Deep learning, especially through convolutional neural networks (CNN) such as the U-Net 3D model, has revolutionized fault identification from seismic data, representing a significant leap over traditional methods. Ou...Deep learning, especially through convolutional neural networks (CNN) such as the U-Net 3D model, has revolutionized fault identification from seismic data, representing a significant leap over traditional methods. Our review traces the evolution of CNN, emphasizing the adaptation and capabilities of the U-Net 3D model in automating seismic fault delineation with unprecedented accuracy. We find: 1) The transition from basic neural networks to sophisticated CNN has enabled remarkable advancements in image recognition, which are directly applicable to analyzing seismic data. The U-Net 3D model, with its innovative architecture, exemplifies this progress by providing a method for detailed and accurate fault detection with reduced manual interpretation bias. 2) The U-Net 3D model has demonstrated its superiority over traditional fault identification methods in several key areas: it has enhanced interpretation accuracy, increased operational efficiency, and reduced the subjectivity of manual methods. 3) Despite these achievements, challenges such as the need for effective data preprocessing, acquisition of high-quality annotated datasets, and achieving model generalization across different geological conditions remain. Future research should therefore focus on developing more complex network architectures and innovative training strategies to refine fault identification performance further. Our findings confirm the transformative potential of deep learning, particularly CNN like the U-Net 3D model, in geosciences, advocating for its broader integration to revolutionize geological exploration and seismic analysis.展开更多
With the increasing demand of computational power in artificial intelligence(AI)algorithms,dedicated accelerators have become a necessity.However,the complexity of hardware architectures,vast design search space,and c...With the increasing demand of computational power in artificial intelligence(AI)algorithms,dedicated accelerators have become a necessity.However,the complexity of hardware architectures,vast design search space,and complex tasks of accelerators have posed significant challenges.Tra-ditional search methods can become prohibitively slow if the search space continues to be expanded.A design space exploration(DSE)method is proposed based on transfer learning,which reduces the time for repeated training and uses multi-task models for different tasks on the same processor.The proposed method accurately predicts the latency and energy consumption associated with neural net-work accelerator design parameters,enabling faster identification of optimal outcomes compared with traditional methods.And compared with other DSE methods by using multilayer perceptron(MLP),the required training time is shorter.Comparative experiments with other methods demonstrate that the proposed method improves the efficiency of DSE without compromising the accuracy of the re-sults.展开更多
eight planets,various asteroids and comets in the solar system.Amount of deep-space scientific experiments promoted people to understand about the origin and evolution of the universe.With the rapid developments of eq...eight planets,various asteroids and comets in the solar system.Amount of deep-space scientific experiments promoted people to understand about the origin and evolution of the universe.With the rapid developments of equipment and spacecraft with high-accuracy detector and long-term energy,more and more ambitious deep-space exploration plans have also been scheduled or under discussion about space resources utilization and space migration,e.g.,manned landing on the Mars,guard infrastructures on the Moon and human-flight to the edge of the solar system(>100 AU),etc.展开更多
The ability of achieving a semantic understanding of workspaces is an important capability for mobile robot. A method is proposed to categorize different places in a typical indoor environment by using a Kinect sensor...The ability of achieving a semantic understanding of workspaces is an important capability for mobile robot. A method is proposed to categorize different places in a typical indoor environment by using a Kinect sensors for mobile robot exploration. At first,the invariant feature based images stitching approach is adopted to form a panoramic image according to Kinect visual information,and the translation between Kinect depth information and obstacle distance information is performed to obtain virtual LIDAR data. Then,the semantic classifier is designed by using convolutional neural networks( CNN) for indoor place categorization based on Kinect visual observations with panoramic view. At last,a frontier-based exploration method is applied to carry out indoor autonomous exploration of mobile robots,which integrates the CNN-based categorization approach. The proposed method has been implemented and tested on a real robot,and experiment results demonstrate the approach effectiveness on solving the semantic categorization problem for mobile robot exploration.展开更多
文摘Deep learning, especially through convolutional neural networks (CNN) such as the U-Net 3D model, has revolutionized fault identification from seismic data, representing a significant leap over traditional methods. Our review traces the evolution of CNN, emphasizing the adaptation and capabilities of the U-Net 3D model in automating seismic fault delineation with unprecedented accuracy. We find: 1) The transition from basic neural networks to sophisticated CNN has enabled remarkable advancements in image recognition, which are directly applicable to analyzing seismic data. The U-Net 3D model, with its innovative architecture, exemplifies this progress by providing a method for detailed and accurate fault detection with reduced manual interpretation bias. 2) The U-Net 3D model has demonstrated its superiority over traditional fault identification methods in several key areas: it has enhanced interpretation accuracy, increased operational efficiency, and reduced the subjectivity of manual methods. 3) Despite these achievements, challenges such as the need for effective data preprocessing, acquisition of high-quality annotated datasets, and achieving model generalization across different geological conditions remain. Future research should therefore focus on developing more complex network architectures and innovative training strategies to refine fault identification performance further. Our findings confirm the transformative potential of deep learning, particularly CNN like the U-Net 3D model, in geosciences, advocating for its broader integration to revolutionize geological exploration and seismic analysis.
基金the National Key R&D Program of China(No.2018AAA0103300)the National Natural Science Foundation of China(No.61925208,U20A20227,U22A2028)+1 种基金the Chinese Academy of Sciences Project for Young Scientists in Basic Research(No.YSBR-029)the Youth Innovation Promotion Association Chinese Academy of Sciences.
文摘With the increasing demand of computational power in artificial intelligence(AI)algorithms,dedicated accelerators have become a necessity.However,the complexity of hardware architectures,vast design search space,and complex tasks of accelerators have posed significant challenges.Tra-ditional search methods can become prohibitively slow if the search space continues to be expanded.A design space exploration(DSE)method is proposed based on transfer learning,which reduces the time for repeated training and uses multi-task models for different tasks on the same processor.The proposed method accurately predicts the latency and energy consumption associated with neural net-work accelerator design parameters,enabling faster identification of optimal outcomes compared with traditional methods.And compared with other DSE methods by using multilayer perceptron(MLP),the required training time is shorter.Comparative experiments with other methods demonstrate that the proposed method improves the efficiency of DSE without compromising the accuracy of the re-sults.
文摘eight planets,various asteroids and comets in the solar system.Amount of deep-space scientific experiments promoted people to understand about the origin and evolution of the universe.With the rapid developments of equipment and spacecraft with high-accuracy detector and long-term energy,more and more ambitious deep-space exploration plans have also been scheduled or under discussion about space resources utilization and space migration,e.g.,manned landing on the Mars,guard infrastructures on the Moon and human-flight to the edge of the solar system(>100 AU),etc.
基金Supported by the National Key Basic Research Program of China(No.2013CB035503)
文摘The ability of achieving a semantic understanding of workspaces is an important capability for mobile robot. A method is proposed to categorize different places in a typical indoor environment by using a Kinect sensors for mobile robot exploration. At first,the invariant feature based images stitching approach is adopted to form a panoramic image according to Kinect visual information,and the translation between Kinect depth information and obstacle distance information is performed to obtain virtual LIDAR data. Then,the semantic classifier is designed by using convolutional neural networks( CNN) for indoor place categorization based on Kinect visual observations with panoramic view. At last,a frontier-based exploration method is applied to carry out indoor autonomous exploration of mobile robots,which integrates the CNN-based categorization approach. The proposed method has been implemented and tested on a real robot,and experiment results demonstrate the approach effectiveness on solving the semantic categorization problem for mobile robot exploration.
