An intelligent lithology identification method is proposed based on deep learning of the rock microscopic images.Based on the characteristics of rock images in the dataset,we used Xception,MobileNet_v2,Inception_ResNet_v2,Inception_v3,Densenet121,ResNet101_v2,and ResNet-101 to develop microscopic image classification models,and then the network structures of seven different convolutional neural networks(CNNs)were compared.It shows that the multi-layer representation of rock features can be represented through convolution structures,thus better feature robustness can be achieved.For the loss function,cross-entropy is used to back propagate the weight parameters layer by layer,and the accuracy of the network is improved by frequent iterative training.We expanded a self-built dataset by using transfer learning and data augmentation.Next,accuracy(acc)and frames per second(fps)were used as the evaluation indexes to assess the accuracy and speed of model identification.The results show that the Xception-based model has the optimum performance,with an accuracy of 97.66%in the training dataset and 98.65%in the testing dataset.Furthermore,the fps of the model is 50.76,and the model is feasible to deploy under different hardware conditions and meets the requirements of rapid lithology identification.This proposed method is proved to be robust and versatile in generalization performance,and it is suitable for both geologists and engineers to identify lithology quickly.
基金supported by the National Key Research and Development Program of China (Grant No. 2021YFA1400204)National Natural Science Foundation of China (Grant Nos. 12174350, 11905200, 12105268,11805179, 11905201, and 12174347)+2 种基金Program for Science&Technology Innovation Talents of Henan Province (Grant No. 202102310001)Science and Technology on Plasma Physics Laboratory (Grant No. 6142A04200107)Excellent Youth Foundation of Henan Scientific Committee (Grant No. 202300410356)。
文摘Optical vortices(OVs) with unique square symmetry are widely used in various applications including particle manipulation,microscopy, and image processing. However, the undesired higher-order foci introduced by the conventional vortex lens such as square spiral zone plates(SSZPs) may lead to additional artifacts and thus degrade contrast sensitivity. In this endeavor, herein,we propose a methodology to combine the merit of SSZPs and the advantage of Gabor zone plates(GZPs) in establishing a specific single optical element, termed binary single focused square spiral zone plates(BSSZPs). In contrast to the abrupt transitions of the SSZPs, our central idea aims to realize the sinusoidal transmittance along the radial direction of SSZPs by a series of randomly distributed annulus-quadrangle-shaped nanometer structure apertures. The innovative design can simultaneously generate OVs with unique square symmetry, and eliminate the interference of higher-order foci along the propagation direction. Guided by our theoretical predication, the focusing property of such optics was further experimentally demonstrated.These findings are expected to direct new avenue towards improving the performance of optical image processing and alignment system.
