Convolutional neural networks have been widely used for analyzing image data in industry,especially in the oil and gas area.Brazil has an extensive hydrocarbon reserve on its coast and has also benefited from these ne...Convolutional neural networks have been widely used for analyzing image data in industry,especially in the oil and gas area.Brazil has an extensive hydrocarbon reserve on its coast and has also benefited from these neural network models.Image data from petrographic thin section can be essential to provide information about reservoir quality,highlighting important features such as carbonate lithology.However,the automatic identification of lithology in reservoir rocks is still a significant challenge,mainly due to the heterogeneity that is part of the lithologies of the Brazilian pre-salt.Within this context,this work presents an approach using one-class or specialist models to identify four classes of lithology present in reservoir rocks in the Brazilian pre-salt.The proposed methodology had the challenge of dealing with a small number of images for training the neural networks,in addition to the complexity involved in the analyzed data.An auto-machine learning tool called AutoKeras was used to define the hyperparameters of the implemented models.The results found were satisfactory and presented an accuracy greater than 70%for image samples belonging to other wells not seen during the model building,which increases the applicability of the implemented model.Finally,a comparison was made between the proposed methodology and multiple-class models,demonstrating the superiority of one-class models.展开更多
COMPUTATIONAL knowledge vision[1]is emphasized as a novel perspective or field in this paper.It first proposes the visual hierarchy and its connection to knowledge,stating that knowledge is a justified true belief.To ...COMPUTATIONAL knowledge vision[1]is emphasized as a novel perspective or field in this paper.It first proposes the visual hierarchy and its connection to knowledge,stating that knowledge is a justified true belief.To further the previous research,we concisely summarize our recent works and suggest a new direction that knowledge is also a thought framework in vision.展开更多
Visual object recognition in humans and nonhuman primates is achieved by the ventral visual pathway(ventral occipital-temporal cortex,VOTC),which shows a well-documented object domain structure.An on-going question is...Visual object recognition in humans and nonhuman primates is achieved by the ventral visual pathway(ventral occipital-temporal cortex,VOTC),which shows a well-documented object domain structure.An on-going question is what type of information is processed in the higher-order VOTC that underlies such observations,with recent evidence suggesting effects of certain visual features.Combining computational vision models,fMRI experiment using a parametric-modulation approach,and natural image statistics of common objects,we depicted the neural distribution of a comprehensive set of visual features in the VOTC,identifying voxel sensitivities with specific feature sets across geometry/shape,Fourier power,and color.The visual feature combination pattern in the VOTC is significantly explained by their relationships to different types of response-action computation(fight-or-flight,navigation,and manipulation),as derived from behavioral ratings and natural image statistics.These results offer a comprehensive visual feature map in the VOTC and a plausible theoretical explanation as a mapping onto different types of downstream response-action systems.展开更多
Neuromorphic computing vision is the most promising technological solution to overcome the arithmetic bottleneck in machine vision applications.All-in-one neuromorphic sensors have been attracting increased attention ...Neuromorphic computing vision is the most promising technological solution to overcome the arithmetic bottleneck in machine vision applications.All-in-one neuromorphic sensors have been attracting increased attention because they can integrate visual perception,processing,and memory functionalities into one single device.However,the limited responsivity and data retention time of all-in-one neuromorphic sensors usually hinder their potential in multispectral machine vision,especially in the near-infrared(NIR)band which contains critical information for pattern recognition.Here,we demonstrate physisorption-assistant optoelectronic synaptic transistors based on Ta_(2)NiSe_(5)/SnS_(2) heterojunction,which present tunable synaptic functionality in broadband(375–1310 nm).We propose a strategy about the physisorption-assistant persistent photoconductivity(PAPPC)effect to effectively solve the problem in detecting and storing the NIR light information.Under this strategy,the responsivity and data retention time of our devices were significantly enhanced and prolonged in broadband from 375 to 1310 nm.Further,the devices realize multilevel non-volatile optoelectronic memory through the modulation of several optical and back-gate signals to simulate emotion-controlled learning and memory processes,optical writing-electric erasing,and associative learning.Moreover,we developed a simplified human visual system to simulate color-cognitive perception and memory functions.Our approach offers a route for creating advanced all-in-one neuromorphic sensors and developing neuromorphic computing vision.展开更多
基金supported by the R&D cooperation agreement be-tween Petrobras and CBPF(Contract No.0050.0121790.22.9)Brazilian Research Council(CNPq)for the scholarships for students.
文摘Convolutional neural networks have been widely used for analyzing image data in industry,especially in the oil and gas area.Brazil has an extensive hydrocarbon reserve on its coast and has also benefited from these neural network models.Image data from petrographic thin section can be essential to provide information about reservoir quality,highlighting important features such as carbonate lithology.However,the automatic identification of lithology in reservoir rocks is still a significant challenge,mainly due to the heterogeneity that is part of the lithologies of the Brazilian pre-salt.Within this context,this work presents an approach using one-class or specialist models to identify four classes of lithology present in reservoir rocks in the Brazilian pre-salt.The proposed methodology had the challenge of dealing with a small number of images for training the neural networks,in addition to the complexity involved in the analyzed data.An auto-machine learning tool called AutoKeras was used to define the hyperparameters of the implemented models.The results found were satisfactory and presented an accuracy greater than 70%for image samples belonging to other wells not seen during the model building,which increases the applicability of the implemented model.Finally,a comparison was made between the proposed methodology and multiple-class models,demonstrating the superiority of one-class models.
基金supported in part by the Natural Science Foundation of China(62303361)in part by the Hainan Provincial Natural Science Foundation of China(623QN266)+2 种基金the Fundamental Research Funds for the Central Universities(WUT:233110002)in part by the University-Industry Collaborative Education Program(231002531131826)in part by the National Key R&D Program of China(2018AAA0101502)
文摘COMPUTATIONAL knowledge vision[1]is emphasized as a novel perspective or field in this paper.It first proposes the visual hierarchy and its connection to knowledge,stating that knowledge is a justified true belief.To further the previous research,we concisely summarize our recent works and suggest a new direction that knowledge is also a thought framework in vision.
基金by the National Natural Science Foundation of China(31671128,31925020,31700999,31700943,and 31500882)the Changjiang Scholar Professorship Award(T2016031)Fundamental Research Funds for the Central Universities(2017EYT35).
文摘Visual object recognition in humans and nonhuman primates is achieved by the ventral visual pathway(ventral occipital-temporal cortex,VOTC),which shows a well-documented object domain structure.An on-going question is what type of information is processed in the higher-order VOTC that underlies such observations,with recent evidence suggesting effects of certain visual features.Combining computational vision models,fMRI experiment using a parametric-modulation approach,and natural image statistics of common objects,we depicted the neural distribution of a comprehensive set of visual features in the VOTC,identifying voxel sensitivities with specific feature sets across geometry/shape,Fourier power,and color.The visual feature combination pattern in the VOTC is significantly explained by their relationships to different types of response-action computation(fight-or-flight,navigation,and manipulation),as derived from behavioral ratings and natural image statistics.These results offer a comprehensive visual feature map in the VOTC and a plausible theoretical explanation as a mapping onto different types of downstream response-action systems.
基金support from the National Natural Science Foundation of China(grant Nos.62334010,62121005,62022081,and 62304221)the National Key Research and Development Program(grant number 2021YFA0717600)+1 种基金the Natural Science Foundation of Jilin Province(20240101377JC)the International Fund Program of Changchun Institute of Optics,Fine Mechanics and Physics,Chinese Academy of Sciences.
文摘Neuromorphic computing vision is the most promising technological solution to overcome the arithmetic bottleneck in machine vision applications.All-in-one neuromorphic sensors have been attracting increased attention because they can integrate visual perception,processing,and memory functionalities into one single device.However,the limited responsivity and data retention time of all-in-one neuromorphic sensors usually hinder their potential in multispectral machine vision,especially in the near-infrared(NIR)band which contains critical information for pattern recognition.Here,we demonstrate physisorption-assistant optoelectronic synaptic transistors based on Ta_(2)NiSe_(5)/SnS_(2) heterojunction,which present tunable synaptic functionality in broadband(375–1310 nm).We propose a strategy about the physisorption-assistant persistent photoconductivity(PAPPC)effect to effectively solve the problem in detecting and storing the NIR light information.Under this strategy,the responsivity and data retention time of our devices were significantly enhanced and prolonged in broadband from 375 to 1310 nm.Further,the devices realize multilevel non-volatile optoelectronic memory through the modulation of several optical and back-gate signals to simulate emotion-controlled learning and memory processes,optical writing-electric erasing,and associative learning.Moreover,we developed a simplified human visual system to simulate color-cognitive perception and memory functions.Our approach offers a route for creating advanced all-in-one neuromorphic sensors and developing neuromorphic computing vision.
