Deep learning for typhoon intensity classification using satellite cloud images

AbstractTropical cyclone, also known as typhoon, is one of the most destructive weather phenomena. Its intense cyclonic eddy circulations often cause serious damages to coastal areas. Accurate classification or prediction for typhoon intensity is crucial to the disaster warning and mitigation management. But typhoon intensity-related feature extraction is a challenging task as it requires significant pre-processing and human intervention for analysis, and its recognition rate is poor due to various physical factors such as tropical disturbance. In this study, we built a Typhoon-CNNs framework, an automatic classifier for typhoon intensity based on convolutional neural network (CNN). Typhoon-CNNs framework utilized a cyclical convolution strategy supplemented with dropout zero-set, which extracted sensitive features of existing spiral cloud band (SCB) more effectively and reduces over-fitting phenomenon. To further optimize the performance of Typhoon-CNNs, we also proposed the improved activation function (T-ReLU) and the loss function (CE-FMCE). The improved Typhoon-CNNs was trained and validated using more than 10,000 multiple sensor satellite cloud images of National Institute of Informatics. The classification accuracy reached to 88.74%. Compared with other deep learning methods, the accuracy of our improved Typhoon-CNNs was 7.43% higher than ResNet50, 10.27% higher than InceptionV3 and 14.71% higher than VGG16. Finally, by visualizing hierarchic feature maps derived from Typhoon-CNNs, we can easily identify the sensitive characteristics such as typhoon eyes, dense-shadowing cloud areas and SCBs, which facilitates classify and forecast typhoon intensity.

Download Full-text

Domain knowledge integration into deep learning for typhoon intensity classification

Scientific Reports ◽

10.1038/s41598-021-92286-w ◽

2021 ◽

Vol 11 (1) ◽

Author(s):

Maiki Higa ◽

Shinya Tanahara ◽

Yoshitaka Adachi ◽

Natsumi Ishiki ◽

Shin Nakama ◽

...

Keyword(s):

Deep Learning ◽

Domain Knowledge ◽

Knowledge Integration ◽

Cluster Formation ◽

Satellite Image ◽

Image Features ◽

Feature Maps ◽

Typhoon Intensity ◽

Intensity Class ◽

Intensity Classification

AbstractIn this report, we propose a deep learning technique for high-accuracy estimation of the intensity class of a typhoon from a single satellite image, by incorporating meteorological domain knowledge. By using the Visual Geometric Group’s model, VGG-16, with images preprocessed with fisheye distortion, which enhances a typhoon’s eye, eyewall, and cloud distribution, we achieved much higher classification accuracy than that of a previous study, even with sequential-split validation. Through comparison of t-distributed stochastic neighbor embedding (t-SNE) plots for the feature maps of VGG with the original satellite images, we also verified that the fisheye preprocessing facilitated cluster formation, suggesting that our model could successfully extract image features related to the typhoon intensity class. Moreover, gradient-weighted class activation mapping (Grad-CAM) was applied to highlight the eye and the cloud distributions surrounding the eye, which are important regions for intensity classification; the results suggest that our model qualitatively gained a viewpoint similar to that of domain experts. A series of analyses revealed that the data-driven approach using only deep learning has limitations, and the integration of domain knowledge could bring new breakthroughs.

Download Full-text

Deep Learning based Effective Steganalysis

International Journal of Innovative Technology and Exploring Engineering - Special Issue ◽

10.35940/ijitee.d1004.0394s220 ◽

2020 ◽

Vol 9 (4S2) ◽

pp. 82-85

Keyword(s):

Neural Networks ◽

Feature Extraction ◽

Deep Learning ◽

Convolutional Neural Networks ◽

Error Detection ◽

Activation Function ◽

Sigmoid Function ◽

Learning Networks ◽

Feature Maps ◽

Gating Mechanism

There is an evident paradigm shift in steganalysis techniques with discovery of deep learning networks. As steganalysis is a classification task, it is done by machine learning classifiers and ensembles of them. But with the proliferation of deep learning and Convolutional Neural Networks in many areas, the performance of steganalysis techniques have jumped up to a another high, because of the application of Convolutional Neural Networks. The traditional steganalysis techniques consists two important steps, i.e., feature extraction and classification; where as deep learning networks learn the features automatically, eliminating the need of extraction of handcrafted features. Because of this feature CNNs were highly successful in image recognition and image classification techniques. In addition to that, feature extraction and classification are combined together in deep learning hence classification would be more effective because of the learning of the features which are really important for classification. But in Steganalysis the task is to detect very subtle and weak noise created by the hidden data with steganography techniques. We have designed a deep CNN architecture customized for steganalysis task based on existing residual neural networks frame. We have introduced a descriptor to capture the inter pixel dependencies and which acts as an indicator for weightage of a particular feature maps. Thus the classifier can give more weightage to effective feature maps instead of treating all the feature maps equally. We have also used a gating mechanism by using sigmoid function after nonlinear activation function sandwiched between two fully connected layers. This enhancement to the existing deep residual neural networks has given better results in terms of error detection rate compared to the other deep learning based steganalysis techniques.

Download Full-text

A Novel Parameter Initialization Technique Using RBM-NN for Human Action Recognition

Computational Intelligence and Neuroscience ◽

10.1155/2020/8852404 ◽

2020 ◽

Vol 2020 ◽

pp. 1-30

Author(s):

Deepika Roselind Johnson ◽

V.Rhymend Uthariaraj

Keyword(s):

Deep Learning ◽

Action Recognition ◽

Recognition Rate ◽

Human Action Recognition ◽

Activation Function ◽

Human Action ◽

Learning Technologies ◽

Svm Classifier ◽

Global Features ◽

Benchmark Datasets

Human action recognition is a trending topic in the field of computer vision and its allied fields. The goal of human action recognition is to identify any human action that takes place in an image or a video dataset. For instance, the actions include walking, running, jumping, throwing, and much more. Existing human action recognition techniques have their own set of limitations when it concerns model accuracy and flexibility. To overcome these limitations, deep learning technologies were implemented. In the deep learning approach, a model learns by itself to improve its recognition accuracy and avoids problems such as gradient eruption, overfitting, and underfitting. In this paper, we propose a novel parameter initialization technique using the Maxout activation function. Firstly, human action is detected and tracked from the video dataset to learn the spatial-temporal features. Secondly, the extracted feature descriptors are trained using the RBM-NN. Thirdly, the local features are encoded into global features using an integrated forward and backward propagation process via RBM-NN. Finally, an SVM classifier recognizes the human actions in the video dataset. The experimental analysis performed on various benchmark datasets showed an improved recognition rate when compared to other state-of-the-art learning models.

Download Full-text

A Lightweight Fusion Distillation Network for Image Deblurring and Deraining

Sensors ◽

10.3390/s21165312 ◽

2021 ◽

Vol 21 (16) ◽

pp. 5312

Author(s):

Yanni Zhang ◽

Yiming Liu ◽

Qiang Li ◽

Jianzhong Wang ◽

Miao Qi ◽

...

Keyword(s):

Deep Learning ◽

Image Deblurring ◽

Image Features ◽

Image Feature ◽

Model Complexity ◽

Small Scale ◽

Feature Maps ◽

Learning Framework ◽

Channel Information ◽

Scale Spaces

Recently, deep learning-based image deblurring and deraining have been well developed. However, most of these methods fail to distill the useful features. What is more, exploiting the detailed image features in a deep learning framework always requires a mass of parameters, which inevitably makes the network suffer from a high computational burden. We propose a lightweight fusion distillation network (LFDN) for image deblurring and deraining to solve the above problems. The proposed LFDN is designed as an encoder–decoder architecture. In the encoding stage, the image feature is reduced to various small-scale spaces for multi-scale information extraction and fusion without much information loss. Then, a feature distillation normalization block is designed at the beginning of the decoding stage, which enables the network to distill and screen valuable channel information of feature maps continuously. Besides, an information fusion strategy between distillation modules and feature channels is also carried out by the attention mechanism. By fusing different information in the proposed approach, our network can achieve state-of-the-art image deblurring and deraining results with a smaller number of parameters and outperform the existing methods in model complexity.

Download Full-text

Real-Time Environment Monitoring Using a Lightweight Image Super-Resolution Network

International Journal of Environmental Research and Public Health ◽

10.3390/ijerph18115890 ◽

2021 ◽

Vol 18 (11) ◽

pp. 5890

Author(s):

Qiang Yu ◽

Feiqiang Liu ◽

Long Xiao ◽

Zitao Liu ◽

Xiaomin Yang

Keyword(s):

Deep Learning ◽

Real Time ◽

Super Resolution ◽

Model Complexity ◽

Practical Application ◽

Single Image ◽

Feature Maps ◽

Benchmark Datasets ◽

Image Super Resolution ◽

Single Image Super Resolution

Deep-learning (DL)-based methods are of growing importance in the field of single image super-resolution (SISR). The practical application of these DL-based models is a remaining problem due to the requirement of heavy computation and huge storage resources. The powerful feature maps of hidden layers in convolutional neural networks (CNN) help the model learn useful information. However, there exists redundancy among feature maps, which can be further exploited. To address these issues, this paper proposes a lightweight efficient feature generating network (EFGN) for SISR by constructing the efficient feature generating block (EFGB). Specifically, the EFGB can conduct plain operations on the original features to produce more feature maps with parameters slightly increasing. With the help of these extra feature maps, the network can extract more useful information from low resolution (LR) images to reconstruct the desired high resolution (HR) images. Experiments conducted on the benchmark datasets demonstrate that the proposed EFGN can outperform other deep-learning based methods in most cases and possess relatively lower model complexity. Additionally, the running time measurement indicates the feasibility of real-time monitoring.

Download Full-text

Image Enhanced Mask R-CNN: A Deep Learning Pipeline with New Evaluation Measures for Wind Turbine Blade Defect Detection and Classification

Journal of Imaging ◽

10.3390/jimaging7030046 ◽

2021 ◽

Vol 7 (3) ◽

pp. 46

Author(s):

Jiajun Zhang ◽

Georgina Cosma ◽

Jason Watkins

Keyword(s):

Deep Learning ◽

Wind Turbine ◽

Turbine Blade ◽

Defect Detection ◽

Recognition Rate ◽

Weighted Average ◽

Wind Turbine Blade ◽

Evaluation Measures ◽

Augmentation Techniques ◽

The Mean

Demand for wind power has grown, and this has increased wind turbine blade (WTB) inspections and defect repairs. This paper empirically investigates the performance of state-of-the-art deep learning algorithms, namely, YOLOv3, YOLOv4, and Mask R-CNN for detecting and classifying defects by type. The paper proposes new performance evaluation measures suitable for defect detection tasks, and these are: Prediction Box Accuracy, Recognition Rate, and False Label Rate. Experiments were carried out using a dataset, provided by the industrial partner, that contains images from WTB inspections. Three variations of the dataset were constructed using different image augmentation settings. Results of the experiments revealed that on average, across all proposed evaluation measures, Mask R-CNN outperformed all other algorithms when transformation-based augmentations (i.e., rotation and flipping) were applied. In particular, when using the best dataset, the mean Weighted Average (mWA) values (i.e., mWA is the average of the proposed measures) achieved were: Mask R-CNN: 86.74%, YOLOv3: 70.08%, and YOLOv4: 78.28%. The paper also proposes a new defect detection pipeline, called Image Enhanced Mask R-CNN (IE Mask R-CNN), that includes the best combination of image enhancement and augmentation techniques for pre-processing the dataset, and a Mask R-CNN model tuned for the task of WTB defect detection and classification.

Download Full-text

Inverse design of layered periodic wave barriers based on deep learning

Proceedings of the Institution of Mechanical Engineers Part L Journal of Materials Design and Applications ◽

10.1177/14644207211016886 ◽

2021 ◽

pp. 146442072110168

Author(s):

Chen-Xu Liu ◽

Gui-Lan Yu

Keyword(s):

Deep Learning ◽

Activation Function ◽

Periodic Wave ◽

P Wave ◽

Soil Parameters ◽

S Wave ◽

Output Frequency ◽

Great Generality ◽

Typical Range ◽

Deep Learning Model

This study presents an approach based on deep learning to design layered periodic wave barriers with consideration of typical range of soil parameters. Three cases are considered where P wave and S wave exist separately or simultaneously. The deep learning model is composed of an autoencoder with a pretrained decoder which has three branches to output frequency attenuation domains for three different cases. A periodic activation function is used to improve the design accuracy, and condition variables are applied in the code layer of the autoencoder to meet the requirements of practical multi working conditions. Forty thousand sets of data are generated to train, validate, and test the model, and the designed results are highly consistent with the targets. The presented approach has great generality, feasibility, rapidity, and accuracy on designing layered periodic wave barriers which exhibit good performance in wave suppression in targeted frequency range.

Download Full-text

Pedestrian detection algorithm in traffic scene based on weakly supervised hierarchical deep model

International Journal of Advanced Robotic Systems ◽

10.1177/1729881417692311 ◽

2016 ◽

Vol 14 (1) ◽

pp. 172988141769231 ◽

Cited By ~ 2

Author(s):

Yingfeng Cai ◽

Youguo He ◽

Hai Wang ◽

Xiaoqiang Sun ◽

Long Chen ◽

...

Keyword(s):

Deep Learning ◽

Pedestrian Detection ◽

Recognition Rate ◽

Detection Algorithm ◽

Training Methods ◽

Two Dimensional ◽

Data Set ◽

Deep Model ◽

Unsupervised Training ◽

Weakly Supervised

The emergence and development of deep learning theory in machine learning field provide new method for visual-based pedestrian recognition technology. To achieve better performance in this application, an improved weakly supervised hierarchical deep learning pedestrian recognition algorithm with two-dimensional deep belief networks is proposed. The improvements are made by taking into consideration the weaknesses of structure and training methods of existing classifiers. First, traditional one-dimensional deep belief network is expanded to two-dimensional that allows image matrix to be loaded directly to preserve more information of a sample space. Then, a determination regularization term with small weight is added to the traditional unsupervised training objective function. By this modification, original unsupervised training is transformed to weakly supervised training. Subsequently, that gives the extracted features discrimination ability. Multiple sets of comparative experiments show that the performance of the proposed algorithm is better than other deep learning algorithms in recognition rate and outperforms most of the existing state-of-the-art methods in non-occlusion pedestrian data set while performs fair in weakly and heavily occlusion data set.

Download Full-text

Automated Sensing System for Real-Time Recognition of Trucks in River Dredging Areas Using Computer Vision and Convolutional Deep Learning

Sensors ◽

10.3390/s21020555 ◽

2021 ◽

Vol 21 (2) ◽

pp. 555

Author(s):

Jui-Sheng Chou ◽

Chia-Hsuan Liu

Keyword(s):

Deep Learning ◽

Real Time ◽

Character Recognition ◽

Initial Period ◽

Control Point ◽

Recognition Rate ◽

Classification Model ◽

License Plate ◽

Construction Site ◽

Illegal Mining

Sand theft or illegal mining in river dredging areas has been a problem in recent decades. For this reason, increasing the use of artificial intelligence in dredging areas, building automated monitoring systems, and reducing human involvement can effectively deter crime and lighten the workload of security guards. In this investigation, a smart dredging construction site system was developed using automated techniques that were arranged to be suitable to various areas. The aim in the initial period of the smart dredging construction was to automate the audit work at the control point, which manages trucks in river dredging areas. Images of dump trucks entering the control point were captured using monitoring equipment in the construction area. The obtained images and the deep learning technique, YOLOv3, were used to detect the positions of the vehicle license plates. Framed images of the vehicle license plates were captured and were used as input in an image classification model, C-CNN-L3, to identify the number of characters on the license plate. Based on the classification results, the images of the vehicle license plates were transmitted to a text recognition model, R-CNN-L3, that corresponded to the characters of the license plate. Finally, the models of each stage were integrated into a real-time truck license plate recognition (TLPR) system; the single character recognition rate was 97.59%, the overall recognition rate was 93.73%, and the speed was 0.3271 s/image. The TLPR system reduces the labor force and time spent to identify the license plates, effectively reducing the probability of crime and increasing the transparency, automation, and efficiency of the frontline personnel’s work. The TLPR is the first step toward an automated operation to manage trucks at the control point. The subsequent and ongoing development of system functions can advance dredging operations toward the goal of being a smart construction site. By intending to facilitate an intelligent and highly efficient management system of dredging-related departments by providing a vehicle LPR system, this paper forms a contribution to the current body of knowledge in the sense that it presents an objective approach for the TLPR system.

Download Full-text

ResBCDU-Net: A Deep Learning Framework for Lung CT Image Segmentation

Sensors ◽

10.3390/s21010268 ◽

2021 ◽

Vol 21 (1) ◽

pp. 268

Author(s):

Yeganeh Jalali ◽

Mansoor Fateh ◽

Mohsen Rezvani ◽

Vahid Abolghasemi ◽

Mohammad Hossein Anisi

Keyword(s):

Image Segmentation ◽

Deep Learning ◽

Ct Images ◽

Medical Image Segmentation ◽

Great Success ◽

Morphological Operations ◽

Ct Image ◽

Feature Maps ◽

Learning Framework ◽

Lung Ct

Lung CT image segmentation is a key process in many applications such as lung cancer detection. It is considered a challenging problem due to existing similar image densities in the pulmonary structures, different types of scanners, and scanning protocols. Most of the current semi-automatic segmentation methods rely on human factors therefore it might suffer from lack of accuracy. Another shortcoming of these methods is their high false-positive rate. In recent years, several approaches, based on a deep learning framework, have been effectively applied in medical image segmentation. Among existing deep neural networks, the U-Net has provided great success in this field. In this paper, we propose a deep neural network architecture to perform an automatic lung CT image segmentation process. In the proposed method, several extensive preprocessing techniques are applied to raw CT images. Then, ground truths corresponding to these images are extracted via some morphological operations and manual reforms. Finally, all the prepared images with the corresponding ground truth are fed into a modified U-Net in which the encoder is replaced with a pre-trained ResNet-34 network (referred to as Res BCDU-Net). In the architecture, we employ BConvLSTM (Bidirectional Convolutional Long Short-term Memory)as an advanced integrator module instead of simple traditional concatenators. This is to merge the extracted feature maps of the corresponding contracting path into the previous expansion of the up-convolutional layer. Finally, a densely connected convolutional layer is utilized for the contracting path. The results of our extensive experiments on lung CT images (LIDC-IDRI database) confirm the effectiveness of the proposed method where a dice coefficient index of 97.31% is achieved.

Download Full-text