- 标题
- 摘要
- 关键词
- 实验方案
- 产品
过滤筛选
- 2018
- Conditional Random Fields (CRF)
- Convolutional Neural Network (CNN)
- Fine Classification
- Airborne hyperspectral
- green tide
- Elegant End-to-End Fully Convolutional Network (E3FCN)
- deep learning
- remote sensing
- Moderate Resolution Imaging Spectroradiometer (MODIS)
- Optoelectronic Information Science and Engineering
- Ocean University of China
- Wuhan University
- Central South University
- Hubei University
-
[IEEE 2018 15th European Radar Conference (EuRAD) - Madrid, Spain (2018.9.26-2018.9.28)] 2018 15th European Radar Conference (EuRAD) - Deep Learning based Human Activity Classification in Radar Micro-Doppler Image
摘要： A convolutional neural network (CNN) based deep learning (DL) approach to classify human activities in micro-Doppler spectrogram of radar is investigated. MOCAP dataset, from Carnegie Mellon University, is used for spectrogram simulation. Seven activities are classified with the proposed CNN network. Our network outperforms several previously published DL-based approaches. To understand the network’s impact on classification performance, we investigate some key parameters of the proposed network. Experiment result demonstrates that a deeper network does not necessarily result in a higher accuracy. We also examine the network size and the number of output feature maps to find out their impact on the result.
关键词： Deep Learning,Convolutional Neural Network,Human Activity Classification,Micro-Doppler Spectrogram,Radar image
更新于2025-09-23 15:21:21
-
[IEEE IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium - Valencia, Spain (2018.7.22-2018.7.27)] IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium - Multitask Classification of Remote Sensing Scenes Using Deep Neural Networks
摘要： The problem of scene classification in remote sensing (RS) images has attracted a lot of attention recently. Many datasets have been presented in the literature for this purpose with each claiming to be the benchmark dataset. In this paper, we propose a different approach to the RS community. Instead of putting our effort in building larger and large scene datasets, we argue that it is better to build a machine learning framework that can learn from all available datasets. We formulate this as a multitask learning problem where each dataset represents a task. Then, we present a deep learning solution that can perform multitask learning. We test the proposed multitask network on three popular scene datasets, namely UC Merced, KSA, and AID datasets. Preliminary results show the promising capabilities of this solution at sharing information between tasks and improving the classification accuracy.
关键词： Deep learning,Scene classification,Multitask classification,Convolutional Neural Network
更新于2025-09-23 15:21:21
-
[IEEE IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium - Valencia, Spain (2018.7.22-2018.7.27)] IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium - Time-Scale Transferring Deep Convolutional Neural Network for Mapping Early Rice
摘要： In recent years, the use of deep learning in remote sensing domain has made it possible to automate mapping in large-scale. In this paper, we propose a transfer learning method which pre-train a convolutional neural network (CNN) with middle-resolution remote sensing data in 2016, and fine-tune it in following years with a spot of high-resolution remote sensing data in 2017. We used the fine-tuned model to mapping the early-rice in 25 countries which cost only 21 minutes, and yielded an overall accuracy of 81.68%. The result demonstrate that the convolutional neural network model can transfer in different time period with little adjustment in a very high accuracy.
关键词： middle-resolution data,convolutional neural network,time-scale,transfer learning
更新于2025-09-23 15:21:21
-
[IEEE IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium - Valencia, Spain (2018.7.22-2018.7.27)] IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium - Rotated Region Based Fully Convolutional Network for Ship Detection
摘要： Ship detection from high-resolution optical remote sensing images has been a prevalent domain in recent years. Unlike objects in natural images, ships of interest can be anywhere in optical remote sensing images with multi-scale and multi-oriented which makes it more difficult to be detected. In this paper, we propose a novel method based on the fully convolutional network to detect ships. Our method has three important components: 1) we design a network merging different levels of feature map to fuse multi-scale information. Determining the existence of large ship require features from deep layers in the network, while predicting rotated bounding box enclosing small ships needs shallow layers information; 2) The network can be trained end-to-end to generate score maps which indicates the confidence score for the ship region of interest in pixel-wise level through all locations and scale of an image; 3) We design a rotated bounding box regression model to localize the ships. The experimental results on our dataset collected from Google Earth has demonstrated our proposed method achieves promising performance on ship detection in terms of both efficiency and accuracy in high-resolution optical remote sensing images.
关键词： Ship detection,Rotated region,Fully Convolutional network
更新于2025-09-23 15:21:21
-
[IEEE IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium - Valencia, Spain (2018.7.22-2018.7.27)] IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium - Introducing Eurosat: A Novel Dataset and Deep Learning Benchmark for Land Use and Land Cover Classification
摘要： In this paper, we address the challenge of land use and land cover classification using Sentinel-2 satellite images. The key contributions are as follows. We present a novel dataset based on Sentinel-2 satellite images covering 13 different spectral bands and consisting of 10 classes with in total 27,000 labeled images. We evaluate state-of-the-art deep Convolutional Neural Networks (CNNs) on this novel dataset with its different spectral bands. We also evaluate deep CNNs on existing remote sensing datasets and compare the obtained results. With the proposed novel dataset, we achieved an overall classification accuracy of 98.57%. The classification system resulting from the proposed research opens a gate towards various Earth observation applications. We demonstrate how the classification system can assist in improving geographical maps.
关键词： Deep Learning,Land Use Classification,Earth Observation,Convolutional Neural Network,Machine Learning,Dataset,Land Cover Classification
更新于2025-09-23 15:21:21
-
[IEEE 2018 IEEE Innovative Smart Grid Technologies - Asia (ISGT Asia) - Singapore (2018.5.22-2018.5.25)] 2018 IEEE Innovative Smart Grid Technologies - Asia (ISGT Asia) - Solar Plant Integration to Utility Grid with Improved Power Quality by using RNN-Hebbian-LMS Current Controller
摘要： In this paper a new topology of current control technique is proposed, to transfer the improved quality of power produced from the solar plant to the utility grid. Also, a high-gain high-efficient converter driving with Kalman MPPT is used to boost the low voltage levels of the PV array. The proposed control uses recurrent neural network RNN-Hebbian -LMS based current controller to achieve the better performance in terms of power quality. The RNN network uses feedback signals to control the current flow from solar plant to the utility grid. The Hebbian - LMS (least mean square) algorithm is used to update the weights of the RNN based current controller. The main advantage of the RNN-Hebbian-LMS current control technique is to maintain the constant voltage. Besides, it also provides system stability over wide range of parameter variations and damp out the oscillations quickly. The proposed algorithm is able to overcome the stability and sensitivity problems incurred with the conventional PI current controller. The simulation results will be compared with the conventional PI and proposed RNN-Hebbiab-LMS current controllers. Finally, the proposed current controller shows the improved power quality, quick settling time and more stable comparing with the conventional controllers.
关键词： Least Mean Square (LMS),Utility Grid,Solar Plant,High-Gain converter,PV array,Maximum Power Point Tracking (MPPT),Recurrent Neural Network (RNN)
更新于2025-09-23 15:21:21
-
[Lecture Notes in Computer Science] Pattern Recognition and Computer Vision Volume 11256 (First Chinese Conference, PRCV 2018, Guangzhou, China, November 23-26, 2018, Proceedings, Part I) || Hand Dorsal Vein Recognition Based on Deep Hash Network
摘要： As a unique biometric technology that has emerged in recent decades, hand dorsal vein recognition has received increasing attention due to its higher safety and convenience. In order to further improve the recognition accuracy, in this paper we propose an end-to-end method for recognizing Hand dorsal vein Based on Deep hash network (DHN), called HBD. The hand dorsal vein image is input into the simpli?ed Convolutional Neural Networks-Fast (SCNN-F) to obtain convolution features. At the last fully connected layer, for the outputs of 128 neurons, sgn function is used to encode each image as 128-bit code. By comparing the distances between images after coding, it can be judged whether they are from the same person. Using a special loss function and training strategy, we verify the effectiveness of HBD on the NCUT, GPDS, and NCUT+GPDS database, respectively. The experimental results show that the HBD method can achieve comparable accuracy to the state-of-the-arts. In NCUT database, when the ratio of training and test set is 7:3, the Equal Error Rate (EER) of the test set is 0.08%, which is an order of magnitude lower than other algorithms. More importantly, due to the adoption of a simpler network structure and hash coding, HBD operates more ef?ciently and has superior performance gains over other deep learning methods while ensuring the accuracy.
关键词： Hand dorsal vein recognition,Deep hash network,Biometrics
更新于2025-09-23 15:21:21
-
[Advances in Intelligent Systems and Computing] Image Processing and Communications Challenges 10 Volume 892 (10th International Conference, IP&C’2018 Bydgoszcz, Poland, November 2018, Proceedings) || Air-Gap Data Transmission Using Backlight Modulation of Screen
摘要： Novel technique for data transmission from air–gap secured computer is considered in this paper. Backlight modulation of screen using BFSK allows data transmission that is not visible for human. The application of digital camera equipped and telescope allows data recovery during the lack of the user’s activity. Demodulation scheme with automatic selection of demodulation ?lters is presented. Di?erent con?guration of data transmission parameters and acquisition hardware were tested.
关键词： BFSK,Image processing,Air–gap transmission,Network security,Digital demodulation
更新于2025-09-23 15:21:21
-
Deep convolutional representations and kernel extreme learning machines for image classification
摘要： Convolutional Neural Networks (CNNs) have been established as a powerful class of models for image classification and related tasks. However, the fully-connected layers in CNN are not robust enough to serve as a classifier to discriminate deep convolutional features, due to the local minima problem of back-propagation. Kernel Extreme Learning Machines (KELMs), known as an outstanding classifier, can not only converge extremely fast but also ensure an outstanding generalization performance. In this paper, we propose a novel image classification framework, in which CNN and KELM are well integrated. In our work, Densely connected network (DenseNet) is employed as the feature extractor, while a radial basis function kernel ELM instead of linear fully connected layer is adopted as a classifier to discriminate categories of extracted features to promote the image classification performance. Experiments conducted on four publicly available datasets demonstrate the promising performance of the proposed framework against the state-of-the-art methods.
关键词： Extreme learning machine,Neural network,Image classification
更新于2025-09-23 15:21:21
-
Road Segmentation Based on Hybrid Convolutional Network for High-Resolution Visible Remote Sensing Image
摘要： Road segmentation plays an important role in many applications, such as intelligent transportation system and urban planning. Various road segmentation methods have been proposed for visible remote sensing images, especially the popular convolutional neural network-based methods. However, high-accuracy road segmentation from high-resolution visible remote sensing images is still a challenging problem due to complex background and multiscale roads in these images. To handle this problem, a hybrid convolutional network (HCN), fusing multiple subnetworks, is proposed in this letter. The HCN contains a fully convolutional network, a modi?ed U-Net, and a VGG subnetwork; these subnetworks obtain a coarse-grained, a medium-grained, and a ?ne-grained road segmentation map. Moreover, the HCN uses a shallow convolutional subnetwork to fuse these multigrained segmentation maps for ?nal road segmentation. Bene?tting from multigrained segmentation, our HCN shows impressing results in processing both multiscale roads and complex background. Four testing indicators, including pixel accuracy, mean accuracy, mean region intersection over union (IU), and frequency weighted IU, are computed to evaluate the proposed HCN on two testing data sets. Compared with ?ve state-of-the-art road segmentation methods, our HCN has higher segmentation accuracy than them.
关键词： high-resolution visible remote sensing image,Convolutional neural network (CNN),road segmentation
更新于2025-09-23 15:21:21