scholarly journals Few-shot Classification of Aerial Scene Images via Meta-learning

2020 ◽  
Author(s):  
Pei Zhang ◽  
YIng Li ◽  
Dong Wang ◽  
Yunpeng Bai
Keyword(s):  
Classification Problem ◽  
Training Data ◽  
Scene Classification ◽  
Shot Classification ◽  
Training Stage ◽  
Testing Stage ◽  
Meta Learning ◽  
Cosine Distance ◽  
Query Sample

CNN-based methods have dominated the field of aerial scene classification for the past few years. While achieving remarkable success, CNN-based methods suffer from excessive parameters and notoriously rely on large amounts of training data. In this work, we introduce few-shot learning to the aerial scene classification problem. Few-shot learning aims to learn a model on base-set that can quickly adapt to unseen categories in novel-set, using only a few labeled samples. To this end, we proposed a meta-learning method for few-shot classification of aerial scene images. First, we train a feature extractor on all base categories to learn a representation of inputs. Then in the meta-training stage, the classifier is optimized in the metric space by cosine distance with a learnable scale parameter. At last, in the meta-testing stage, the query sample in the unseen category is predicted by the adapted classifier given a few support samples. We conduct extensive experiments on two challenging datasets: NWPU-RESISC45 and RSD46-WHU. The experimental results show that our method outperforms three state-of-the-art few-shot algorithms and one typical CNN-based method, D-CNN. Furthermore, several ablation experiments are conducted to investigate the effects of dataset scale and support shots; the experiment results confirm that our model is specifically effective in few-shot settings.

scholarly journals Few-Shot Classification of Aerial Scene Images via Meta-Learning

2020 ◽  
Vol 13 (1) ◽  
pp. 108
Author(s):  
Pei Zhang ◽  
Yunpeng Bai ◽  
Dong Wang ◽  
Bendu Bai ◽  
Ying Li
Keyword(s):  
Classification Problem ◽  
Training Data ◽  
Scene Classification ◽  
Shot Classification ◽  
Training Stage ◽  
Testing Stage ◽  
Meta Learning ◽  
Cosine Distance ◽  
The Impact

Convolutional neural network (CNN) based methods have dominated the field of aerial scene classification for the past few years. While achieving remarkable success, CNN-based methods suffer from excessive parameters and notoriously rely on large amounts of training data. In this work, we introduce few-shot learning to the aerial scene classification problem. Few-shot learning aims to learn a model on base-set that can quickly adapt to unseen categories in novel-set, using only a few labeled samples. To this end, we proposed a meta-learning method for few-shot classification of aerial scene images. First, we train a feature extractor on all base categories to learn a representation of inputs. Then in the meta-training stage, the classifier is optimized in the metric space by cosine distance with a learnable scale parameter. At last, in the meta-testing stage, the query sample in the unseen category is predicted by the adapted classifier given a few support samples. We conduct extensive experiments on two challenging datasets: NWPU-RESISC45 and RSD46-WHU. The experimental results show that our method yields state-of-the-art performance. Furthermore, several ablation experiments are conducted to investigate the effects of dataset scale, the impact of different metrics and the number of support shots; the experiment results confirm that our model is specifically effective in few-shot settings.

scholarly journals RS-SSKD: Self-Supervision Equipped with Knowledge Distillation for Few-Shot Remote Sensing Scene Classification

Sensors ◽  
2021 ◽  
Vol 21 (5) ◽  
pp. 1566
Author(s):  
Pei Zhang ◽  
Ying Li ◽  
Dong Wang ◽  
Jiyue Wang
Keyword(s):  
Remote Sensing ◽  
State Of The Art ◽  
Ground Truth ◽  
Training Data ◽  
Scene Classification ◽  
Training Time ◽  
Shot Classification ◽  
Meta Learning ◽  
Knowledge Distillation ◽  
Few Data

While growing instruments generate more and more airborne or satellite images, the bottleneck in remote sensing (RS) scene classification has shifted from data limits toward a lack of ground truth samples. There are still many challenges when we are facing unknown environments, especially those with insufficient training data. Few-shot classification offers a different picture under the umbrella of meta-learning: digging rich knowledge from a few data are possible. In this work, we propose a method named RS-SSKD for few-shot RS scene classification from a perspective of generating powerful representation for the downstream meta-learner. Firstly, we propose a novel two-branch network that takes three pairs of original-transformed images as inputs and incorporates Class Activation Maps (CAMs) to drive the network mining, the most relevant category-specific region. This strategy ensures that the network generates discriminative embeddings. Secondly, we set a round of self-knowledge distillation to prevent overfitting and boost the performance. Our experiments show that the proposed method surpasses current state-of-the-art approaches on two challenging RS scene datasets: NWPU-RESISC45 and RSD46-WHU. Finally, we conduct various ablation experiments to investigate the effect of each component of the proposed method and analyze the training time of state-of-the-art methods and ours.

scholarly journals Meta Learning for Few-Shot One-Class Classification

AI ◽  
2021 ◽  
Vol 2 (2) ◽  
pp. 195-208
Author(s):  
Gabriel Dahia ◽  
Maurício Pamplona Segundo
Keyword(s):  
Feature Representation ◽  
Support Vector ◽  
Support Vector Data Description ◽  
Target Class ◽  
Feature Representations ◽  
Shot Classification ◽  
Training Stage ◽  
Comparable Performance ◽  
Meta Learning ◽  
One Class Classification

We propose a method that can perform one-class classification given only a small number of examples from the target class and none from the others. We formulate the learning of meaningful features for one-class classification as a meta-learning problem in which the meta-training stage repeatedly simulates one-class classification, using the classification loss of the chosen algorithm to learn a feature representation. To learn these representations, we require only multiclass data from similar tasks. We show how the Support Vector Data Description method can be used with our method, and also propose a simpler variant based on Prototypical Networks that obtains comparable performance, indicating that learning feature representations directly from data may be more important than which one-class algorithm we choose. We validate our approach by adapting few-shot classification datasets to the few-shot one-class classification scenario, obtaining similar results to the state-of-the-art of traditional one-class classification, and that improves upon that of one-class classification baselines employed in the few-shot setting.

scholarly journals Meta-FSEO: A Meta-Learning Fast Adaptation with Self-Supervised Embedding Optimization for Few-Shot Remote Sensing Scene Classification

2021 ◽  
Vol 13 (14) ◽  
pp. 2776
Author(s):  
Yong Li ◽  
Zhenfeng Shao ◽  
Xiao Huang ◽  
Bowen Cai ◽  
Song Peng
Keyword(s):  
Remote Sensing ◽  
Deep Learning ◽  
Remote Sensing Data ◽  
Scene Classification ◽  
Generalization Capability ◽  
Urban Scenes ◽  
Shot Classification ◽  
Meta Learning ◽  
Land Surfaces ◽  
Fast Adaptation

The performance of deep learning is heavily influenced by the size of the learning samples, whose labeling process is time consuming and laborious. Deep learning algorithms typically assume that the training and prediction data are independent and uniformly distributed, which is rarely the case given the attributes and properties of different data sources. In remote sensing images, representations of urban land surfaces can vary across regions and by season, demanding rapid generalization of these surfaces in remote sensing data. In this study, we propose Meta-FSEO, a novel model for improving the performance of few-shot remote sensing scene classification in varying urban scenes. The proposed Meta-FSEO model deploys self-supervised embedding optimization for adaptive generalization in new tasks such as classifying features in new urban regions that have never been encountered during the training phase, thus balancing the requirements for feature classification tasks between multiple images collected at different times and places. We also created a loss function by weighting the contrast losses and cross-entropy losses. The proposed Meta-FSEO demonstrates a great generalization capability in remote sensing scene classification among different cities. In a five-way one-shot classification experiment with the Sentinel-1/2 Multi-Spectral (SEN12MS) dataset, the accuracy reached 63.08%. In a five-way five-shot experiment on the same dataset, the accuracy reached 74.29%. These results indicated that the proposed Meta-FSEO model outperformed both the transfer learning-based algorithm and two popular meta-learning-based methods, i.e., MAML and Meta-SGD.

scholarly journals The Artificial Power System Networks Stability Control Using the Technology of Neural Network

2019 ◽  
Vol 124 ◽  
pp. 05002
Author(s):  
A M-N Alzakkar ◽  
I.M. Valeev ◽  
N.P. Mestnikov ◽  
E.G. Nurullin
Keyword(s):  
Neural Network ◽  
Stability Control ◽  
Training Data ◽  
Electric Voltage ◽  
Normal Loading ◽  
Training Stage ◽  
Testing Stage ◽  
On Line ◽  
Two Stages ◽  
Speed And Accuracy

In the present work, the electric voltage stability at Muharda station in Syria was studied during the normal and up to normal loading states. The results were obtained using artificial neural network, which consists of three layers (input-hidden-output). This network is characterized by the speed and accuracy in processing before failure and supply turn-off, which may lead to economical problems. This study was carried out using two different generating schemes in this station (single double). The performance of this network consists of two stages: training stage (off-line) and testing stage (on-line), and a comparison between these stages is carried out, which leads to optimization the load in testing cases depending on the training data.

scholarly journals Prototype Calibration with Feature Generation for Few-Shot Remote Sensing Image Scene Classification

2021 ◽  
Vol 13 (14) ◽  
pp. 2728
Author(s):  
Qingjie Zeng ◽  
Jie Geng ◽  
Kai Huang ◽  
Wen Jiang ◽  
Jun Guo
Keyword(s):  
Remote Sensing ◽  
Remote Sensing Image ◽  
Classification Performance ◽  
Scene Classification ◽  
Remote Sensing Images ◽  
Training Set ◽  
Shot Classification ◽  
Support Set ◽  
Feature Expression

Few-shot classification of remote sensing images has attracted attention due to its important applications in various fields. The major challenge in few-shot remote sensing image scene classification is that limited labeled samples can be utilized for training. This may lead to the deviation of prototype feature expression, and thus the classification performance will be impacted. To solve these issues, a prototype calibration with a feature-generating model is proposed for few-shot remote sensing image scene classification. In the proposed framework, a feature encoder with self-attention is developed to reduce the influence of irrelevant information. Then, the feature-generating module is utilized to expand the support set of the testing set based on prototypes of the training set, and prototype calibration is proposed to optimize features of support images that can enhance the representativeness of each category features. Experiments on NWPU-RESISC45 and WHU-RS19 datasets demonstrate that the proposed method can yield superior classification accuracies for few-shot remote sensing image scene classification.

scholarly journals CNN for Image Identification of Hiragana Based on Pattern Recognition using CNN

2021 ◽  
Vol 6 (2) ◽  
pp. 62-71
Author(s):  
Chaerul Umam ◽  
Andi Danang Krismawan ◽  
Rabei Raad Ali
Keyword(s):  
Neural Network ◽  
Pattern Recognition ◽  
Convolutional Neural Network ◽  
Training Data ◽  
Training Process ◽  
Image Identification ◽  
Training Stage ◽  
Testing Stage ◽  
Testing Data ◽  
Network Method

Hiragana is one of the letters in Japanese. In this study, CNN (Convolutional Neural Network) method used as identication method, while he preprocessing used thresholding. Then carry out the normalization stage and the filtering stage to remove noise in the image. At the training stage use maxpooling and danse methods as a liaison in the training process, wherea in testing stage using the Adam Optimizer method. Here, we use 1000 images from 50 hiragana characters with a ratio of 950: 50, 950 as training data and 50 data as testing data. Our experiment yield accuracy in 95%.

scholarly journals A Dual Attention Network with Semantic Embedding for Few-Shot Learning

2019 ◽  
Vol 33 ◽  
pp. 9079-9086 ◽  
Author(s):  
Shipeng Yan ◽  
Songyang Zhang ◽  
Xuming He
Keyword(s):  
Complex Model ◽  
Training Data ◽  
Future Research ◽  
Learning Approaches ◽  
Learning Loss ◽  
Attention Network ◽  
Shot Classification ◽  
Recent Success ◽  
Meta Learning ◽  
Visual Concepts

Despite recent success of deep neural networks, it remains challenging to efficiently learn new visual concepts from limited training data. To address this problem, a prevailing strategy is to build a meta-learner that learns prior knowledge on learning from a small set of annotated data. However, most of existing meta-learning approaches rely on a global representation of images and a meta-learner with complex model structures, which are sensitive to background clutter and difficult to interpret. We propose a novel meta-learning method for few-shot classification based on two simple attention mechanisms: one is a spatial attention to localize relevant object regions and the other is a task attention to select similar training data for label prediction. We implement our method via a dual-attention network and design a semantic-aware meta-learning loss to train the meta-learner network in an end-to-end manner. We validate our model on three few-shot image classification datasets with extensive ablative study, and our approach shows competitive performances over these datasets with fewer parameters. For facilitating the future research, code and data split are available: https://github.com/tonysy/STANet-PyTorch

Sign in / Sign up

Export Citation Format

Share Document