scholarly journals Image based Plant leaf disease detection using Deep learning

2021 ◽  
Vol 3 (1) ◽  
pp. 53-65
Author(s):  
Poornam S ◽  
Francis Saviour Devaraj A
Keyword(s):  
Deep Learning ◽  
Activation Function ◽  
Training Phase ◽  
Image Preprocessing ◽  
Plant Leaf ◽  
Pest Infestation ◽  
Network Training ◽  
Detection And Identification ◽  
Final Layer ◽  
Fully Connected

Agriculture is important for India. Every year growing variety of crops is at loss due to inefficiency in shipping, cultivation, pest infestation in crop and storage of government-subsidized crops.  There is reduction in production of good crops in both quality and quantity due to Plants being affected by diseases. Hence it is important for early detection and identification of diseases in plants. The proposed methodology consists of collection of Plant leaf dataset, Image preprocessing, Image Augmentation and Neural network training. The dataset is collected from ImageNet for training phase. The CNN technique is used to differentiate the healthy leaf from disease affected leaf. In image preprocessing resizing the image is carried out to reduce the training phase time. Image augmentation is performed in training phase by applying various transformation function on Plant images. The Network is trained by Caffenet deep learning framework. CNN is trained with ReLu (Rectified Linear Unit). The convolution base of CNN generates features from image through the multiple convolution layers and pooling layers. The classifier part of CNN classifies the image based on the features extracted from the convolution base. The classification is performed through the fully connected layers. The performance is measured using 10-fold cross validation function. The final layer uses activation function like softmax to categorize the outputs.

scholarly journals Flatten-T Swish: a thresholded ReLU-Swish-like activation function for deep learning

2018 ◽  
Vol 4 (2) ◽  
pp. 76 ◽  
Author(s):  
Hock Hung Chieng ◽  
Noorhaniza Wahid ◽  
Ong Pauline ◽  
Sai Raj Kishore Perla
Keyword(s):  
Deep Learning ◽  
Learning Community ◽  
Classification Accuracy ◽  
Deep Neural Network ◽  
Threshold Value ◽  
Activation Function ◽  
Activation Functions ◽  
Rectified Linear Unit ◽  
Overall Performance ◽  
Fully Connected

Activation functions are essential for deep learning methods to learn and perform complex tasks such as image classification. Rectified Linear Unit (ReLU) has been widely used and become the default activation function across the deep learning community since 2012. Although ReLU has been popular, however, the hard zero property of the ReLU has heavily hindering the negative values from propagating through the network. Consequently, the deep neural network has not been benefited from the negative representations. In this work, an activation function called Flatten-T Swish (FTS) that leverage the benefit of the negative values is proposed. To verify its performance, this study evaluates FTS with ReLU and several recent activation functions. Each activation function is trained using MNIST dataset on five different deep fully connected neural networks (DFNNs) with depth vary from five to eight layers. For a fair evaluation, all DFNNs are using the same configuration settings. Based on the experimental results, FTS with a threshold value, T=-0.20 has the best overall performance. As compared with ReLU, FTS (T=-0.20) improves MNIST classification accuracy by 0.13%, 0.70%, 0.67%, 1.07% and 1.15% on wider 5 layers, slimmer 5 layers, 6 layers, 7 layers and 8 layers DFNNs respectively. Apart from this, the study also noticed that FTS converges twice as fast as ReLU. Although there are other existing activation functions are also evaluated, this study elects ReLU as the baseline activation function.

scholarly journals Automatic Segmentation of Choroid Layer Using Deep Learning on Spectral Domain Optical Coherence Tomography

2021 ◽  
Vol 11 (12) ◽  
pp. 5488
Author(s):  
Wei Ping Hsia ◽  
Siu Lun Tse ◽  
Chia Jen Chang ◽  
Yu Len Huang
Keyword(s):  
Optical Coherence Tomography ◽  
Deep Learning ◽  
Choroidal Thickness ◽  
Automatic Segmentation ◽  
Average Error ◽  
Good Prediction ◽  
Optical Coherence ◽  
Learning Method ◽  
Subfoveal Choroidal Thickness ◽  
Fully Connected

The purpose of this article is to evaluate the accuracy of the optical coherence tomography (OCT) measurement of choroidal thickness in healthy eyes using a deep-learning method with the Mask R-CNN model. Thirty EDI-OCT of thirty patients were enrolled. A mask region-based convolutional neural network (Mask R-CNN) model composed of deep residual network (ResNet) and feature pyramid networks (FPNs) with standard convolution and fully connected heads for mask and box prediction, respectively, was used to automatically depict the choroid layer. The average choroidal thickness and subfoveal choroidal thickness were measured. The results of this study showed that ResNet 50 layers deep (R50) model and ResNet 101 layers deep (R101). R101 U R50 (OR model) demonstrated the best accuracy with an average error of 4.85 pixels and 4.86 pixels, respectively. The R101 ∩ R50 (AND model) took the least time with an average execution time of 4.6 s. Mask-RCNN models showed a good prediction rate of choroidal layer with accuracy rates of 90% and 89.9% for average choroidal thickness and average subfoveal choroidal thickness, respectively. In conclusion, the deep-learning method using the Mask-RCNN model provides a faster and accurate measurement of choroidal thickness. Comparing with manual delineation, it provides better effectiveness, which is feasible for clinical application and larger scale of research on choroid.

Breast cancer diagnosis using multiple activation deep neural network

2021 ◽  
pp. 1063293X2110251
Author(s):  
K Vijayakumar ◽  
Vinod J Kadam ◽  
Sudhir Kumar Sharma
Keyword(s):  
Breast Cancer ◽  
Neural Network ◽  
Deep Neural Network ◽  
Breast Cancer Diagnosis ◽  
Activation Function ◽  
Linear Functions ◽  
Cancer Data ◽  
Final Layer ◽  
Improved Performance ◽  
Hidden Layer

Deep Neural Network (DNN) stands for multilayered Neural Network (NN) that is capable of progressively learn the more abstract and composite representations of the raw features of the input data received, with no need for any feature engineering. They are advanced NNs having repetitious hidden layers between the initial input and the final layer. The working principle of such a standard deep classifier is based on a hierarchy formed by the composition of linear functions and a defined nonlinear Activation Function (AF). It remains uncertain (not clear) how the DNN classifier can function so well. But it is clear from many studies that within DNN, the AF choice has a notable impact on the kinetics of training and the success of tasks. In the past few years, different AFs have been formulated. The choice of AF is still an area of active study. Hence, in this study, a novel deep Feed forward NN model with four AFs has been proposed for breast cancer classification: hidden layer 1: Swish, hidden layer, 2:-LeakyReLU, hidden layer 3: ReLU, and final output layer: naturally Sigmoidal. The purpose of the study is twofold. Firstly, this study is a step toward a more profound understanding of DNN with layer-wise different AFs. Secondly, research is also aimed to explore better DNN-based systems to build predictive models for breast cancer data with improved accuracy. Therefore, the benchmark UCI dataset WDBC was used for the validation of the framework and evaluated using a ten-fold CV method and various performance indicators. Multiple simulations and outcomes of the experimentations have shown that the proposed solution performs in a better way than the Sigmoid, ReLU, and LeakyReLU and Swish activation DNN in terms of different parameters. This analysis contributes to producing an expert and precise clinical dataset classification method for breast cancer. Furthermore, the model also achieved improved performance compared to many established state-of-the-art algorithms/models.

scholarly journals Audio-Based Drone Detection and Identification Using Deep Learning Techniques with Dataset Enhancement through Generative Adversarial Networks

Sensors ◽  
2021 ◽  
Vol 21 (15) ◽  
pp. 4953
Author(s):  
Sara Al-Emadi ◽  
Abdulla Al-Ali ◽  
Abdulaziz Al-Ali
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Recurrent Neural Network ◽  
Learning Algorithms ◽  
Generative Adversarial Networks ◽  
Generative Adversarial Network ◽  
Adversarial Networks ◽  
Detection And Identification ◽  
Learning Techniques ◽  
The Impact

Drones are becoming increasingly popular not only for recreational purposes but in day-to-day applications in engineering, medicine, logistics, security and others. In addition to their useful applications, an alarming concern in regard to the physical infrastructure security, safety and privacy has arisen due to the potential of their use in malicious activities. To address this problem, we propose a novel solution that automates the drone detection and identification processes using a drone’s acoustic features with different deep learning algorithms. However, the lack of acoustic drone datasets hinders the ability to implement an effective solution. In this paper, we aim to fill this gap by introducing a hybrid drone acoustic dataset composed of recorded drone audio clips and artificially generated drone audio samples using a state-of-the-art deep learning technique known as the Generative Adversarial Network. Furthermore, we examine the effectiveness of using drone audio with different deep learning algorithms, namely, the Convolutional Neural Network, the Recurrent Neural Network and the Convolutional Recurrent Neural Network in drone detection and identification. Moreover, we investigate the impact of our proposed hybrid dataset in drone detection. Our findings prove the advantage of using deep learning techniques for drone detection and identification while confirming our hypothesis on the benefits of using the Generative Adversarial Networks to generate real-like drone audio clips with an aim of enhancing the detection of new and unfamiliar drones.

scholarly journals Deep Learning for Automated Detection and Identification of Migrating American Eel Anguilla rostrata from Imaging Sonar Data

2021 ◽  
Vol 13 (14) ◽  
pp. 2671
Author(s):  
Xiaoqin Zang ◽  
Tianzhixi Yin ◽  
Zhangshuan Hou ◽  
Robert P. Mueller ◽  
Zhiqun Daniel Deng ◽  
...  
Keyword(s):  
Deep Learning ◽  
Fish Passage ◽  
Automated Detection ◽  
Anguilla Rostrata ◽  
Automated Method ◽  
Detection And Identification ◽  
And Migration ◽  
Imaging Sonar ◽  
American Eels ◽  
Deep Learning Model

Adult American eels (Anguilla rostrata) are vulnerable to hydropower turbine mortality during outmigration from growth habitat in inland waters to the ocean where they spawn. Imaging sonar is a reliable and proven technology for monitoring of fish passage and migration; however, there is no efficient automated method for eel detection. We designed a deep learning model for automated detection of adult American eels from sonar data. The method employs convolution neural network (CNN) to distinguish between 14 images of eels and non-eel objects. Prior to image classification with CNN, background subtraction and wavelet denoising were applied to enhance sonar images. The CNN model was first trained and tested on data obtained from a laboratory experiment, which yielded overall accuracies of >98% for image-based classification. Then, the model was trained and tested on field data that were obtained near the Iroquois Dam located on the St. Lawrence River; the accuracy achieved was commensurate with that of human experts.

scholarly journals Classification of Shoulder X-ray Images with Deep Learning Ensemble Models

2021 ◽  
Vol 11 (6) ◽  
pp. 2723
Author(s):  
Fatih Uysal ◽  
Fırat Hardalaç ◽  
Ozan Peker ◽  
Tolga Tolunay ◽  
Nil Tokgöz
Keyword(s):  
Deep Learning ◽  
Performance Test ◽  
The Body ◽  
Test Accuracy ◽  
Cohen’S Kappa ◽  
X Ray ◽  
Cohen's Kappa ◽  
Auc Value ◽  
Magnetic Resonance Imaging Mri ◽  
Fully Connected

Fractures occur in the shoulder area, which has a wider range of motion than other joints in the body, for various reasons. To diagnose these fractures, data gathered from X-radiation (X-ray), magnetic resonance imaging (MRI), or computed tomography (CT) are used. This study aims to help physicians by classifying shoulder images taken from X-ray devices as fracture/non-fracture with artificial intelligence. For this purpose, the performances of 26 deep learning-based pre-trained models in the detection of shoulder fractures were evaluated on the musculoskeletal radiographs (MURA) dataset, and two ensemble learning models (EL1 and EL2) were developed. The pre-trained models used are ResNet, ResNeXt, DenseNet, VGG, Inception, MobileNet, and their spinal fully connected (Spinal FC) versions. In the EL1 and EL2 models developed using pre-trained models with the best performance, test accuracy was 0.8455, 0.8472, Cohen’s kappa was 0.6907, 0.6942 and the area that was related with fracture class under the receiver operating characteristic (ROC) curve (AUC) was 0.8862, 0.8695. As a result of 28 different classifications in total, the highest test accuracy and Cohen’s kappa values were obtained in the EL2 model, and the highest AUC value was obtained in the EL1 model.

Inverse design of layered periodic wave barriers based on deep learning

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.

Application of Deep Learning Technique to Predict Downhole Pressure Differential in Eccentric Annulus of Ultra-Deep Well

2021 ◽  
Author(s):  
Shwetank Krishna ◽  
Syahrir Ridha ◽  
Suhaib Umer Ilyas ◽  
Scott Campbell ◽  
Uday Bhan ◽  
...  
Keyword(s):  
Deep Learning ◽  
Pressure Gradient ◽  
Field Data ◽  
Activation Function ◽  
Pressure Differential ◽  
Optimal Arrangement ◽  
Eccentric Annulus ◽  
Learning Framework ◽  
Deep Well ◽  
Deep Wells

Abstract Accurate prediction of downhole pressure differential (surge/swab pressure gradient) in the eccentric annulus of ultra-deep wells during tripping operation is a necessity to optimize well geometry, reduction of drilling anomalies, and prevention of hazardous drilling accidents. Therefore, a new predictive model is developed to forecast surge/swab pressure gradient by using feed-forward and backpropagation deep neural networks (FFBP-DNN). A theoretical-based model is developed that follows the physical and mechanical aspects of surge/swab pressure generation in eccentric annulus during tripping operation. The data generated from this model, field data, and experimental data are used to train and test the FFBP-DNN networks. The network is developed used Keras’s deep learning framework. After testing the models, the most optimal arrangement of FFBP-DNN is the ReLU algorithm as an activation function, 4-hidden layers, the learning rate of 0.003, and 2300 of training numbers. The optimum FFBP-DNN model is validated by comparing it with field data (Wells K 470 and K 480, North Sea). It shows an excellent argument between predicted data and field data with an error range of ±7.68 %.

Sign in / Sign up

Export Citation Format

Share Document