VGG16 Network Research Articles

A Review on Potato Disease Detection and Classification by exploiting Deep Learning Techniques

The edible potato comes in fifth for human consumption and fourth among main food crops. Since it is a crop that is vegetative grown, many pests and disease can be passed along from one generation to the next. Crop diseases, which have a negative impact on food security as well as economic losses, have a significant impact on the production and quality of yields from potato crops. Thus, the application of unique and precise deep learning-based algorithms for disease detection and classification is highly required. Identifying weaknesses in agricultural products, particularly potatoes requires the use of machine vision and image processing techniques. Deep learning and image processing have been used in agriculture to classify and number of disease and pests affecting potatoes has grown, and study in this area is still continuing. The use of artificial intelligence and image processing in agriculture for the classification and identification of potato pests and disease has grown, and work in this area is still ongoing. Different deep learning techniques, such as VGG19, VGG16, Google Net, Alex Net, and convolution neural network methods, can be used to address the disease problem in potatoes. These methods also examined multiple classes of potato diseases as: Healthy, Black Leg, Black Scurf, Pink Rot, Common Scab, etc. Food safety could be seriously threatened by the spread of potato disease. In this article, deep learning techniques for early detection of potato disease are discussed.

A Comparative Study of Deep Learning Parameters for Arcus Senilis Classification

Deep learning technique has recently yielded positive results that have increased productivity system for artificial intelligent task, especially the digital image processing and advance machine vision. The popularity of deep learning technique has a major impact on solving complex problems in many fields, particularly the medical science owing to its applications in medical imaging, disease diagnosis, and much more. However, successful application of deep learning depends upon the appropriate setting of the parameters to achieve better result. Therefore, this paper presents a comparative analysis of different base learning rate and batch size configurations for arcus senilis (AS) classification using deep learning techniques. In this analytical study, a dataset of 402 eye images comprising 158 normal and 244 abnormal eye images was employed. Two well-known ResNet-50, VGG-19 and pre-trained convolutional neural network (CNN) models have been trained and validated with 10-fold cross validation using the proposed dataset. Furthermore, base learning rate and batch size were adjusted accordingly to determine the optimal convergence of each model by observing the validation accuracy and error. Experimental result shows that the best combined system has achieved an overall accuracy of 99.78% with a base learning rate of 0.0001 and a batch size of 20 on CNN pre-trained model validation set. Moreover, CNN produces the best result on F1-score and standard deviation of 99.77% and 0.464 respectively. Thus, it can be concluded that CNN requires a considerably smaller number of parameters and reasonable computing time to achieve state-of-the-art performances. This study shows that CNN has the tendency to consistently improve inaccuracy with growing number of epochs, with no signs of overfitting and performance.

Fractal Texture Enhancement of Simulated Infrared Images Using a CNN-Based Neural Style Transfer Algorithm with a Histogram Matching Technique.

Here, we propose a CNN-based infrared image enhancement method to transform pseudo-realistic regions of simulation-based infrared images into real infrared texture. The proposed algorithm consists of the following three steps. First, target infrared features based on a real infrared image are extracted through pretrained VGG-19 networks. Next, by implementing a neural style-transfer algorithm to a simulated infrared image, fractal nature features from the real infrared image are progressively applied to the image. Therefore, the fractal characteristics of the simulated image are improved. Finally, based on the results of fractal analysis, peak signal-to-noise (PSNR), structural similarity index measure (SSIM), and natural image quality evaluator (NIQE) texture evaluations are performed to know how the simulated infrared image is properly transformed as it contains the real infrared fractal features. We verified the proposed methodology using a simulation with three different simulation conditions with a real mid-wave infrared (MWIR) image. As a result, the enhanced simulated infrared images based on the proposed algorithm have better NIQE and SSIM score values in both brightness and fractal characteristics, indicating the closest similarity to the given actual infrared image. The proposed image fractal feature analysis technique can be widely used not only for the simulated infrared images but also for general synthetic images.

A Graph Neural Network based Fake Currency Recognition Approach

In general, fake currencies are being circulated around, for collapsing the country’s economy. It is achieved by spreading the money to foreign tourist people and vision-impaired people. This proposed work aims to assist such people on detecting the genuineness of the currencies given to them. The work is designed using a customized Graph Neural Network (GNN) -based approach for estimations, and its performance is compared to the traditional Convolution Neural Network, K-Nearest Neighbor, and VGG16 network. This experimental study has been performed utilizing an openly available dataset called Indian currency notes from Kaggle website. The outcome of the study indicates that the suggested GNN model performs better than the existing approaches.

Crowd Counting Based on Context-Aware and Multi Scale Feature Fusion

Crowd counting plays an important role in public security. Estimating the number of people in an image with congested crowd accurately is a challenging task. The crowd counting method based on fully convolutional network can perform well in crowd image with complex scene. In this paper, to address the counting problems of occlusion，background clutter and perspective effect, we proposed a simple but effective method called Context-aware Multi scale Fusion Network(CMF Net).The CMF Net applied VGG network as backbone to extract coarse features. Then, three context-aware multi-scale fusion modules (CMFM) are adopted. Each CMFM consist of multi-scale feature extraction module (MEM) and context-aware feature extraction module (CEM). In addition, we propose adaptively dense connection to promoted information transmission in the counting network. Experiments on four datasets demonstrate that our network achieves competitive and effective results.

Chest X-Ray Images to Differentiate COVID-19 from Pneumonia with Artificial Intelligence Techniques.

This paper presents an automated and noninvasive technique to discriminate COVID-19 patients from pneumonia patients using chest X-ray images and artificial intelligence. The reverse transcription-polymerase chain reaction (RT-PCR) test is commonly administered to detect COVID-19. However, the RT-PCR test necessitates person-to-person contact to administer, requires variable time to produce results, and is expensive. Moreover, this test is still unreachable to the significant global population. The chest X-ray images can play an important role here as the X-ray machines are commonly available at any healthcare facility. However, the chest X-ray images of COVID-19 and viral pneumonia patients are very similar and often lead to misdiagnosis subjectively. This investigation has employed two algorithms to solve this problem objectively. One algorithm uses lower-dimension encoded features extracted from the X-ray images and applies them to the machine learning algorithms for final classification. The other algorithm relies on the inbuilt feature extractor network to extract features from the X-ray images and classifies them with a pretrained deep neural network VGG16. The simulation results show that the proposed two algorithms can extricate COVID-19 patients from pneumonia with the best accuracy of 100% and 98.1%, employing VGG16 and the machine learning algorithm, respectively. The performances of these two algorithms have also been collated with those of other existing state-of-the-art methods.

VANet: a medical image fusion model based on attention mechanism to assist disease diagnosis

BackgroundToday’s biomedical imaging technology has been able to present the morphological structure or functional metabolic information of organisms at different scale levels, such as organ, tissue, cell, molecule and gene. However, different imaging modes have different application scope, advantages and disadvantages. In order to improve the role of medical image in disease diagnosis, the fusion of biomedical image information at different imaging modes and scales has become an important research direction in medical image. Traditional medical image fusion methods are all designed to measure the activity level and fusion rules. They are lack of mining the context features of different modes of image, which leads to the obstruction of improving the quality of fused images.MethodIn this paper, an attention-multiscale network medical image fusion model based on contextual features is proposed. The model selects five backbone modules in the VGG-16 network to build encoders to obtain the contextual features of medical images. It builds the attention mechanism branch to complete the fusion of global contextual features and designs the residual multiscale detail processing branch to complete the fusion of local contextual features. Finally, it completes the cascade reconstruction of features by the decoder to obtain the fused image.ResultsTen sets of images related to five diseases are selected from the AANLIB database to validate the VANet model. Structural images are derived from MR images with high resolution and functional images are derived from SPECT and PET images that are good at describing organ blood flow levels and tissue metabolism. Fusion experiments are performed on twelve fusion algorithms including the VANet model. The model selects eight metrics from different aspects to build a fusion quality evaluation system to complete the performance evaluation of the fused images. Friedman’s test and the post-hoc Nemenyi test are introduced to conduct professional statistical tests to demonstrate the superiority of VANet model.ConclusionsThe VANet model completely captures and fuses the texture details and color information of the source images. From the fusion results, the metabolism and structural information of the model are well expressed and there is no interference of color information on the structure and texture; in terms of the objective evaluation system, the metric value of the VANet model is generally higher than that of other methods.; in terms of efficiency, the time consumption of the model is acceptable; in terms of scalability, the model is not affected by the input order of source images and can be extended to tri-modal fusion.

Using cascade CNN-LSTM-FCNs to identify AI-altered video based on eye state sequence.

Deep learning is notably successful in data analysis, computer vision, and human control. Nevertheless, this approach has inevitably allowed the development of DeepFake video sequences and images that could be altered so that the changes are not easily or explicitly detectable. Such alterations have been recently used to spread false news or disinformation. This study aims to identify Deepfaked videos and images and alert viewers to the possible falsity of the information. The current work presented a novel means of revealing fake face videos by cascading the convolution network with recurrent neural networks and fully connected network (FCN) models. The system detection approach utilizes the eye-blinking state in temporal video frames. Notwithstanding, it is deemed challenging to precisely depict (i) artificiality in fake videos and (ii) spatial information within the individual frame through this physiological signal. Spatial features were extracted using the VGG16 network and trained with the ImageNet dataset. The temporal features were then extracted in every 20 sequences through the LSTM network. On another note, the pre-processed eye-blinking state served as a probability to generate a novel BPD dataset. This newly-acquired dataset was fed to three models for training purposes with each entailing four, three, and six hidden layers, respectively. Every model constitutes a unique architecture and specific dropout value. Resultantly, the model optimally and accurately identified tampered videos within the dataset. The study model was assessed using the current BPD dataset based on one of the most complex datasets (FaceForensic++) with 90.8% accuracy. Such precision was successfully maintained in datasets that were not used in the training process. The training process was also accelerated by lowering the computation prerequisites.

Robust zero-watermarking scheme based on a depthwise overparameterized VGG network in healthcare information security

As healthcare information technology has rapidly evolved, securely storing and transmitting medical data online and successfully protecting patient privacy are currently the research focus in the healthcare information field. To better protect the security of medical data, this paper introduces deep neural network and convolutional block attention module (CBAM) into the study of watermarking techniques and proposes a medical image zero-watermarking scheme based on depthwise overparameterized VGG (DO-VGG). First, we extract the high-dimensional abstract feature information of medical images using the pretrained DO-VGG model. Then, the construction of the zero-watermarking scheme utilizes the mean-perceptual hashing algorithm, which can efficiently resist both common and geometric attacks. Meanwhile, using the improved logistic mapping to encrypt the watermarking image effectively improves the security of the scheme. Experimental results indicate that all NC values of the proposed scheme are maintained above 0.8 under various degrees of attacks, which has good robustness and invisibility. The proposed scheme can satisfy the special requirements of medical image integrity and effectively protect the private information of medical images.

Identification of Corrosion on the Inner Walls of Water Pipes Using a VGG Model Incorporating Attentional Mechanisms

To accurately classify and identify the different corrosion patterns on the inner walls of water-supply pipes with different morphologies and complex and variable backgrounds, an improved VGG16 convolutional neural network classification model is proposed. Firstly, the S.E attention mechanism is added to the traditional VGG network model, which can be used to distinguish the importance of each channel of the feature map and re-weight the feature map through the globally calculated channel attention. Secondly, the joint-loss-function method is used to improve the loss function and further improve the classification performance of the model. The experimental results show that the proposed model can effectively identify different pipe-corrosion patterns with an accuracy of 95.266%, higher than the unimproved VGG and AlexNet models.

Detection of trachoma using machine learning approaches.

Though significant progress in disease elimination has been made over the past decades, trachoma is the leading infectious cause of blindness globally. Further efforts in trachoma elimination are paradoxically being limited by the relative rarity of the disease, which makes clinical training for monitoring surveys difficult. In this work, we evaluate the plausibility of an Artificial Intelligence model to augment or replace human image graders in the evaluation/diagnosis of trachomatous inflammation-follicular (TF). We utilized a dataset consisting of 2300 images with a 5% positivity rate for TF. We developed classifiers by implementing two state-of-the-art Convolutional Neural Network architectures, ResNet101 and VGG16, and applying a suite of data augmentation/oversampling techniques to the positive images. We then augmented our data set with additional images from independent research groups and evaluated performance. Models performed well in minimizing the number of false negatives, given the constraint of the low numbers of images in which TF was present. The best performing models achieved a sensitivity of 95% and positive predictive value of 50-70% while reducing the number images requiring skilled grading by 66-75%. Basic oversampling and data augmentation techniques were most successful at improving model performance, while techniques that are grounded in clinical experience, such as highlighting follicles, were less successful. The developed models perform well and significantly reduce the burden on graders by minimizing the number of false negative identifications. Further improvements in model skill will benefit from data sets with more TF as well as a range in image quality and image capture techniques used. While these models approach/meet the community-accepted standard for skilled field graders (i.e., Cohen's Kappa >0.7), they are insufficient to be deployed independently/clinically at this time; rather, they can be utilized to significantly reduce the burden on skilled image graders.

Transfer Learning in Keratoconus Classification

Early detection of keratoconus will provide more treatment choices, avoid heavy treatments, and help stop the rapid progression of the disease. Unlike traditional methods of keratoconus classification, this study presents a machine learning-based keratoconus classification approach, using transfer learning, applied on corneal topographic images. Classification is performed considering the three corneal classes already cited : normal, suspicious and keratoconus. Keratoconus classification is carried out using six pretrained convolutional neural networks (CNN) VGG16, InceptionV3, MobileNet, DenseNet201, Xception and EfficientNetB0. Each of these different classifiers is trained individually on five different datasets, generated from an original dataset of 2924 corneal topographic images. Original corneal topographic images have been subjected to a special preprocessing before their use by different models in the learning phase. Images of corneal maps are separated in five different datasets while removing noise and textual annotation from images. Most of models used in the classification allow good discrimination between normal cornea, suspicious and keratoconus one. Obtained results reached classification accuracy of 99.31% and 98.51% by DenseNet201 and VGG16 respectively. Obtained results indicate that transfer learning technique could well improve performance of keratoconus classification systems.

Deep Learning Activation Layer-Based Wall Quality Recognition Using Conv2D ResNet Exponential Transfer Learning Model

Crack detection is essential for observing structural health and guaranteeing structural safety. The manual crack and other damage detection process is time-consuming and subject to surveyors’ biased judgments. The proposed Conv2D ResNet Exponential model for wall quality detection was trained with 5000 wall images, including various imperfections such as cracks, holes, efflorescence, damp patches, and spalls. The model was trained with initial weights to form the trained layers of the base model and was integrated with Xception, VGG19, DenseNet, and ResNet convolutional neural network (CNN) models to retrieve the general high-level features. A transfer deep-learning-based approach was implemented to create a custom layer of CNN models. The base model was combined with custom layers to estimate wall quality. Xception, VGG19, DenseNet, and ResNet models were fitted with different activation layers such as softplus, softsign, tanh, selu, elu, and exponential, along with transfer learning. The performance of Conv2D was evaluated using model loss, precision, accuracy, recall, and F-score measures. The model was validated by comparing the performances of Xception, VGG19, DenseNet, ResNet, and Conv2D ResNet Exponential. The experimental results show that the Conv2D ResNet model with an exponential activation layer outperforms it with an F-score value of 0.9978 and can potentially be a viable substitute for classifying various wall defects.

OpenPose-Based Yoga Pose Classification Using Convolutional Neural Network

Human action recognition has various implementation, such surveillance system, elders care and construction alert, which arouse lots of interest of research in classification of still image. This paper mainly focusses on detecting the pose of Yoga. Comparing with traditional method using convolutional neural network, which is using original image as input to train the VGG network, extracting the skeleton images and feed them into Mobile net can impressively increase the accuracy. Dataset is collected from Kaggle website which contains five categories of labeled Yoga image. Openpose is an open-source API that can extract the human skeleton structure form the Yoga image based on the pose. With these skeleton image as input, the convolutional neural network will perceive everything important such as pose and angle of joints, rather than irrelevant features such as color and environment. Using Mobile net instead of common method to do classification with VGG, calculation time has been remarkably reduced and size of model is lighter which is able to be apply on single chip device. The result of model is impressive, showing high accuracy in both training data set and testing data set, which means no overfitting problem occurred in the experiment. Model size and demanding of hardware are also acceptable for a common personal computer.

From Bibliometrics to Altmetrics: Examining the Relationship Between Citation Count and Altmetric Score in Publications on Artificial Intelligence

In this research work, we studied two main research questions. First, we explored how artificial intelligence (AI) publications are seen on various altmetric platforms. Second, by looking at publications in the field of AI, we examined whether bibliometrics and altmetrics are correlated. For each of the 8000 AI publications on the Web of Science (WoS) database with altmetric scores, we recorded the citation counts and altmetric scores. Pearson’s correlation for the two lists of variables was found to be 0.2. This result is near to zero, indicating that there is weak correlation between citations and altmetric attention score. Further, in the two top lists of publications according to citation count and altmetric score, respectively, we found only one paper in common. For the time being, altmetric scores are weakly correlated with citation in the AI publications. In terms of research impact, altmetrics should be regarded as complementary to traditional bibliometrics rather than as a replacement measure. Beside these results, this article gives an overview of this area of research (scientometrics and informetrics). All related terminologies are defined, and for the purpose of this study we focus on bibliometrics and altmetrics.

Optimal deep transfer learning based ethnicity recognition on face images

In recent times, deep learning driven face image analysis has gained significant interest among several application areas like surveillance, security, biometrics, etc. The facial analysis intends to compute facial soft biometrics like ethnicity, expression, identification, age, gender, and so on. Among several biometrics, ethnicity recognition remains a hot research area. Recent advancements in computer vision (CV) and artificial intelligence (AI) models form the basis of an effective design of ethnicity recognition models. With this motivation, this paper introduces a novel Harris Hawks optimization with deep transfer learning based fusion model for face ethnicity recognition (HHODTLF-FER) model. The proposed HHODTLF-FER model is to determine the different kinds of ethnicity for applied facial images. A fusion of three pre-trained DL models, namely VGG16, Inception v3, and capsule networks (CapsNet) models, are employed. In addition, bidirectional long short term memory (BiLSTM) model is applied for ethnicity recognition and Classification. Finally, HHO algorithm is utilized to fine tune the hyperparameters contained in the BiLSTM model, showing the novelty of the work. In order to ensure the improved recognition performance of the HHODTLF-FER model, a wide ranging experimental analysis is performed using benchmark databases. The comprehensive comparative study highlighted the promising performance of the HHODTLF-FER model over the other approaches. • To develop an ethnicity recognition on face images using harris hawks optimization basd deep transfer learning. • To improve the classification performance, bidirectional long short term memory (BiLSTM) model is applied for ethnicity recognition and Classification. • To improve the classification performance, HHO algorithm is utilized to fine tune the hyperparameters contained in the BiLSTM model, showing the novelty of the work.

Smoke target detection based on dynamic multi-frame decimation with cavity convolution

Compared with traditional smoke alarms, smoke detection based on video images has the advantages of low cost, fast response time, wide applicability, etc. Smoke detection algorithms are becoming an effective and widely used tool for early warning of fires by recognizing smoke graphic features. In this paper, we use the VGG-16 neural network as the benchmark network and improve it by using null convolution and combining it with the dynamic multi-frame subtraction method for smoke detection. The experimental results show that the proposed method is useful for early warning of smoke in the early stage of a fire before it becomes intense.

Employing Neural Style Transfer for Generating Deep Dream Images

In recent years, deep dream and neural style transfer emerged as hot topics in deep learning. Hence, mixing those two techniques support the art and enhance the images that simulate hallucinations among psychiatric patients and drug addicts. In this study, our model combines deep dream and neural style transfer (NST) to produce a new image that combines the two technologies. VGG-19 and Inception v3 pre-trained networks are used for NST and deep dream, respectively. Gram matrix is a vital process for style transfer. The loss is minimized in style transfer while maximized in a deep dream using gradient descent for the first case and gradient ascent for the second. We found that different image produces different loss values depending on the degree of clarity of that images. Distorted images have higher loss values in NST and lower loss values with deep dreams. The opposite happened for the clear images that did not contain mixed lines, circles, colors, or other shapes.

A Non-Idealities Aware Software–Hardware Co-Design Framework for Edge-AI Deep Neural Network Implemented on Memristive Crossbar

In this work, a non-idealities aware software-hardware co-design framework for deep neural network (DNN) implemented on memristive crossbar is presented. The device level non-ideal factors such as device conductance variation, nonuniform quantization levels, device-to-device variation and programming failure probability are included in the model. At array level, the impact of line resistance and sneak path are considered using a new fast and accurate line resistance estimation model. The non-linearity and offset of the peripheral circuits are also considered. By incorporating these factors into a unified DNN training process, the neural network performance can be evaluated holistically. Furthermore, the proposed training process can effectively mitigate the impact of these non-idealities and reduce the inference accuracy degradations. Implemented in a hybrid fashion of Python and PyTorch, the proposed framework is evaluated with a simplified 5-layer VGG network implemented on measured <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$128\times 128$ </tex-math></inline-formula> RRAM array with 3-level weight resolution. For CIFAR-10 tasks, 83% inference accuracy is achieved with less than 3% accuracy drop compared to the ideal model.

CLASSIFICATION OF CYTOLOGICAL IMAGES OF WHITE BLOOD CELLS USING DEEP LEARNING

Leukocytes, also known as white blood cells, are an important part of the immune system and represent a group of cells that protect the body from infections. The classification of leukocytes is widely used to diagnose various diseases, such as AIDS, leukemia, myeloma, anemia and others. However, traditional methods of classification of leukocytes require a lot of time, and their results are prone to errors. The article implements a computer approach to classification and detection of white blood cells on images of blood cells using deep learning methods based on the application of the method of transferring deep learning models and finetuning them. Deep learning has become one of the most popular areas of artificial intelligence. There are many neural network architectures used in deep learning in solving computer vision problems. The purpose of this study is to develop a computer system for efficient automatic detection and classification of leukocytes of four types of blood cells: eosinophils, lymphocytes, monocytes, neutrophils. This article used popular models of accurate neural networks VGG16, ResNet50, DensNet201, MobileNetV3, InceptionResnetV2, pretrained on ImageNet and finetuned on WBC images dataset.The paper also proposes a custom model of a deep neural network using separable convolutional blocks SeparableConv2D in its architecture. The model is optimized using methods of preprocessing normalization and data augmentation. The is model gave classification metrics of accuracy = 98.84 %, precision = 99.56 %, recall = 98.89 % and f1-score = 99.22 %. The developed model allows, in most cases, to determine with high speed whether a leukocyte belongs to one of the four classes in the image, which indicates the possibility of using the system as an auxiliary tool for hematological blood analysis.