scholarly journals Remote Sensing Scene Classification and Explanation Using RSSCNet and LIME

2020 ◽  
Vol 10 (18) ◽  
pp. 6151
Author(s):  
Sheng-Chieh Hung ◽  
Hui-Ching Wu ◽  
Ming-Hseng Tseng
Keyword(s):  
Remote Sensing ◽  
Traffic Control ◽  
Network Models ◽  
Scene Classification ◽  
Neural Network Models ◽  
Proposed Model ◽  
Rate Policy ◽  
Speed Up ◽  
Interpretable Model ◽  
Model Training

Classification is needed in disaster investigation, traffic control, and land-use resource management. How to quickly and accurately classify such remote sensing imagery has become a popular research topic. However, the application of large, deep neural network models for the training of classifiers in the hope of obtaining good classification results is often very time-consuming. In this study, a new CNN (convolutional neutral networks) architecture, i.e., RSSCNet (remote sensing scene classification network), with high generalization capability was designed. Moreover, a two-stage cyclical learning rate policy and the no-freezing transfer learning method were developed to speed up model training and enhance accuracy. In addition, the manifold learning t-SNE (t-distributed stochastic neighbor embedding) algorithm was used to verify the effectiveness of the proposed model, and the LIME (local interpretable model, agnostic explanation) algorithm was applied to improve the results in cases where the model made wrong predictions. Comparing the results of three publicly available datasets in this study with those obtained in previous studies, the experimental results show that the model and method proposed in this paper can achieve better scene classification more quickly and more efficiently.

scholarly journals Convolutional Neural Network for Remote-Sensing Scene Classification: Transfer Learning Analysis

2019 ◽  
Vol 12 (1) ◽  
pp. 86 ◽  
Author(s):  
Rafael Pires de Lima ◽  
Kurt Marfurt
Keyword(s):  
Neural Network ◽  
Remote Sensing ◽  
Convolutional Neural Network ◽  
Transfer Learning ◽  
Network Models ◽  
Remote Sensing Image ◽  
Natural Images ◽  
Natural Image ◽  
Scene Classification ◽  
Neural Network Models

Remote-sensing image scene classification can provide significant value, ranging from forest fire monitoring to land-use and land-cover classification. Beginning with the first aerial photographs of the early 20th century to the satellite imagery of today, the amount of remote-sensing data has increased geometrically with a higher resolution. The need to analyze these modern digital data motivated research to accelerate remote-sensing image classification. Fortunately, great advances have been made by the computer vision community to classify natural images or photographs taken with an ordinary camera. Natural image datasets can range up to millions of samples and are, therefore, amenable to deep-learning techniques. Many fields of science, remote sensing included, were able to exploit the success of natural image classification by convolutional neural network models using a technique commonly called transfer learning. We provide a systematic review of transfer learning application for scene classification using different datasets and different deep-learning models. We evaluate how the specialization of convolutional neural network models affects the transfer learning process by splitting original models in different points. As expected, we find the choice of hyperparameters used to train the model has a significant influence on the final performance of the models. Curiously, we find transfer learning from models trained on larger, more generic natural images datasets outperformed transfer learning from models trained directly on smaller remotely sensed datasets. Nonetheless, results show that transfer learning provides a powerful tool for remote-sensing scene classification.

scholarly journals Road Extraction from Very-High-Resolution Remote Sensing Images via a Nested SE-Deeplab Model

2020 ◽  
Vol 12 (18) ◽  
pp. 2985 ◽  
Author(s):  
Yeneng Lin ◽  
Dongyun Xu ◽  
Nan Wang ◽  
Zhou Shi ◽  
Qiuxiao Chen
Keyword(s):  
Remote Sensing ◽  
Deep Learning ◽  
High Resolution ◽  
Network Models ◽  
Road Extraction ◽  
Remote Sensing Images ◽  
Proposed Model ◽  
Wide Range ◽  
Network Modules ◽  
Very High

Automatic road extraction from very-high-resolution remote sensing images has become a popular topic in a wide range of fields. Convolutional neural networks are often used for this purpose. However, many network models do not achieve satisfactory extraction results because of the elongated nature and varying sizes of roads in images. To improve the accuracy of road extraction, this paper proposes a deep learning model based on the structure of Deeplab v3. It incorporates squeeze-and-excitation (SE) module to apply weights to different feature channels, and performs multi-scale upsampling to preserve and fuse shallow and deep information. To solve the problems associated with unbalanced road samples in images, different loss functions and backbone network modules are tested in the model’s training process. Compared with cross entropy, dice loss can improve the performance of the model during training and prediction. The SE module is superior to ResNext and ResNet in improving the integrity of the extracted roads. Experimental results obtained using the Massachusetts Roads Dataset show that the proposed model (Nested SE-Deeplab) improves F1-Score by 2.4% and Intersection over Union by 2.0% compared with FC-DenseNet. The proposed model also achieves better segmentation accuracy in road extraction compared with other mainstream deep-learning models including Deeplab v3, SegNet, and UNet.

Identification of Tomato Leaf Disease Detection using Pretrained Deep Convolutional Neural Network Models

2020 ◽  
Vol 21 (4) ◽  
pp. 625-635
Author(s):  
Anandhakrishnan T ◽  
Jaisakthi S.M Murugaiyan
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Early Stage ◽  
Computing Time ◽  
Network Models ◽  
Deep Convolutional Neural Network ◽  
Neural Network Models ◽  
Leaf Disease ◽  
Proposed Model ◽  
Deep Cnn

In this paper, we proposed a plant leaf disease identification model based on a Pretrained deep convolutional neural network (Deep CNN). The Deep CNN model is trained using an open dataset with 10 different classes of tomato leaves We observed that overall architectures which can increase the best performance of the model. The proposed model was trained using different training epochs, batch sizes and dropouts. The Xception has attained maximum accuracy compare with all other approaches. After an extensive simulation, the proposed model achieves classification accuracy better. This accuracy of the proposed work is greater than the accuracy of all other Pretrained approaches. The proposed model is also tested with respect to its consistency and reliability. The set of data used for this work was collected from the plant village dataset, including sick and healthy images. Models for detection of plant disease should predict the disease quickly and accurately in the early stage itself so that a proper precautionary measures can be applied to avoid further spread of the diseases. So, to reduce the main issue about the leaf diseases, we can analyze distinct kinds of deep neural network architectures in this research. From the outcomes, Xception has a constantly improving more to enhance the accuracy by increasing the number of epochs, without any indications of overfitting and decreasein quality. And Xception also generated a fine 99.45% precision in less computing time.

scholarly journals A Novel CapsNet based Image Reconstruction and Regression Analysis

2020 ◽  
Vol 2 (3) ◽  
pp. 156-164 ◽  
Author(s):  
Dr. Akey Sungheetha ◽  
Dr. Rajesh Sharma R
Keyword(s):  
Neural Network ◽  
Regression Analysis ◽  
Network Models ◽  
Superior Performance ◽  
Neural Network Models ◽  
Innovative Method ◽  
Proposed Model ◽  
Retrieval Efficiency ◽  
High Level ◽  
Better Than

In the field of image processing, all types of computation models are almost evolved to solve the issues through encoded neurons. However, compared with decoding orientation and regression analysis, still the doors are open due to its complexity. At present technologies uses two steps such as, decoding the intermediate terms and reconstruction using decoded information. The performance in terms of regression analysis is lagging due to the decoded intermediate terms. Conventional neural network models perform better in feature classification and representation, though the performance is reduced while handling high level features. Considering these issues in image classification and regression, the proposed model is designed with capsule network as an innovative method which is suitable to handle high level features. The experimental results of the proposed model are compared with conventional neural network models such as BPNN and CNN to validate the superior performance. The proposed model achieves better retrieval efficiency of 95.4% which is much better than other neural network models.

scholarly journals Image-based taxonomic classification of bulk biodiversity samples using deep learning and domain adaptation

2021 ◽  
Author(s):  
Tomochika Fujisawa ◽  
Victor Noguerales ◽  
Emmanouil Meramveliotakis ◽  
Anna Papadopoulou ◽  
Alfried P Vogler
Keyword(s):  
Deep Learning ◽  
High Throughput ◽  
Domain Adaptation ◽  
Network Models ◽  
Neural Network Models ◽  
Data Set ◽  
Model Training ◽  
Trained Neural Network ◽  
Domain Transfer

Complex bulk samples of invertebrates from biodiversity surveys present a great challenge for taxonomic identification, especially if obtained from unexplored ecosystems. High-throughput imaging combined with machine learning for rapid classification could overcome this bottleneck. Developing such procedures requires that taxonomic labels from an existing source data set are used for model training and prediction of an unknown target sample. Yet the feasibility of transfer learning for the classification of unknown samples remains to be tested. Here, we assess the efficiency of deep learning and domain transfer algorithms for family-level classification of below-ground bulk samples of Coleoptera from understudied forests of Cyprus. We trained neural network models with images from local surveys versus global databases of above-ground samples from tropical forests and evaluated how prediction accuracy was affected by: (a) the quality and resolution of images, (b) the size and complexity of the training set and (c) the transferability of identifications across very disparate source-target pairs that do not share any species or genera. Within-dataset classification accuracy reached 98% and depended on the number and quality of training images and on dataset complexity. The accuracy of between-datasets predictions was reduced to a maximum of 82% and depended greatly on the standardisation of the imaging procedure. When the source and target images were of similar quality and resolution, albeit from different faunas, the reduction of accuracy was minimal. Application of algorithms for domain adaptation significantly improved the prediction performance of models trained by non-standardised, low-quality images. Our findings demonstrate that existing databases can be used to train models and successfully classify images from unexplored biota, when the imaging conditions and classification algorithms are carefully considered. Also, our results provide guidelines for data acquisition and algorithmic development for high-throughput image-based biodiversity surveys.

scholarly journals Remote Sensing Image Scene Classification Based on Global Self-Attention Module

2021 ◽  
Vol 13 (22) ◽  
pp. 4542
Author(s):  
Qingwen Li ◽  
Dongmei Yan ◽  
Wanrong Wu
Keyword(s):  
Remote Sensing ◽  
Semantic Information ◽  
Remote Sensing Image ◽  
Global Information ◽  
Classification Methods ◽  
Scene Classification ◽  
Local Perception ◽  
Proposed Model ◽  
Public Datasets ◽  
Fully Connected

The complexity of scene images makes the research on remote-sensing image scene classification challenging. With the wide application of deep learning in recent years, many remote-sensing scene classification methods using a convolutional neural network (CNN) have emerged. Current CNN usually output global information by integrating the depth features extricated from the convolutional layer through the fully connected layer; however, the global information extracted is not comprehensive. This paper proposes an improved remote-sensing image scene classification method based on a global self-attention module to address this problem. The global information is derived from the depth characteristics extracted by the CNN. In order to better express the semantic information of the remote-sensing image, the multi-head self-attention module is introduced for global information augmentation. Meanwhile, the local perception unit is utilized to improve the self-attention module’s representation capabilities for local objects. The proposed method’s effectiveness is validated through comparative experiments with various training ratios and different scales on public datasets (UC Merced, AID, and NWPU-NESISC45). The precision of our proposed model is significantly improved compared to other methods for remote-sensing image scene classification.

A Novel Hybrid Model Using RBF and PSO for Net Asset Value Prediction

2018 ◽  
pp. 54-72
Author(s):  
C. M. Anish ◽  
Babita Majhi ◽  
Ritanjali Majhi
Keyword(s):  
Neural Network ◽  
Network Models ◽  
Performance Measure ◽  
Neural Network Models ◽  
Proposed Model ◽  
Derivative Free ◽  
Asset Value ◽  
Net Asset Value ◽  
Small Investors ◽  
The Cost

Net asset value (NAV) prediction is an important area of research as small investors are doing investment in there, Literature survey reveals that very little work has been done in this field. The reported literature mainly used various neural network models for NAV prediction. But the derivative based learning algorithms of these reported models have the problem of trapping into the local solution. Hence in chapter derivative free algorithm, particle swarm optimization is used to update the parameters of radial basis function neural network for prediction of NAV. The positions of particles represent the centers, spreads and weights of the RBF model and the minimum MSE is used as the cost function. The convergence characteristics are obtained to show the performance of the model during training phase. The MAPE and RMSE value are calculated during testing phase to show the performance of the proposed RBF-PSO model. These performance measure exhibits that the proposed model is better model in comparison to MLANN, FLANN and RBFNN models.

A Novel Hybrid Model Using RBF and PSO for Net Asset Value Prediction

2018 ◽  
pp. 1031-1049 ◽  
Author(s):  
C. M. Anish ◽  
Babita Majhi ◽  
Ritanjali Majhi
Keyword(s):  
Neural Network ◽  
Network Models ◽  
Performance Measure ◽  
Neural Network Models ◽  
Proposed Model ◽  
Derivative Free ◽  
Asset Value ◽  
Net Asset Value ◽  
Small Investors ◽  
The Cost

Net asset value (NAV) prediction is an important area of research as small investors are doing investment in there, Literature survey reveals that very little work has been done in this field. The reported literature mainly used various neural network models for NAV prediction. But the derivative based learning algorithms of these reported models have the problem of trapping into the local solution. Hence in chapter derivative free algorithm, particle swarm optimization is used to update the parameters of radial basis function neural network for prediction of NAV. The positions of particles represent the centers, spreads and weights of the RBF model and the minimum MSE is used as the cost function. The convergence characteristics are obtained to show the performance of the model during training phase. The MAPE and RMSE value are calculated during testing phase to show the performance of the proposed RBF-PSO model. These performance measure exhibits that the proposed model is better model in comparison to MLANN, FLANN and RBFNN models.

Intelligent Modeling of Drilling Process: A Knowledge-Based Neural Strategy With Fuzzy Switches

2001 ◽  
Author(s):  
Ye Sheng ◽  
Oliver Nelles ◽  
Masayoshi Tomizuka
Keyword(s):  
Thrust Force ◽  
Network Models ◽  
Drilling Process ◽  
Neural Network Models ◽  
Intelligent Modeling ◽  
Knowledge Based ◽  
Gain Variation ◽  
Proposed Model ◽  
Modeling Strategy ◽  
Wide Operating

Abstract In this paper, an intelligent modeling strategy for thrust force in drilling process is developed. Neural network models and a fuzzy switching strategy are presented to deal with the gain variation problem due to transitions from one drilling stage to another. Gain variation due to drill wear and the related modeling strategy are studied. Simulation and experimental results show that the proposed model works well over a wide operating range.

Sign in / Sign up

Export Citation Format

Share Document