AN APPLICATION OF SEGNET FOR DETECTING LANDSLIDE AREAS BY USING FULLY POLARIMETRIC SAR DATA

The study location of landslide is in Hokkaido, Japan which occurred due to the Iburi Earthquake 2018. In this study the landslide has been estimated by the fully Polarimetric SAR (Pol-SAR) technique based on ALOS-2 PALSAR-2 data using the Yamaguchi’s decomposition. The Yamaguchi's decomposition is proposed by Yoshio Yamaguchi et.al. The data has been analyzed using the deep learning process with SegNet architecture with color composite. In this research, the performance of SegNet is fast and efficient in memory usage. However, the result is not good, based on the Intersection over Union (IoU) evaluation obtained the lowest value is 0.0515 and the highest value is 0.1483. That is because of difficulty to make training datasets and of a small number of datasets. The greater difference between accuracy and loss graph along with higher epochs represents overfitting. The overfitting can be caused by the limited amount of training data and failure of the network to generalize the feature set over the training images.

Download Full-text

SAR Target Classification Based on Sample Spectral Regularization

Remote Sensing ◽

10.3390/rs12213628 ◽

2020 ◽

Vol 12 (21) ◽

pp. 3628

Author(s):

Wei Liang ◽

Tengfei Zhang ◽

Wenhui Diao ◽

Xian Sun ◽

Liangjin Zhao ◽

...

Keyword(s):

Deep Learning ◽

Transfer Learning ◽

Regularization Method ◽

Training Data ◽

Image Feature ◽

Sar Image ◽

Target Classification ◽

Spectral Regularization ◽

Sar Data ◽

Restriction Problem

Synthetic Aperture Radar (SAR) target classification is an important branch of SAR image interpretation. The deep learning based SAR target classification algorithms have made remarkable achievements. But the acquisition and annotation of SAR target images are time-consuming and laborious, and it is difficult to obtain sufficient training data in many cases. The insufficient training data can make deep learning based models suffering from over-fitting, which will severely limit their wide application in SAR target classification. Motivated by the above problem, this paper employs transfer-learning to transfer the prior knowledge learned from a simulated SAR dataset to a real SAR dataset. To overcome the sample restriction problem caused by the poor feature discriminability for real SAR data. A simple and effective sample spectral regularization method is proposed, which can regularize the singular values of each SAR image feature to improve the feature discriminability. Based on the proposed regularization method, we design a transfer-learning pipeline to leverage the simulated SAR data as well as acquire better feature discriminability. The experimental results indicate that the proposed method is feasible for the sample restriction problem in SAR target classification. Furthermore, the proposed method can improve the classification accuracy when relatively sufficient training data is available, and it can be plugged into any convolutional neural network (CNN) based SAR classification models.

Download Full-text

Adaptive Data Augmentation to Achieve Noise Robustness and Overcome Data Deficiency for Deep Learning

Applied Sciences ◽

10.3390/app11125586 ◽

2021 ◽

Vol 11 (12) ◽

pp. 5586

Author(s):

Eunkyeong Kim ◽

Jinyong Kim ◽

Hansoo Lee ◽

Sungshin Kim

Keyword(s):

Deep Learning ◽

Classification Accuracy ◽

Data Augmentation ◽

Robot Vision ◽

Training Data ◽

Noise Robustness ◽

Automatic Data ◽

Optimal Perturbation ◽

Training Images ◽

Smart Factories

Artificial intelligence technologies and robot vision systems are core technologies in smart factories. Currently, there is scholarly interest in automatic data feature extraction in smart factories using deep learning networks. However, sufficient training data are required to train these networks. In addition, barely perceptible noise can affect classification accuracy. Therefore, to increase the amount of training data and achieve robustness against noise attacks, a data augmentation method implemented using the adaptive inverse peak signal-to-noise ratio was developed in this study to consider the influence of the color characteristics of the training images. This method was used to automatically determine the optimal perturbation range of the color perturbation method for generating images using weights based on the characteristics of the training images. The experimental results showed that the proposed method could generate new training images from original images, classify noisy images with greater accuracy, and generally improve the classification accuracy. This demonstrates that the proposed method is effective and robust to noise, even when the training data are deficient.

Download Full-text

Real Time Face Recognition with limited training data: Feature Transfer Learning integrating CNN and Sparse Approximation

10.1101/2021.03.17.435457 ◽

2021 ◽

Author(s):

Gargi Mishra ◽

Supriya Bajpai

Keyword(s):

Deep Learning ◽

Face Recognition ◽

Real Time ◽

High Performance ◽

Sparse Approximation ◽

Training Data ◽

Test Image ◽

Learning Methods ◽

Training Time ◽

Training Images

It is highly challenging to obtain high performance with limited and unconstrained data in real time face recognition applications. Sparse Approximation is a fast and computationally efficient method for the above application as it requires no training time as compared to deep learning methods. It eliminates the training time by assuming that the test image can be approximated by the sum of individual contributions of the training images from different classes and the class with maximum contribution is closest to the test image. The efficiency of the Sparse Approximation method can be further increased by providing high quality features as input for classification. Hence, we propose to integrate pre-trained CNN architecture to extract the highly discriminative features from the image dataset for Sparse classification. The proposed approach provides better performance even for one training image per class in complex environment as compared to the existing methods. Highlight of the present approach is the results obtained for LFW dataset with one and thirteen training images per class are 84.86% and 96.14% respectively, whereas the existing deep learning methods use a large amount of training data to achieve comparable results.

Download Full-text

Zebrafish Larvae Phenotype Classification from Bright-field Microscopic Images Using a Two-Tier Deep-Learning Pipeline

Applied Sciences ◽

10.3390/app10041247 ◽

2020 ◽

Vol 10 (4) ◽

pp. 1247 ◽

Cited By ~ 3

Author(s):

Shang Shang ◽

Sijie Lin ◽

Fengyu Cong

Keyword(s):

Deep Learning ◽

Environmental Influence ◽

Training Data ◽

Convolutional Network ◽

Training Images ◽

Zebrafish Larvae ◽

Machine Learning Methods ◽

Phenotype Classification ◽

Convolution Kernels

Classification of different zebrafish larvae phenotypes is useful for studying the environmental influence on embryo development. However, the scarcity of well-annotated training images and fuzzy inter-phenotype differences hamper the application of machine-learning methods in phenotype classification. This study develops a deep-learning approach to address these challenging problems. A convolutional network model with compressed separable convolution kernels is adopted to address the overfitting issue caused by insufficient training data. A two-tier classification pipeline is designed to improve the classification accuracy based on fuzzy phenotype features. Our method achieved an averaged accuracy of 91% for all the phenotypes and maximum accuracy of 100% for some phenotypes (e.g., dead and chorion). We also compared our method with the state-of-the-art methods based on the same dataset. Our method obtained dramatic accuracy improvement up to 22% against the existing method. This study offers an effective deep-learning solution for classifying difficult zebrafish larvae phenotypes based on very limited training data.

Download Full-text

CNN performance dependence on linear image processing

Electronic Imaging ◽

10.2352/issn.2470-1173.2020.10.ipas-182 ◽

2020 ◽

Vol 2020 (10) ◽

pp. 310-1-310-7

Author(s):

Khalid Omer ◽

Luca Caucci ◽

Meredith Kupinski

Keyword(s):

Image Processing ◽

Texture Classification ◽

Full Rank ◽

Detection Performance ◽

Ideal Observer ◽

Training Data ◽

Image Texture ◽

Training Images ◽

Analytic Expressions ◽

Linear Compression

This work reports on convolutional neural network (CNN) performance on an image texture classification task as a function of linear image processing and number of training images. Detection performance of single and multi-layer CNNs (sCNN/mCNN) are compared to optimal observers. Performance is quantified by the area under the receiver operating characteristic (ROC) curve, also known as the AUC. For perfect detection AUC = 1.0 and AUC = 0.5 for guessing. The Ideal Observer (IO) maximizes AUC but is prohibitive in practice because it depends on high-dimensional image likelihoods. The IO performance is invariant to any fullrank, invertible linear image processing. This work demonstrates the existence of full-rank, invertible linear transforms that can degrade both sCNN and mCNN even in the limit of large quantities of training data. A subsequent invertible linear transform changes the images’ correlation structure again and can improve this AUC. Stationary textures sampled from zero mean and unequal covariance Gaussian distributions allow closed-form analytic expressions for the IO and optimal linear compression. Linear compression is a mitigation technique for high-dimension low sample size (HDLSS) applications. By definition, compression strictly decreases or maintains IO detection performance. For small quantities of training data, linear image compression prior to the sCNN architecture can increase AUC from 0.56 to 0.93. Results indicate an optimal compression ratio for CNN based on task difficulty, compression method, and number of training images.

Download Full-text

Siamese Reconstruction Network: Accurate Image Reconstruction from Human Brain Activity by Learning to Compare

Applied Sciences ◽

10.3390/app9224749 ◽

2019 ◽

Vol 9 (22) ◽

pp. 4749

Author(s):

Lingyun Jiang ◽

Kai Qiao ◽

Linyuan Wang ◽

Chi Zhang ◽

Jian Chen ◽

...

Keyword(s):

Deep Learning ◽

Human Brain ◽

Brain Activity ◽

Feature Space ◽

Training Data ◽

Reconstruction Method ◽

Learning Method ◽

Training Samples ◽

Visual Reconstruction ◽

Relationship Of

Decoding human brain activities, especially reconstructing human visual stimuli via functional magnetic resonance imaging (fMRI), has gained increasing attention in recent years. However, the high dimensionality and small quantity of fMRI data impose restrictions on satisfactory reconstruction, especially for the reconstruction method with deep learning requiring huge amounts of labelled samples. When compared with the deep learning method, humans can recognize a new image because our human visual system is naturally capable of extracting features from any object and comparing them. Inspired by this visual mechanism, we introduced the mechanism of comparison into deep learning method to realize better visual reconstruction by making full use of each sample and the relationship of the sample pair by learning to compare. In this way, we proposed a Siamese reconstruction network (SRN) method. By using the SRN, we improved upon the satisfying results on two fMRI recording datasets, providing 72.5% accuracy on the digit dataset and 44.6% accuracy on the character dataset. Essentially, this manner can increase the training data about from n samples to 2n sample pairs, which takes full advantage of the limited quantity of training samples. The SRN learns to converge sample pairs of the same class or disperse sample pairs of different class in feature space.

Download Full-text

Domain randomization-enhanced deep learning models for bird detection

Scientific Reports ◽

10.1038/s41598-020-80101-x ◽

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.

Download Full-text

U-Infuse: Democratization of Customizable Deep Learning for Object Detection

Sensors ◽

10.3390/s21082611 ◽

2021 ◽

Vol 21 (8) ◽

pp. 2611

Author(s):

Andrew Shepley ◽

Greg Falzon ◽

Christopher Lawson ◽

Paul Meek ◽

Paul Kwan

Keyword(s):

Deep Learning ◽

Intellectual Property ◽

Object Detection ◽

Image Data ◽

Learning Technologies ◽

Training Data ◽

Learning Models ◽

Ecological Data ◽

Single Class ◽

Large Numbers

Image data is one of the primary sources of ecological data used in biodiversity conservation and management worldwide. However, classifying and interpreting large numbers of images is time and resource expensive, particularly in the context of camera trapping. Deep learning models have been used to achieve this task but are often not suited to specific applications due to their inability to generalise to new environments and inconsistent performance. Models need to be developed for specific species cohorts and environments, but the technical skills required to achieve this are a key barrier to the accessibility of this technology to ecologists. Thus, there is a strong need to democratize access to deep learning technologies by providing an easy-to-use software application allowing non-technical users to train custom object detectors. U-Infuse addresses this issue by providing ecologists with the ability to train customised models using publicly available images and/or their own images without specific technical expertise. Auto-annotation and annotation editing functionalities minimize the constraints of manually annotating and pre-processing large numbers of images. U-Infuse is a free and open-source software solution that supports both multiclass and single class training and object detection, allowing ecologists to access deep learning technologies usually only available to computer scientists, on their own device, customised for their application, without sharing intellectual property or sensitive data. It provides ecological practitioners with the ability to (i) easily achieve object detection within a user-friendly GUI, generating a species distribution report, and other useful statistics, (ii) custom train deep learning models using publicly available and custom training data, (iii) achieve supervised auto-annotation of images for further training, with the benefit of editing annotations to ensure quality datasets. Broad adoption of U-Infuse by ecological practitioners will improve ecological image analysis and processing by allowing significantly more image data to be processed with minimal expenditure of time and resources, particularly for camera trap images. Ease of training and use of transfer learning means domain-specific models can be trained rapidly, and frequently updated without the need for computer science expertise, or data sharing, protecting intellectual property and privacy.

Download Full-text

Deep Learning-Based Differentiation between Mucinous Cystic Neoplasm and Serous Cystic Neoplasm in the Pancreas Using Endoscopic Ultrasonography

Diagnostics ◽

10.3390/diagnostics11061052 ◽

2021 ◽

Vol 11 (6) ◽

pp. 1052

Author(s):

Leang Sim Nguon ◽

Kangwon Seo ◽

Jung-Hyun Lim ◽

Tae-Jun Song ◽

Sung-Hyun Cho ◽

...

Keyword(s):

Decision Making ◽

Deep Learning ◽

Network Model ◽

Endoscopic Ultrasonography ◽

Data Augmentation ◽

Clinical Information ◽

Training Data ◽

Fine Tuning ◽

Cystic Neoplasm ◽

Cystic Neoplasms

Mucinous cystic neoplasms (MCN) and serous cystic neoplasms (SCN) account for a large portion of solitary pancreatic cystic neoplasms (PCN). In this study we implemented a convolutional neural network (CNN) model using ResNet50 to differentiate between MCN and SCN. The training data were collected retrospectively from 59 MCN and 49 SCN patients from two different hospitals. Data augmentation was used to enhance the size and quality of training datasets. Fine-tuning training approaches were utilized by adopting the pre-trained model from transfer learning while training selected layers. Testing of the network was conducted by varying the endoscopic ultrasonography (EUS) image sizes and positions to evaluate the network performance for differentiation. The proposed network model achieved up to 82.75% accuracy and a 0.88 (95% CI: 0.817–0.930) area under curve (AUC) score. The performance of the implemented deep learning networks in decision-making using only EUS images is comparable to that of traditional manual decision-making using EUS images along with supporting clinical information. Gradient-weighted class activation mapping (Grad-CAM) confirmed that the network model learned the features from the cyst region accurately. This study proves the feasibility of diagnosing MCN and SCN using a deep learning network model. Further improvement using more datasets is needed.

Download Full-text

Cotton Stand Counting from Unmanned Aerial System Imagery Using MobileNet and CenterNet Deep Learning Models

Remote Sensing ◽

10.3390/rs13142822 ◽

2021 ◽

Vol 13 (14) ◽

pp. 2822

Author(s):

Zhe Lin ◽

Wenxuan Guo

Keyword(s):

Deep Learning ◽

Cotton Plant ◽

Unmanned Aerial System ◽

Learning Models ◽

Training Images ◽

Testing Dataset ◽

Cotton Plants ◽

Detection And Counting ◽

Different Dimensions ◽

The Mean

An accurate stand count is a prerequisite to determining the emergence rate, assessing seedling vigor, and facilitating site-specific management for optimal crop production. Traditional manual counting methods in stand assessment are labor intensive and time consuming for large-scale breeding programs or production field operations. This study aimed to apply two deep learning models, the MobileNet and CenterNet, to detect and count cotton plants at the seedling stage with unmanned aerial system (UAS) images. These models were trained with two datasets containing 400 and 900 images with variations in plant size and soil background brightness. The performance of these models was assessed with two testing datasets of different dimensions, testing dataset 1 with 300 by 400 pixels and testing dataset 2 with 250 by 1200 pixels. The model validation results showed that the mean average precision (mAP) and average recall (AR) were 79% and 73% for the CenterNet model, and 86% and 72% for the MobileNet model with 900 training images. The accuracy of cotton plant detection and counting was higher with testing dataset 1 for both CenterNet and MobileNet models. The results showed that the CenterNet model had a better overall performance for cotton plant detection and counting with 900 training images. The results also indicated that more training images are required when applying object detection models on images with different dimensions from training datasets. The mean absolute percentage error (MAPE), coefficient of determination (R2), and the root mean squared error (RMSE) values of the cotton plant counting were 0.07%, 0.98 and 0.37, respectively, with testing dataset 1 for the CenterNet model with 900 training images. Both MobileNet and CenterNet models have the potential to accurately and timely detect and count cotton plants based on high-resolution UAS images at the seedling stage. This study provides valuable information for selecting the right deep learning tools and the appropriate number of training images for object detection projects in agricultural applications.

Download Full-text