One of the most basic and difficult areas of computer vision and image understanding applications is still object detection. Deep neural network models and enhanced object representation have led to significant progre...One of the most basic and difficult areas of computer vision and image understanding applications is still object detection. Deep neural network models and enhanced object representation have led to significant progress in object detection. This research investigates in greater detail how object detection has changed in the recent years in the deep learning age. We provide an overview of the literature on a range of cutting-edge object identification algorithms and the theoretical underpinnings of these techniques. Deep learning technologies are contributing to substantial innovations in the field of object detection. While Convolutional Neural Networks (CNN) have laid a solid foundation, new models such as You Only Look Once (YOLO) and Vision Transformers (ViTs) have expanded the possibilities even further by providing high accuracy and fast detection in a variety of settings. Even with these developments, integrating CNN, YOLO and ViTs, into a coherent framework still poses challenges with juggling computing demand, speed, and accuracy especially in dynamic contexts. Real-time processing in applications like surveillance and autonomous driving necessitates improvements that take use of each model type’s advantages. The goal of this work is to provide an object detection system that maximizes detection speed and accuracy while decreasing processing requirements by integrating YOLO, CNN, and ViTs. Improving real-time detection performance in changing weather and light exposure circumstances, as well as detecting small or partially obscured objects in crowded cities, are among the goals. We provide a hybrid architecture which leverages CNN for robust feature extraction, YOLO for rapid detection, and ViTs for remarkable global context capture via self-attention techniques. Using an innovative training regimen that prioritizes flexible learning rates and data augmentation procedures, the model is trained on an extensive dataset of urban settings. Compared to solo YOLO, CNN, or ViTs models, the suggested model exhibits an increase in detection accuracy. This improvement is especially noticeable in difficult situations such settings with high occlusion and low light. In addition, it attains a decrease in inference time in comparison to baseline models, allowing real-time object detection without performance loss. This work introduces a novel method of object identification that integrates CNN, YOLO and ViTs, in a synergistic way. The resultant framework extends the use of integrated deep learning models in practical applications while also setting a new standard for detection performance under a variety of conditions. Our research advances computer vision by providing a scalable and effective approach to object identification problems. Its possible uses include autonomous navigation, security, and other areas.展开更多
In 2015,the U.S National Institute of Standards and Technology(NIST)funded the Center of Excellence for Risk-Based Community Resilience Planning(CoE),a fourteen university-based consortium of almost 100 col-laborators...In 2015,the U.S National Institute of Standards and Technology(NIST)funded the Center of Excellence for Risk-Based Community Resilience Planning(CoE),a fourteen university-based consortium of almost 100 col-laborators,including faculty,students,post-doctoral scholars,and NIST researchers.This paper highlights the scientific theory behind the state-of-the-art cloud platform being developed by the CoE-the Interdisciplinary Networked Community Resilience Modeling Environment(IN-CORE).IN-CORE enables communities,consul-tants,and researchers to set up complex interdependent models of an entire community consisting of people,businesses,social institutions,buildings,transportation networks,water networks,and electric power networks and to predict their performance and recovery to hazard scenario events,including uncertainty propagation through the chained models.The modeling environment includes a detailed building inventory,hazard scenario models,building and infrastructure damage(fragility)and recovery functions,social science data-driven house-hold and business models,and computable general equilibrium(CGE)models of local economies.An important aspect of IN-CORE is the characterization of uncertainty and its propagation throughout the chained models of the platform.Three illustrative examples of community testbeds are presented that look at hazard impacts and recovery on population,economics,physical services,and social services.An overview of the IN-CORE technology and scientific implementation is described with a focus on four key community stability areas(CSA)that encompass an array of community resilience metrics(CRM)and support community resilience informed decision-making.Each testbed within IN-CORE has been developed by a team of engineers,social scientists,urban planners,and economists.Community models,begin with a community description,i.e.,people,businesses,buildings,infras-tructure,and progresses to the damage and loss of functions caused by a hazard scenario,i.e.,a flood,tornado,hurricane,or earthquake.This process is accomplished through chaining of modular algorithms,as described.The baseline community characteristics and the hazard-induced damage sets are the initial conditions for the recovery models,which have been the least studied area of community resilience but arguably one of the most important.Communities can then test the effect of mitigation and/or policies and compare the effects of“what if”scenarios on physical,social,and economic metrics with the only requirement being that the change much be able to be numerically modeled in IN-CORE.展开更多
In this paper,we report upon our recent work aimed at improving and adapting machine learning algorithms to automatically classify nanoscience images acquired by the Scanning Electron Microscope(SEM).This is done by c...In this paper,we report upon our recent work aimed at improving and adapting machine learning algorithms to automatically classify nanoscience images acquired by the Scanning Electron Microscope(SEM).This is done by coupling supervised and unsupervised learning approaches.We first investigate supervised learning on a ten-category data set of images and compare the performance of the different models in terms of training accuracy.Then,we reduce the dimensionality of the features through autoencoders to perform unsupervised learning on a subset of images in a selected range of scales(from 1μm to 2μm).Finally,we compare different clustering methods to uncover intrinsic structures in the images.展开更多
文摘One of the most basic and difficult areas of computer vision and image understanding applications is still object detection. Deep neural network models and enhanced object representation have led to significant progress in object detection. This research investigates in greater detail how object detection has changed in the recent years in the deep learning age. We provide an overview of the literature on a range of cutting-edge object identification algorithms and the theoretical underpinnings of these techniques. Deep learning technologies are contributing to substantial innovations in the field of object detection. While Convolutional Neural Networks (CNN) have laid a solid foundation, new models such as You Only Look Once (YOLO) and Vision Transformers (ViTs) have expanded the possibilities even further by providing high accuracy and fast detection in a variety of settings. Even with these developments, integrating CNN, YOLO and ViTs, into a coherent framework still poses challenges with juggling computing demand, speed, and accuracy especially in dynamic contexts. Real-time processing in applications like surveillance and autonomous driving necessitates improvements that take use of each model type’s advantages. The goal of this work is to provide an object detection system that maximizes detection speed and accuracy while decreasing processing requirements by integrating YOLO, CNN, and ViTs. Improving real-time detection performance in changing weather and light exposure circumstances, as well as detecting small or partially obscured objects in crowded cities, are among the goals. We provide a hybrid architecture which leverages CNN for robust feature extraction, YOLO for rapid detection, and ViTs for remarkable global context capture via self-attention techniques. Using an innovative training regimen that prioritizes flexible learning rates and data augmentation procedures, the model is trained on an extensive dataset of urban settings. Compared to solo YOLO, CNN, or ViTs models, the suggested model exhibits an increase in detection accuracy. This improvement is especially noticeable in difficult situations such settings with high occlusion and low light. In addition, it attains a decrease in inference time in comparison to baseline models, allowing real-time object detection without performance loss. This work introduces a novel method of object identification that integrates CNN, YOLO and ViTs, in a synergistic way. The resultant framework extends the use of integrated deep learning models in practical applications while also setting a new standard for detection performance under a variety of conditions. Our research advances computer vision by providing a scalable and effective approach to object identification problems. Its possible uses include autonomous navigation, security, and other areas.
基金The Center for Risk-Based Community Resilience Planning is a NIST-funded Center of Excellencethe Center is funded through a cooperative agreement between the U.S.National Institute of Standards and Tech-nology and Colorado State University(NIST Financial Assistance Award Numbers:70NANB15H044 and 70NANB20H008)。
文摘In 2015,the U.S National Institute of Standards and Technology(NIST)funded the Center of Excellence for Risk-Based Community Resilience Planning(CoE),a fourteen university-based consortium of almost 100 col-laborators,including faculty,students,post-doctoral scholars,and NIST researchers.This paper highlights the scientific theory behind the state-of-the-art cloud platform being developed by the CoE-the Interdisciplinary Networked Community Resilience Modeling Environment(IN-CORE).IN-CORE enables communities,consul-tants,and researchers to set up complex interdependent models of an entire community consisting of people,businesses,social institutions,buildings,transportation networks,water networks,and electric power networks and to predict their performance and recovery to hazard scenario events,including uncertainty propagation through the chained models.The modeling environment includes a detailed building inventory,hazard scenario models,building and infrastructure damage(fragility)and recovery functions,social science data-driven house-hold and business models,and computable general equilibrium(CGE)models of local economies.An important aspect of IN-CORE is the characterization of uncertainty and its propagation throughout the chained models of the platform.Three illustrative examples of community testbeds are presented that look at hazard impacts and recovery on population,economics,physical services,and social services.An overview of the IN-CORE technology and scientific implementation is described with a focus on four key community stability areas(CSA)that encompass an array of community resilience metrics(CRM)and support community resilience informed decision-making.Each testbed within IN-CORE has been developed by a team of engineers,social scientists,urban planners,and economists.Community models,begin with a community description,i.e.,people,businesses,buildings,infras-tructure,and progresses to the damage and loss of functions caused by a hazard scenario,i.e.,a flood,tornado,hurricane,or earthquake.This process is accomplished through chaining of modular algorithms,as described.The baseline community characteristics and the hazard-induced damage sets are the initial conditions for the recovery models,which have been the least studied area of community resilience but arguably one of the most important.Communities can then test the effect of mitigation and/or policies and compare the effects of“what if”scenarios on physical,social,and economic metrics with the only requirement being that the change much be able to be numerically modeled in IN-CORE.
基金This work has been done within the NFFA-EUROPE project and has received funding from the European Union’s Horizon 2020 Research and Innovation Program under grant agreement No.654360 NFFA-EUROPE.
文摘In this paper,we report upon our recent work aimed at improving and adapting machine learning algorithms to automatically classify nanoscience images acquired by the Scanning Electron Microscope(SEM).This is done by coupling supervised and unsupervised learning approaches.We first investigate supervised learning on a ten-category data set of images and compare the performance of the different models in terms of training accuracy.Then,we reduce the dimensionality of the features through autoencoders to perform unsupervised learning on a subset of images in a selected range of scales(from 1μm to 2μm).Finally,we compare different clustering methods to uncover intrinsic structures in the images.
