Improving Image-Based Plant Disease Classification With Generative Adversarial Network Under Limited Training Set

Traditionally, plant disease recognition has mainly been done visually by human. It is often biased, time-consuming, and laborious. Machine learning methods based on plant leave images have been proposed to improve the disease recognition process. Convolutional neural networks (CNNs) have been adopted and proven to be very effective. Despite the good classification accuracy achieved by CNNs, the issue of limited training data remains. In most cases, the training dataset is often small due to significant effort in data collection and annotation. In this case, CNN methods tend to have the overfitting problem. In this paper, Wasserstein generative adversarial network with gradient penalty (WGAN-GP) is combined with label smoothing regularization (LSR) to improve the prediction accuracy and address the overfitting problem under limited training data. Experiments show that the proposed WGAN-GP enhanced classification method can improve the overall classification accuracy of plant diseases by 24.4% as compared to 20.2% using classic data augmentation and 22% using synthetic samples without LSR.

Download Full-text

An Imbalanced Image Classification Method for the Cell Cycle Phase

Information ◽

10.3390/info12060249 ◽

2021 ◽

Vol 12 (6) ◽

pp. 249

Author(s):

Xin Jin ◽

Yuanwen Zou ◽

Zhongbing Huang

Keyword(s):

Cell Cycle ◽

Deep Learning ◽

Image Classification ◽

Classification Accuracy ◽

Data Augmentation ◽

Cycle Phase ◽

Generative Adversarial Network ◽

Adversarial Network ◽

Cellular Life

The cell cycle is an important process in cellular life. In recent years, some image processing methods have been developed to determine the cell cycle stages of individual cells. However, in most of these methods, cells have to be segmented, and their features need to be extracted. During feature extraction, some important information may be lost, resulting in lower classification accuracy. Thus, we used a deep learning method to retain all cell features. In order to solve the problems surrounding insufficient numbers of original images and the imbalanced distribution of original images, we used the Wasserstein generative adversarial network-gradient penalty (WGAN-GP) for data augmentation. At the same time, a residual network (ResNet) was used for image classification. ResNet is one of the most used deep learning classification networks. The classification accuracy of cell cycle images was achieved more effectively with our method, reaching 83.88%. Compared with an accuracy of 79.40% in previous experiments, our accuracy increased by 4.48%. Another dataset was used to verify the effect of our model and, compared with the accuracy from previous results, our accuracy increased by 12.52%. The results showed that our new cell cycle image classification system based on WGAN-GP and ResNet is useful for the classification of imbalanced images. Moreover, our method could potentially solve the low classification accuracy in biomedical images caused by insufficient numbers of original images and the imbalanced distribution of original images.

Download Full-text

Underwater Acoustic Target Recognition Based on Generative Adversarial Network Data Augmentation

INTER-NOISE and NOISE-CON Congress and Conference Proceedings ◽

10.3397/in-2021-2737 ◽

2021 ◽

Vol 263 (2) ◽

pp. 4558-4564

Author(s):

Minghong Zhang ◽

Xinwei Luo

Keyword(s):

Data Augmentation ◽

Target Recognition ◽

Training Data ◽

Small Samples ◽

Generative Adversarial Network ◽

Data Set ◽

Underwater Acoustic ◽

Adversarial Network ◽

Acoustic Target ◽

The Impact

Underwater acoustic target recognition is an important aspect of underwater acoustic research. In recent years, machine learning has been developed continuously, which is widely and effectively applied in underwater acoustic target recognition. In order to acquire good recognition results and reduce the problem of overfitting, Adequate data sets are essential. However, underwater acoustic samples are relatively rare, which has a certain impact on recognition accuracy. In this paper, in addition of the traditional audio data augmentation method, a new method of data augmentation using generative adversarial network is proposed, which uses generator and discriminator to learn the characteristics of underwater acoustic samples, so as to generate reliable underwater acoustic signals to expand the training data set. The expanded data set is input into the deep neural network, and the transfer learning method is applied to further reduce the impact caused by small samples by fixing part of the pre-trained parameters. The experimental results show that the recognition result of this method is better than the general underwater acoustic recognition method, and the effectiveness of this method is verified.

Download Full-text

Urban Intersection Classification: A Comparative Analysis

Sensors ◽

10.3390/s21186269 ◽

2021 ◽

Vol 21 (18) ◽

pp. 6269

Author(s):

Augusto Luis Ballardini ◽

Álvaro Hernández Saz ◽

Sandra Carrasco Limeros ◽

Javier Lorenzo ◽

Ignacio Parra Alonso ◽

...

Keyword(s):

Ad Hoc ◽

Data Augmentation ◽

Comprehensive Evaluation ◽

Temporal Integration ◽

Individual Performance ◽

Training Data ◽

Real World Data ◽

Generative Adversarial Network ◽

Adversarial Network ◽

Integration Schemes

Understanding the scene in front of a vehicle is crucial for self-driving vehicles and Advanced Driver Assistance Systems, and in urban scenarios, intersection areas are one of the most critical, concentrating between 20% to 25% of road fatalities. This research presents a thorough investigation on the detection and classification of urban intersections as seen from onboard front-facing cameras. Different methodologies aimed at classifying intersection geometries have been assessed to provide a comprehensive evaluation of state-of-the-art techniques based on Deep Neural Network (DNN) approaches, including single-frame approaches and temporal integration schemes. A detailed analysis of most popular datasets previously used for the application together with a comparison with ad hoc recorded sequences revealed that the performances strongly depend on the field of view of the camera rather than other characteristics or temporal-integrating techniques. Due to the scarcity of training data, a new dataset is created by performing data augmentation from real-world data through a Generative Adversarial Network (GAN) to increase generalizability as well as to test the influence of data quality. Despite being in the relatively early stages, mainly due to the lack of intersection datasets oriented to the problem, an extensive experimental activity has been performed to analyze the individual performance of each proposed systems.

Download Full-text

A Hybrid Capsule Network for Pneumonia Detection Using Image Augmentation Based on Generative Adversarial Network

Traitement du signal ◽

10.18280/ts.380309 ◽

2021 ◽

Vol 38 (3) ◽

pp. 619-627

Author(s):

Kazim Firildak ◽

Muhammed Fatih Talu

Keyword(s):

Classification Accuracy ◽

Data Augmentation ◽

Model Parameters ◽

Classification Problems ◽

Generative Adversarial Network ◽

X Ray ◽

Discrimination Capability ◽

Adversarial Network ◽

Original Dataset ◽

Chest X Ray

Pneumonia, featured by inflammation of the air sacs in one or both lungs, is usually detected by examining chest X-ray images. This paper probes into the classification models that can distinguish between normal and pneumonia images. As is known, trained networks like AlexNet and GoogleNet are deep network architectures, which are widely adopted to solve many classification problems. They have been adapted to the target datasets, and employed to classify new data generated through transfer learning. However, the classical architectures are not accurate enough for the diagnosis of pneumonia. Therefore, this paper designs a capsule network with high discrimination capability, and trains the network on Kaggle’s online pneumonia dataset, which contains chest X-ray images of many adults and children. The original dataset consists of 1,583 normal images, and 4,273 pneumonia images. Then, two data augmentation approaches were applied to the dataset, and their effects on classification accuracy were compared in details. The model parameters were optimized through five different experiments. The results show that the highest classification accuracy (93.91% even on small images) was achieved by the capsule network, coupled with data augmentation by generative adversarial network (GAN), using optimized parameters. This network outperformed the classical strategies.

Download Full-text

AI Radar Sensor: Creating Radar Depth Sounder Images Based on Generative Adversarial Network

Sensors ◽

10.3390/s19245479 ◽

2019 ◽

Vol 19 (24) ◽

pp. 5479 ◽

Cited By ~ 1

Author(s):

Maryam Rahnemoonfar ◽

Jimmy Johnson ◽

John Paden

Keyword(s):

Data Augmentation ◽

Synthetic Data ◽

Detection Algorithm ◽

Contour Detection ◽

Training Data ◽

Generative Adversarial Network ◽

Radar Images ◽

Adversarial Network ◽

Radar Imagery

Significant resources have been spent in collecting and storing large and heterogeneous radar datasets during expensive Arctic and Antarctic fieldwork. The vast majority of data available is unlabeled, and the labeling process is both time-consuming and expensive. One possible alternative to the labeling process is the use of synthetically generated data with artificial intelligence. Instead of labeling real images, we can generate synthetic data based on arbitrary labels. In this way, training data can be quickly augmented with additional images. In this research, we evaluated the performance of synthetically generated radar images based on modified cycle-consistent adversarial networks. We conducted several experiments to test the quality of the generated radar imagery. We also tested the quality of a state-of-the-art contour detection algorithm on synthetic data and different combinations of real and synthetic data. Our experiments show that synthetic radar images generated by generative adversarial network (GAN) can be used in combination with real images for data augmentation and training of deep neural networks. However, the synthetic images generated by GANs cannot be used solely for training a neural network (training on synthetic and testing on real) as they cannot simulate all of the radar characteristics such as noise or Doppler effects. To the best of our knowledge, this is the first work in creating radar sounder imagery based on generative adversarial network.

Download Full-text

Visual and Quantitative Evaluation of Amyloid Brain PET Image Synthesis with Generative Adversarial Network

Applied Sciences ◽

10.3390/app10072628 ◽

2020 ◽

Vol 10 (7) ◽

pp. 2628 ◽

Cited By ~ 2

Author(s):

Hyeon Kang ◽

Jang-Sik Park ◽

Kook Cho ◽

Do-Young Kang

Keyword(s):

Quantitative Evaluation ◽

Data Augmentation ◽

Image Synthesis ◽

Turing Test ◽

Training Data ◽

Generative Adversarial Network ◽

Visual Evaluation ◽

Image Distribution ◽

Adversarial Network ◽

Quantitative Measurements

Conventional data augmentation (DA) techniques, which have been used to improve the performance of predictive models with a lack of balanced training data sets, entail an effort to define the proper repeating operation (e.g., rotation and mirroring) according to the target class distribution. Although DA using generative adversarial network (GAN) has the potential to overcome the disadvantages of conventional DA, there are not enough cases where this technique has been applied to medical images, and in particular, not enough cases where quantitative evaluation was used to determine whether the generated images had enough realism and diversity to be used for DA. In this study, we synthesized 18F-Florbetaben (FBB) images using CGAN. The generated images were evaluated using various measures, and we presented the state of the images and the similarity value of quantitative measurement that can be expected to successfully augment data from generated images for DA. The method includes (1) conditional WGAN-GP to learn the axial image distribution extracted from pre-processed 3D FBB images, (2) pre-trained DenseNet121 and model-agnostic metrics for visual and quantitative measurements of generated image distribution, and (3) a machine learning model for observing improvement in generalization performance by generated dataset. The Visual Turing test showed similarity in the descriptions of typical patterns of amyloid deposition for each of the generated images. However, differences in similarity and classification performance per axial level were observed, which did not agree with the visual evaluation. Experimental results demonstrated that quantitative measurements were able to detect the similarity between two distributions and observe mode collapse better than the Visual Turing test and t-SNE.

Download Full-text

Generating the Microstructure of Al-Si Cast Alloys Using Machine Learning

Korean Journal of Metals and Materials ◽

10.3365/kjmm.2021.59.11.838 ◽

2021 ◽

Vol 59 (11) ◽

pp. 838-847

Author(s):

In-Kyu Hwang ◽

Hyun-Ji Lee ◽

Sang-Jun Jeong ◽

In-Sung Cho ◽

Hee-Soo Kim

Keyword(s):

Machine Learning ◽

Data Augmentation ◽

Training Dataset ◽

Initial Training ◽

Microstructural Characteristics ◽

Generative Adversarial Network ◽

The Real ◽

Adversarial Network ◽

Cast Alloys ◽

The Right

In this study, we constructed a deep convolutional generative adversarial network (DCGAN) to generate the microstructural images that imitate the real microstructures of binary Al-Si cast alloys. We prepared four combinations of alloys, Al-6wt%Si, Al-9wt%Si, Al-12wt%Si and Al-15wt%Si for machine learning. DCGAN is composed of a generator and a discriminator. The discriminator has a typical convolutional neural network (CNN), and the generator has an inverse shaped CNN. The fake images generated using DCGAN were similar to real microstructural images. However, they showed some strange morphology, including dendrites without directionality, and deformed Si crystals. Verification with Inception V3 revealed that the fake images generated using DCGAN were well classified into the target categories. Even the visually imperfect images in the initial training iterations showed high similarity to the target. It seems that the imperfect images had enough microstructural characteristics to satisfy the classification, even though human cannot recognize the images. Cross validation was carried out using real, fake and other test images. When the training dataset had the fake images only, the real and test images showed high similarities to the target categories. When the training dataset contained both the real and fake images, the similarity at the target categories were high enough to meet the right answers. We concluded that the DCGAN developed for microstructural images in this study is highly useful for data augmentation for rare microstructures.

Download Full-text

Deep learning data augmentation for Raman spectroscopy cancer tissue classification

Scientific Reports ◽

10.1038/s41598-021-02687-0 ◽

2021 ◽

Vol 11 (1) ◽

Author(s):

Man Wu ◽

Shuwen Wang ◽

Shirui Pan ◽

Andrew C. Terentis ◽

John Strasswimmer ◽

...

Keyword(s):

Raman Spectroscopy ◽

Skin Cancer ◽

Data Augmentation ◽

Training Data ◽

Cancer Tissue ◽

Tissue Classification ◽

Generative Adversarial Network ◽

Tissue Samples ◽

Normal Tissues ◽

Adversarial Network

AbstractRecently, Raman Spectroscopy (RS) was demonstrated to be a non-destructive way of cancer diagnosis, due to the uniqueness of RS measurements in revealing molecular biochemical changes between cancerous vs. normal tissues and cells. In order to design computational approaches for cancer detection, the quality and quantity of tissue samples for RS are important for accurate prediction. In reality, however, obtaining skin cancer samples is difficult and expensive due to privacy and other constraints. With a small number of samples, the training of the classifier is difficult, and often results in overfitting. Therefore, it is important to have more samples to better train classifiers for accurate cancer tissue classification. To overcome these limitations, this paper presents a novel generative adversarial network based skin cancer tissue classification framework. Specifically, we design a data augmentation module that employs a Generative Adversarial Network (GAN) to generate synthetic RS data resembling the training data classes. The original tissue samples and the generated data are concatenated to train classification modules. Experiments on real-world RS data demonstrate that (1) data augmentation can help improve skin cancer tissue classification accuracy, and (2) generative adversarial network can be used to generate reliable synthetic Raman spectroscopic data.

Download Full-text

Improved Classification of White Blood Cells with the Generative Adversarial Network and Deep Convolutional Neural Network

Computational Intelligence and Neuroscience ◽

10.1155/2020/6490479 ◽

2020 ◽

Vol 2020 ◽

pp. 1-12

Author(s):

Khaled Almezhghwi ◽

Sertan Serte

Keyword(s):

Blood Cells ◽

Data Augmentation ◽

White Blood Cells ◽

Training Data ◽

Generative Adversarial Networks ◽

Generative Adversarial Network ◽

Human Errors ◽

Morphological Variations ◽

Adversarial Network

White blood cells (leukocytes) are a very important component of the blood that forms the immune system, which is responsible for fighting foreign elements. The five types of white blood cells include neutrophils, eosinophils, lymphocytes, monocytes, and basophils, where each type constitutes a different proportion and performs specific functions. Being able to classify and, therefore, count these different constituents is critical for assessing the health of patients and infection risks. Generally, laboratory experiments are used for determining the type of a white blood cell. The staining process and manual evaluation of acquired images under the microscope are tedious and subject to human errors. Moreover, a major challenge is the unavailability of training data that cover the morphological variations of white blood cells so that trained classifiers can generalize well. As such, this paper investigates image transformation operations and generative adversarial networks (GAN) for data augmentation and state-of-the-art deep neural networks (i.e., VGG-16, ResNet, and DenseNet) for the classification of white blood cells into the five types. Furthermore, we explore initializing the DNNs’ weights randomly or using weights pretrained on the CIFAR-100 dataset. In contrast to other works that require advanced image preprocessing and manual feature extraction before classification, our method works directly with the acquired images. The results of extensive experiments show that the proposed method can successfully classify white blood cells. The best DNN model, DenseNet-169, yields a validation accuracy of 98.8%. Particularly, we find that the proposed approach outperforms other methods that rely on sophisticated image processing and manual feature engineering.

Download Full-text

Generative Adversarial Network for Class-Conditional Data Augmentation

Applied Sciences ◽

10.3390/app10238415 ◽

2020 ◽

Vol 10 (23) ◽

pp. 8415

Author(s):

Jeongmin Lee ◽

Younkyoung Yoon ◽

Junseok Kwon

Keyword(s):

Classification Accuracy ◽

Data Augmentation ◽

Denoising Autoencoder ◽

Generative Adversarial Network ◽

Minority Class ◽

Adversarial Network ◽

Latent Space ◽

Benchmark Datasets ◽

Class Information ◽

Classification Tasks

We propose a novel generative adversarial network for class-conditional data augmentation (i.e., GANDA) to mitigate data imbalance problems in image classification tasks. The proposed GANDA generates minority class data by exploiting majority class information to enhance the classification accuracy of minority classes. For stable GAN training, we introduce a new denoising autoencoder initialization with explicit class conditioning in the latent space, which enables the generation of definite samples. The generated samples are visually realistic and have a high resolution. Experimental results demonstrate that the proposed GANDA can considerably improve classification accuracy, especially when datasets are highly imbalanced on standard benchmark datasets (i.e., MNIST and CelebA). Our generated samples can be easily used to train conventional classifiers to enhance their classification accuracy.

Download Full-text