scholarly journals Constructing Geospatial Concept Graphs from Tagged Images for Geo-Aware Fine-Grained Image Recognition

2020 ◽  
Vol 9 (6) ◽  
pp. 354
Author(s):  
Naoko Nitta ◽  
Kazuaki Nakamura ◽  
Noboru Babaguchi
Keyword(s):  
Image Recognition ◽  
Prior Information ◽  
Training Data ◽  
Coarse Grained ◽  
Main Role ◽  
Fine Grained ◽  
Image Sharing ◽  
Additional Information ◽  
Visual Concepts ◽  
Geographical Locations

While visual appearances play a main role in recognizing the concepts captured in images, additional information can provide complementary information for fine-grained image recognition, where concepts with similar visual appearances such as species of birds need to be distinguished. Especially for recognizing geospatial concepts, which are observed only at specific places, geographical locations of the images can improve the recognition accuracy. However, such geo-aware fine-grained image recognition requires prior information about the visual and geospatial features of each concept or the training data composed of high-quality images for each concept associated with correct geographical locations. By using a large number of images photographed in various places and described with textual tags which can be collected from image sharing services such as Flickr, this paper proposes a method for constructing a geospatial concept graph which contains the necessary prior information for realizing the geo-aware fine-grained image recognition, such as a set of visually recognizable fine-grained geospatial concepts, their visual and geospatial features, and the coarse-grained representative visual concepts whose visual features can be transferred to several fine-grained geospatial concepts. Leveraging the information from the images captured by many people can automatically extract diverse types of geospatial concepts with proper features for realizing efficient and effective geo-aware fine-grained image recognition.

scholarly journals Generalized Zero-Shot Vehicle Detection in Remote Sensing Imagery via Coarse-to-Fine Framework

2019 ◽  
Author(s):  
Hong Chen ◽  
Yongtan Luo ◽  
Liujuan Cao ◽  
Baochang Zhang ◽  
Guodong Guo ◽  
...  
Keyword(s):  
Remote Sensing ◽  
Vehicle Detection ◽  
Training Data ◽  
Coarse Grained ◽  
Vehicle Classification ◽  
Vehicle Localization ◽  
Fine Grained ◽  
Coarse To Fine ◽  
Generalized Zero ◽  
Detection And Recognition

Vehicle detection and recognition in remote sensing images are challenging, especially when only limited training data are available to accommodate various target categories. In this paper, we introduce a novel coarse-to-fine framework, which decomposes vehicle detection into segmentation-based vehicle localization and generalized zero-shot vehicle classification. Particularly, the proposed framework can well handle the problem of generalized zero-shot vehicle detection, which is challenging due to the requirement of recognizing vehicles that are even unseen during training. Specifically, a hierarchical DeepLab v3 model is proposed in the framework, which fully exploits fine-grained features to locate the target on a pixel-wise level, then recognizes vehicles in a coarse-grained manner. Additionally, the hierarchical DeepLab v3 model is beneficially compatible to combine the generalized zero-shot recognition. To the best of our knowledge, there is no publically available dataset to test comparative methods, we therefore construct a new dataset to fill this gap of evaluation. The experimental results show that the proposed framework yields promising results on the imperative yet difficult task of zero-shot vehicle detection and recognition.

scholarly journals Relevance-guided Supervision for OpenQA with ColBERT

2021 ◽  
Vol 9 ◽  
pp. 929-944
Author(s):  
Omar Khattab ◽  
Christopher Potts ◽  
Matei Zaharia
Keyword(s):  
Question Answering ◽  
State Of The Art ◽  
Training Data ◽  
Coarse Grained ◽  
Retrieval Model ◽  
Open Domain ◽  
Weak Supervision ◽  
Fine Grained ◽  
Vector Representations ◽  
Large Corpus

Abstract Systems for Open-Domain Question Answering (OpenQA) generally depend on a retriever for finding candidate passages in a large corpus and a reader for extracting answers from those passages. In much recent work, the retriever is a learned component that uses coarse-grained vector representations of questions and passages. We argue that this modeling choice is insufficiently expressive for dealing with the complexity of natural language questions. To address this, we define ColBERT-QA, which adapts the scalable neural retrieval model ColBERT to OpenQA. ColBERT creates fine-grained interactions between questions and passages. We propose an efficient weak supervision strategy that iteratively uses ColBERT to create its own training data. This greatly improves OpenQA retrieval on Natural Questions, SQuAD, and TriviaQA, and the resulting system attains state-of-the-art extractive OpenQA performance on all three datasets.

scholarly journals Relative Information Gain: Shannon entropy-based measure of the relative structural conservation in RNA alignments

2021 ◽  
Vol 3 (1) ◽  
Author(s):  
Marco Pietrosanto ◽  
Marta Adinolfi ◽  
Andrea Guarracino ◽  
Fabrizio Ferrè ◽  
Gabriele Ausiello ◽  
...  
Keyword(s):  
Secondary Structure ◽  
Information Gain ◽  
Structural Information ◽  
Coarse Grained ◽  
Fine Grained ◽  
Additional Information ◽  
Levels Of Detail ◽  
Relative Information ◽  
Substitution Matrices ◽  
Secondary Structure Models

Abstract Structural characterization of RNAs is a dynamic field, offering many modelling possibilities. RNA secondary structure models are usually characterized by an encoding that depicts structural information of the molecule through string representations or graphs. In this work, we provide a generalization of the BEAR encoding (a context-aware structural encoding we previously developed) by expanding the set of alignments used for the construction of substitution matrices and then applying it to secondary structure encodings ranging from fine-grained to more coarse-grained representations. We also introduce a re-interpretation of the Shannon Information applied on RNA alignments, proposing a new scoring metric, the Relative Information Gain (RIG). The RIG score is available for any position in an alignment, showing how different levels of detail encoded in the RNA representation can contribute differently to convey structural information. The approaches presented in this study can be used alongside state-of-the-art tools to synergistically gain insights into the structural elements that RNAs and RNA families are composed of. This additional information could potentially contribute to their improvement or increase the degree of confidence in the secondary structure of families and any set of aligned RNAs.

Correlation between dislocation, grain boundary and interface of duplex SS in stress corrosion cracking(SCC)

1990 ◽  
Vol 48 (4) ◽  
pp. 928-929
Author(s):  
Wang Zheng-fang ◽  
Z.F. Wang
Keyword(s):  
Stainless Steel ◽  
Thermal Cycle ◽  
Secondary Phase ◽  
Corrosion Cracking ◽  
Coarse Grained ◽  
Test Environment ◽  
Fine Grained ◽  
Evaluation Test ◽  
Welding Thermal Cycle ◽  
Micro Analysis

The main purpose of this study highlights on the evaluation of chloride SCC resistance of the material,duplex stainless steel,OOCr18Ni5Mo3Si2 (18-5Mo) and its welded coarse grained zone(CGZ).18-5Mo is a dual phases (A+F) stainless steel with yield strength:512N/mm2 .The proportion of secondary Phase(A phase) accounts for 30-35% of the total with fine grained and homogeneously distributed A and F phases(Fig.1).After being welded by a specific welding thermal cycle to the material,i.e. Tmax=1350°C and t8/5=20s,microstructure may change from fine grained morphology to coarse grained morphology and from homogeneously distributed of A phase to a concentration of A phase(Fig.2).Meanwhile,the proportion of A phase reduced from 35% to 5-10°o.For this reason it is known as welded coarse grained zone(CGZ).In association with difference of microstructure between base metal and welded CGZ,so chloride SCC resistance also differ from each other.Test procedures:Constant load tensile test(CLTT) were performed for recording Esce-t curve by which corrosion cracking growth can be described, tf,fractured time,can also be recorded by the test which is taken as a electrochemical behavior and mechanical property for SCC resistance evaluation. Test environment:143°C boiling 42%MgCl2 solution is used.Besides, micro analysis were conducted with light microscopy(LM),SEM,TEM,and Auger energy spectrum(AES) so as to reveal the correlation between the data generated by the CLTT results and micro analysis.

scholarly journals Domain randomization-enhanced deep learning models for bird detection

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Xin Mao ◽  
Jun Kang Chow ◽  
Pin Siang Tan ◽  
Kuan-fu Liu ◽  
Jimmy Wu ◽  
...  
Keyword(s):  
Deep Learning ◽  
Continuous Monitoring ◽  
Bird Species ◽  
Training Data ◽  
Learning Models ◽  
Fine Grained ◽  
Bird Detection ◽  
Relationship Of ◽  
The Relationship

AbstractAutomatic bird detection in ornithological analyses is limited by the accuracy of existing models, due to the lack of training data and the difficulties in extracting the fine-grained features required to distinguish bird species. Here we apply the domain randomization strategy to enhance the accuracy of the deep learning models in bird detection. Trained with virtual birds of sufficient variations in different environments, the model tends to focus on the fine-grained features of birds and achieves higher accuracies. Based on the 100 terabytes of 2-month continuous monitoring data of egrets, our results cover the findings using conventional manual observations, e.g., vertical stratification of egrets according to body size, and also open up opportunities of long-term bird surveys requiring intensive monitoring that is impractical using conventional methods, e.g., the weather influences on egrets, and the relationship of the migration schedules between the great egrets and little egrets.

scholarly journals Balanced Sparsity for Efficient DNN Inference on GPU

2019 ◽  
Vol 33 ◽  
pp. 5676-5683 ◽  
Author(s):  
Zhuliang Yao ◽  
Shijie Cao ◽  
Wencong Xiao ◽  
Chen Zhang ◽  
Lanshun Nie
Keyword(s):  
Deep Neural Networks ◽  
General Purpose ◽  
Coarse Grained ◽  
Efficient Computation ◽  
Model Accuracy ◽  
Sparse Model ◽  
Model Inference ◽  
Fine Grained ◽  
Practical Inference ◽  
Speed Up

In trained deep neural networks, unstructured pruning can reduce redundant weights to lower storage cost. However, it requires the customization of hardwares to speed up practical inference. Another trend accelerates sparse model inference on general-purpose hardwares by adopting coarse-grained sparsity to prune or regularize consecutive weights for efficient computation. But this method often sacrifices model accuracy. In this paper, we propose a novel fine-grained sparsity approach, Balanced Sparsity, to achieve high model accuracy with commercial hardwares efficiently. Our approach adapts to high parallelism property of GPU, showing incredible potential for sparsity in the widely deployment of deep learning services. Experiment results show that Balanced Sparsity achieves up to 3.1x practical speedup for model inference on GPU, while retains the same high model accuracy as finegrained sparsity.

scholarly journals Submarine lava deltas of the 2018 eruption of Kīlauea volcano

2021 ◽  
Vol 83 (4) ◽  
Author(s):  
S. Adam Soule ◽  
Michael Zoeller ◽  
Carolyn Parcheta
Keyword(s):  
Grain Size ◽  
Pore Pressure ◽  
Mechanistic Model ◽  
Large Fraction ◽  
Volcanic Hazards ◽  
Lava Flows ◽  
Coarse Grained ◽  
Fine Grained ◽  
Fine Grain ◽  
Delta Formation

AbstractHawaiian and other ocean island lava flows that reach the coastline can deposit significant volumes of lava in submarine deltas. The catastrophic collapse of these deltas represents one of the most significant, but least predictable, volcanic hazards at ocean islands. The volume of lava deposited below sea level in delta-forming eruptions and the mechanisms of delta construction and destruction are rarely documented. Here, we report on bathymetric surveys and ROV observations following the Kīlauea 2018 eruption that, along with a comparison to the deltas formed at Pu‘u ‘Ō‘ō over the past decade, provide new insight into delta formation. Bathymetric differencing reveals that the 2018 deltas contain more than half of the total volume of lava erupted. In addition, we find that the 2018 deltas are comprised largely of coarse-grained volcanic breccias and intact lava flows, which contrast with those at Pu‘u ‘Ō‘ō that contain a large fraction of fine-grained hyaloclastite. We attribute this difference to less efficient fragmentation of the 2018 ‘a‘ā flows leading to fragmentation by collapse rather than hydrovolcanic explosion. We suggest a mechanistic model where the characteristic grain size influences the form and stability of the delta with fine grain size deltas (Pu‘u ‘Ō‘ō) experiencing larger landslides with greater run-out supported by increased pore pressure and with coarse grain size deltas (Kīlauea 2018) experiencing smaller landslides that quickly stop as the pore pressure rapidly dissipates. This difference, if validated for other lava deltas, would provide a means to assess potential delta stability in future eruptions.

scholarly journals A cooperative DDoS attack detection scheme based on entropy and ensemble learning in SDN

2021 ◽  
Vol 2021 (1) ◽  
Author(s):  
Shanshan Yu ◽  
Jicheng Zhang ◽  
Ju Liu ◽  
Xiaoqing Zhang ◽  
Yafeng Li ◽  
...  
Keyword(s):  
Ensemble Learning ◽  
Denial Of Service ◽  
Attack Detection ◽  
Coarse Grained ◽  
Communication Overhead ◽  
Detection Scheme ◽  
Fine Grained ◽  
Ddos Attack ◽  
Network Status ◽  
Ddos Attack Detection

AbstractIn order to solve the problem of distributed denial of service (DDoS) attack detection in software-defined network, we proposed a cooperative DDoS attack detection scheme based on entropy and ensemble learning. This method sets up a coarse-grained preliminary detection module based on entropy in the edge switch to monitor the network status in real time and report to the controller if any abnormality is found. Simultaneously, a fine-grained precise attack detection module is designed in the controller, and a ensemble learning-based algorithm is utilized to further identify abnormal traffic accurately. In this framework, the idle computing capability of edge switches is fully utilized with the design idea of edge computing to offload part of the detection task from the control plane to the data plane innovatively. Simulation results of two common DDoS attack methods, ICMP and SYN, show that the system can effectively detect DDoS attacks and greatly reduce the southbound communication overhead and the burden of the controller as well as the detection delay of the attacks.

scholarly journals BeautyNet: Joint Multiscale CNN and Transfer Learning Method for Unconstrained Facial Beauty Prediction

2019 ◽  
Vol 2019 ◽  
pp. 1-14 ◽  
Author(s):  
Yikui Zhai ◽  
He Cao ◽  
Wenbo Deng ◽  
Junying Gan ◽  
Vincenzo Piuri ◽  
...  
Keyword(s):  
Transfer Learning ◽  
Classification Accuracy ◽  
Learning Strategy ◽  
State Of The Art ◽  
Activation Function ◽  
Training Data ◽  
Fine Grained ◽  
Pattern Recognition Problem ◽  
Face Features ◽  
Facial Beauty

Because of the lack of discriminative face representations and scarcity of labeled training data, facial beauty prediction (FBP), which aims at assessing facial attractiveness automatically, has become a challenging pattern recognition problem. Inspired by recent promising work on fine-grained image classification using the multiscale architecture to extend the diversity of deep features, BeautyNet for unconstrained facial beauty prediction is proposed in this paper. Firstly, a multiscale network is adopted to improve the discriminative of face features. Secondly, to alleviate the computational burden of the multiscale architecture, MFM (max-feature-map) is utilized as an activation function which can not only lighten the network and speed network convergence but also benefit the performance. Finally, transfer learning strategy is introduced here to mitigate the overfitting phenomenon which is caused by the scarcity of labeled facial beauty samples and improves the proposed BeautyNet’s performance. Extensive experiments performed on LSFBD demonstrate that the proposed scheme outperforms the state-of-the-art methods, which can achieve 67.48% classification accuracy.

Sign in / Sign up

Export Citation Format

Share Document