VGG16 Architecture Research Articles

Utilising VGG-16 of Convolutional Neural Network for Medical Image Classification

Medical image classification, which involves accurately classifying anomalies or abnormalities within images, is an important area of attention in healthcare domain. It requires a fast and exact classification to ensure appropriate and timely treatment to the patients. This paper introduces a model based on Convolutional Neural Network (CNN) that utilises the VGG16 architecture for medical image classification, specifically in brain tumour and Alzheimer dataset. The VGG16 architecture, is known for its remarkable ability to extract important features, that is crucial in medical image classification. To enhance the precision of diagnosis, a detailed experimental setup is conducted, which includes the careful selection and organisation of a collection of medical images that cover different illnesses and anomalies to the dataset. The architecture of the model is then adjusted to achieve optimal performance in for image classification. The results show the model's efficiency in identifying anomalies in medical images especially for brain tumour dataset. The sensitivity, specificity, and F1-score evaluation metrics are presented, emphasising the model's ability to accurately differentiate between various medical image diseases.

Grid-Based Structural and Dimensional Skin Cancer Classification with Self-Featured Optimized Explainable Deep Convolutional Neural Networks.

Skin cancer is a severe and potentially lethal disease, and early detection is critical for successful treatment. Traditional procedures for diagnosing skin cancer are expensive, time-intensive, and necessitate the expertise of a medical practitioner. In recent years, many researchers have developed artificial intelligence (AI) tools, including shallow and deep machine learning-based approaches, to diagnose skin cancer. However, AI-based skin cancer diagnosis faces challenges in complexity, low reproducibility, and explainability. To address these problems, we propose a novel Grid-Based Structural and Dimensional Explainable Deep Convolutional Neural Network for accurate and interpretable skin cancer classification. This model employs adaptive thresholding for extracting the region of interest (ROI), using its dynamic capabilities to enhance the accuracy of identifying cancerous regions. The VGG-16 architecture extracts the hierarchical characteristics of skin lesion images, leveraging its recognized capabilities for deep feature extraction. Our proposed model leverages a grid structure to capture spatial relationships within lesions, while the dimensional features extract relevant information from various image channels. An Adaptive Intelligent Coney Optimization (AICO) algorithm is employed for self-feature selected optimization and fine-tuning the hyperparameters, which dynamically adapts the model architecture to optimize feature extraction and classification. The model was trained and tested using the ISIC dataset of 10,015 dermascope images and the MNIST dataset of 2357 images of malignant and benign oncological diseases. The experimental results demonstrated that the model achieved accuracy and CSI values of 0.96 and 0.97 for TP 80 using the ISIC dataset, which is 17.70% and 16.49% more than lightweight CNN, 20.83% and 19.59% more than DenseNet, 18.75% and 17.53% more than CNN, 6.25% and 6.18% more than Efficient Net-B0, 5.21% and 5.15% over ECNN, 2.08% and 2.06% over COA-CAN, and 5.21% and 5.15% more than ARO-ECNN. Additionally, the AICO self-feature selected ECNN model exhibited minimal FPR and FNR of 0.03 and 0.02, respectively. The model attained a loss of 0.09 for ISIC and 0.18 for the MNIST dataset, indicating that the model proposed in this research outperforms existing techniques. The proposed model improves accuracy, interpretability, and robustness for skin cancer classification, ultimately aiding clinicians in early diagnosis and treatment.

Improved prostate cancer diagnosis using a modified ResNet50-based deep learning architecture

Prostate cancer, the most common cancer in men, is influenced by age, family history, genetics, and lifestyle factors. Early detection of prostate cancer using screening methods improves outcomes, but the balance between overdiagnosis and early detection remains debated. Using Deep Learning (DL) algorithms for prostate cancer detection offers a promising solution for accurate and efficient diagnosis, particularly in cases where prostate imaging is challenging. In this paper, we propose a Prostate Cancer Detection Model (PCDM) model for the automatic diagnosis of prostate cancer. It proves its clinical applicability to aid in the early detection and management of prostate cancer in real-world healthcare environments. The PCDM model is a modified ResNet50-based architecture that integrates faster R-CNN and dual optimizers to improve the performance of the detection process. The model is trained on a large dataset of annotated medical images, and the experimental results show that the proposed model outperforms both ResNet50 and VGG19 architectures. Specifically, the proposed model achieves high sensitivity, specificity, precision, and accuracy rates of 97.40%, 97.09%, 97.56%, and 95.24%, respectively.

Real-time Facial Expression Recognition to Enhance Emotional Intelligence in Autism

The ability to interpret emotions from facial expressions is essential in social interaction. However, these abilities are complex for individuals with autism spectrum disorder (ASD). This study aims to develop a real-time human facial expression recognition system on a mobile phone application to help individuals with ASD. The expression recognition model was developed using the FER-2013 dataset with the VGG-16 architecture. The experimental results show that the model obtained from the VGG-16 architecture can produce the best accuracy and average F1 scores with 0.91 and 0.91, respectively. We also experiment with three individuals with ASD, resulting in increasing interest in learning the expression.

Catalyzing medical imaging : Exploring the potentials of deep transfer learning

Medical imaging plays a crucial role in modern healthcare by aiding in accurate diagnosis, treatment planning, and disease monitoring. The advancement of deep learning has revolutionized the field of medical image processing, leading to extraordinary breakthroughs in various domains. Among the many subdomains within deep learning, deep transfer learning has emerged as a powerful technique for transferring knowledge between domains, greatly enhancing the accuracy and efficiency of medical image analysis. This paper delves into the evolution of deep transfer learning in the realm of medical imaging. This study delves into leveraging pre-trained models and employing transfer learning techniques - specifically, freezing CNN layers and fine-tuning - on the VGG-16 architecture. The results were impressive, as freezing CNN layers and fine-tuning the VGG-16 model with the MIAS dataset yielded astounding accuracies of 98.33% and 99.89% respectively.

Diagnosis of bacterial leaf blight, leaf smut, and brown spot in rice leafs using VGG16

In the context of agricultural sustainability and food security, timely and precise identification of plant diseases holds paramount significance. This research proposes an innovative approach to diagnose bacterial leaf blight (BLB), rice blast (RB), and brown spot (BS) in rice leaves utilizing the VGG16 convolutional neural network (CNN). By leveraging deep learning capabilities, the proposed model analyzes high-resolution images of rice leaves to classify and distinguish between the different diseases. The VGG16 architecture has shown remarkable performance in image recognition tasks, and this study aims to harness its potential for precision agriculture. Rice (Oryza sativa) is a staple crop that plays a crucial role in global food security. Timely and precise diagnosis of these diseases is vital for effective disease management and crop protection. Our methodology involves the collection and curation of a diverse dataset of rice leaf images. The VGG16 model, a proven deep convolutional neural network (CNN) architecture, is employed for extraction of features and classification. The results of our study demonstrate the efficacy of the VGG16 model in distinguishing between different rice leaf diseases with remarkable precision. This high level of accuracy holds great promise for the field of agriculture, offering a non-invasive and efficient means of disease detection. Early disease identification can facilitate timely intervention, reducing crop losses and mitigating the economic impact on farmers. The classification accuracy of the proposed method is 97.77% using the publicly available dataset.

Content Based Image Retrieval System Using CNN based Deep Learning Models

Digital information search archives on the internet and other platforms have been an essential part of our daily lives. With the use of huge amounts of digital images, it is paramount to develop efficient solutions to retrieve images seamlessly. Due to inadequate textual description, the task of automated image retrieval is quite complicated. Therefore, retrieval of images by analysing their visual content is an exciting task and research-worthy. Text based Image Search (TIBR) is time consuming as initially, every image has to be linked with a keyword or text to be extracted on query search. In the proposed system, an efficient algorithm for Content Based Image Retrieval (CBIR) using pre-trained CNN-based Deep Learning models to extract deep features of an image has been developed, which significantly increases the performance of the image retrieval process. Convolutional Neural Networks (CNN) has evolved as an efficient deep learning solution for the CBIR systems that work in image recognition applications. The results of retrieval are evaluated in terms of precision performance parameters. Emphasis has been given to the comparison of the efficiency of the system using two different pre-trained CNN-based Deep Learning models based on VGG16 and ResNet-50 architecture. Transfer learning enables us to develop robust and efficient working systems using pre-trained models. In the proposed system, pre-trained models have been used to derive features using a deep learning CNN network trained for a large image classification dataset ImageNet. This approach outperforms many contemporary CBIR systems. As most CBIR systems depend on query images and as most users of such systems are non-professional, we have developed a user interface for the built prototype systems. The efficiency of the CBIR systems with pre-trained models VGG16, and ResNet-50 are compared for better image retrieval and reliability. The proposed system has addressed a few existing problems in CBIR system design and developed a suitable solution with less semantic gap and better efficiency in image retrieval.

Transfer Learning-Based Bhutanese Currency Recognition

Currency recognition is an important field in the area of pattern recognition. Recent improvements in the field of deep learning have increased its ability to recognize complex features from images. The goal of this research was to create a first-ever Bhutanese dataset and train using transfer learning techniques. This paper proposes three different pre-trained models to train the Bhutanese currency datasets. A parameter fine-tuning was applied to get better accuracy from the custom dataset. The currency images were gathered using crowdsourcing techniques and then the data augmentation was performed to generate 1000 images per class. The datasets were split into the train, test and validation sets with the ratio of 70:20:10. The train and test sets were used while training the model and after the training, it was validated using the validation set. After training the model for certain epochs, VGG16 architecture outperformed other models with a training accuracy of 99.82%, a test accuracy of 99.12% and a validation accuracy of 95.5%. In the future, a greater number of images need to be included in the datasets and trained using other pre-trained models.

A Deep Learning-based Transfer Learning Approach Fine-Tuned for Detecting Diabetic Retinopathy

Diabetics have blood sugar levels that are abnormal. Because of this, their eyesight can be affected. Untreated diabetes has the potential to result in permanent blindness, a grave complication. Among the severe consequences, early detection stands out as a critical hurdle for ensuring effective treatment. Regrettably, accurately assessing fundus images relies on human expertise to determine the stage of diabetic retinopathy. A simplified detection process could greatly benefit a vast population. Leveraging convolutional neural networks (CNN), which have demonstrated efficacy across various domains, diabetic retinopathy identification has seen advancements. The possibility of preventing complete, irreversible blindness lies in timely identification, underscoring the need for a dependable screening mechanism. To enhance the efficiency of screening and reduce errors that plague current models, a novel system has been introduced, capable of categorizing distinct diabetic retinopathy stages. This study focuses on examining datasets of Indian diabetic retinopathy images, encompassing four classes: Microaneurysms (MA), Soft Exudates (SE), Hard Exudates (EX), and Hemorrhages (HE). By employing the proposed VGG16 architecture for feature extraction and the Logistic Regression classifier for classification, an impressive accuracy of 90.4% was achieved.

Efficient-VGG16: A Novel Ensemble Method for the Classification of COVID-19 X-ray Images in Contrast to Machine and Transfer Learning

Corona Virus Disease 2019 (COVID-19) is a contagious respiratory disease characterized by its high transmissibility and exponential spread, presenting persistent difficulties. This investigation sought to incorporate machine learning (ML) and transfer learning (TL) and introduced Efficient-VGG16, a novel ensemble approach that integrated the EfficientNet-B0 and VGG16 architectures to classify chest X-ray images as either COVID-19 positive or non-COVID. The COVID-Xray-5k dataset used in the research was split into different training and testing sets (with ratios of 80:20, 75:25, and 70:30), preprocessed, and then used to evaluate the approach’s effectiveness. Throughout the investigation, all procedures operated effectively at the same selected image size, 224*224. Preprocessing included Gaussian blur to remove noise and relics, while Contrast Limited Adaptive Histogram Equalization (CLAHE) adjusted contrast. Among the employed ML classifiers, XgBoost attained the highest accuracy, with F1 scores of 95.78% and 71.56%, respectively. In addition, the research investigated the effectiveness of TL using pre-trained convolutional neural networks (CNNs) such as VGG16, MobileNetV3, ResNet101, Densenet201, and EfficientNet-B0. EfficientNet-B0 achieved the highest levels of accuracy and F1 score, 99.31% and 93.72%, respectively. Furthermore, the proposed Efficient-VGG16 achieved accuracy and an F1 score of 99.46% and 98.41%, respectively, and outperformed TL with EfficientNet-B0 and XgBoost. The investigation demonstrated the effectiveness of the Efficient-VGG16 for accurately classifying COVID-19.

An Edge Computing-Based Factor-Aware Novel Framework for Early Detection and Classification of Melanoma Disease Through a Customized VGG16 Architecture With Privacy Preservation and Real-Time Analysis

An Edge Computing-Based Factor-Aware Novel Framework for Early Detection and Classification of Melanoma Disease Through a Customized VGG16 Architecture With Privacy Preservation and Real-Time Analysis

Improved Image Classification Task Using Enhanced Visual Geometry Group of Convolution Neural Networks

Convolutional Neural Networks (CNNs) have become essential to solving image classification tasks. One of the most frequent models of CNNs for image classification is the Visual Geometry Group (VGG). The VGG architecture is made up of multiple layers of convolution and pooling processes followed by fully connected layers. Among the various VGG models, the VGG16 architecture has gained great attention due to its remarkable performance and simplicity. However, the VGG16 architecture is still prone to have many parameters contributing to its complexity. Moreover, the complexity of VGG16 may cause a longer execution time. The complexity of VGG16 architecture is also more highly prone to overfitting and may affect the classification accuracy. This study proposes an enhancement of VGG16 architecture to overcome such drawbacks. The enhancement involved the reduction of the convolution blocks, implementing batch normalization (B.N.) layers, and integrating global average pooling (GAP) layers with the addition of dense and dropout layers in the architecture. The experiment was carried out with six benchmark datasets for image classification tasks. The results from the experiment show that the network parameters are 79% less complex than the standard VGG16. The proposed model also yields better classification accuracy and shorter execution time. Reducing the parameters in the proposed improved VGG architecture allows for more efficient computation and memory usage. Overall, the proposed improved VGG architecture offers a promising solution to the challenges of long execution times and excessive memory usage in VGG16 architecture.

Improved Image Classification Task Using Enhanced Visual Geometry Group of Convolution Neural Networks

Convolutional Neural Networks (CNNs) have become essential to solving image classification tasks. One of the most frequent models of CNNs for image classification is the Visual Geometry Group (VGG). The VGG architecture is made up of multiple layers of convolution and pooling processes followed by fully connected layers. Among the various VGG models, the VGG16 architecture has gained great attention due to its remarkable performance and simplicity. However, the VGG16 architecture is still prone to have many parameters contributing to its complexity. Moreover, the complexity of VGG16 may cause a longer execution time. The complexity of VGG16 architecture is also more highly prone to overfitting and may affect the classification accuracy. This study proposes an enhancement of VGG16 architecture to overcome such drawbacks. The enhancement involved the reduction of the convolution blocks, implementing batch normalization (B.N.) layers, and integrating global average pooling (GAP) layers with the addition of dense and dropout layers in the architecture. The experiment was carried out with six benchmark datasets for image classification tasks. The results from the experiment show that the network parameters are 79% less complex than the standard VGG16. The proposed model also yields better classification accuracy and shorter execution time. Reducing the parameters in the proposed improved VGG architecture allows for more efficient computation and memory usage. Overall, the proposed improved VGG architecture offers a promising solution to the challenges of long execution times and excessive memory usage in VGG16 architecture.

Classification of Pistachio Images Using VGG16 and VGG19 Deep Learning Models

The value of the economy provided by pistachios to the countries where they are grown is increasing day by day. From this point of view, the importance of correct classification of pistachios is known. The more accurately the harvested pistachios are classified, the better the monetary return value. In this study, two different classes of pistachios were classified using VGG16 and VGG19 deep learning architectures. There are 2148 pieces of Kirmizi and Siirt Pistachio in the dataset. Experimental studies were carried out with 5-fold crossvalidation. As a result of the experimental studies, the Accuracy value of 0.802117 and the F1-measure value of 0.830593 were obtained from the average of 5 folds from the VGG16 deep learning model. Likewise, the Accuracy value of 0.779404 and the F-measure value as 0.779404 were obtained from the average of 5 folds from the VGG19 deep learning model.

Classification of diseases in snake plants using convolutional neural network

snake plant has an important role in human life, as well as in increasing the aesthetic value of the environment. Limited knowledge about diseases in snake plants has a crucial result in improper handling and control when the plant is attacked by disease. Advances in deep learning technology and Convolutional Neural Network (CNN) have presented high opportunities with their advantages in recognizing patterns and features from image data. This research will use a CNN model with VGG-19 architecture to classify diseases in the leaves of the snake plant. It is expected that by using the pre-trained VGG-19 model, the model can recognize complex visual patterns in snake plants. Diseases to be classified include several types of diseases that often attack snake plants such as anthracnose, rust, water soaked lesion, and healthy plants for comparison. The highest value of training accuracy reached a value of 98.08%, validation accuracy of 94.02%, and testing accuracy reached 94%.

Revolutionizing cardiovascular risk prediction: A novel image-based approach using fundus analysis and deep learning

Worldwide, CVDs continue to stand as a prominent contributor to mortality. Detecting and assessing CVD risk early is paramount for effective prevention and management. Medical imaging has gained prominence as a tool for evaluating CVD risk factors. In this study, a novel Inception v3 with a VGG16 is proposed to forecast cardiovascular risk rates via readily available and non-invasive fundus images. This approach harnesses advanced image analysis techniques encompassing contrast enhancement and noise reduction. The blood vessel segmentation and Optic disc detection of the pertinent features are extracted from the fundus images. In this context, the Inception v3 architecture is initially employed to capture intricate hierarchical patterns within the images. Alternatively, explore the utilization of the VGG16 architecture. By integrating these features with clinical data, the model is then trained to predict cardiovascular risk rates. Empirical findings underscore the method's remarkable accuracy in risk rate prediction. This non-invasive, image-based methodology holds transformative potential for reshaping early diagnosis and risk management approaches for cardiovascular diseases. Ultimately, this innovation stands to enhance patient care and outcomes.

Ethnicity Classification: Discriminating Between Iranian and Asian Populations Using Hybrid Deep Learning Algorithm

The rapid advancement of deep learning techniques has opened new avenues for solving complex classification problems in various domains. This research explores the application of Convolutional Neural Networks (CNN), specifically the VGG16 architecture, in conjunction with Support Vector Machines (SVM) for the classification of ethnicity using Iranian and Asian facial images. The classification of ethnicity based on facial features is a challenging task due to the subtle and complex variations within and between different ethnic groups. The proposed methodology involves a two-step process. First, the VGG16 CNN model is utilized to extract high-level features from the facial images. The pre-trained VGG16 model, known for its depth and representational power, is fine-tuned on the dataset to capture relevant ethnic features. The extracted features are then fed into an SVM classifier, which is trained to distinguish between Iranian and Asian facial characteristics. A comprehensive dataset consisting of labeled Iranian and Asian facial images is compiled and preprocessed for training and evaluation. The model's performance is assessed using metrics such as accuracy. Various experiments are conducted to optimize hyperparameters and validate the generalization capability of the proposed model. Additionally, visualization techniques are employed to provide insights into the features learned by the CNN and the decision boundaries established by the SVM. The results indicate that the combined approach of CNN-VGG16 and SVM yields promising accuracy in the classification of ethnicity based on facial images.

Performance Evaluation of CNN Models for Alzheimer’s Disease Detection with MRI Scans

The primary symptom of Alzheimer's disease is memory impairment, which is a neurodegenerative condition. The manifestation of these symptoms can be attributed to the impairment of the cerebral nerve responsible for cognitive functions such as learning, thinking and memory. Alzheimer’s disease is a prominent cause of mortality and lacks a definitive curve. However, appropriate medicinal interventions have demonstrated the potential to mitigate the progression and severity of the condition. This study presents the comparison of Convolutional Neural Network (CNN) models, namely ResNet50, VGG19 and VGG16 architectures, as an approach to construct an automated classification system for Alzheimer’s disease in future. The study utilises Magnetic Resonance Imaging (MRI) datasets to identify MRI datasets of individual with Alzheimer’s disease (AD), Cognitively normal (CN), mild cognitive impairment (MCI), early mild cognitive impairment (EMCI), and late mild cognitive impairment (LMCI). In the conducted experiment, the study achieved accuracy rates of 91.18% and 94.56% while utilising an epoch size of 2. The accuracy results indicate that the VGG16 model outperforms the ResNet50 model. The utilisation of automated Alzheimer’s disease classification holds potential as an auxiliary tool for healthcare professionals in determining the stage of Alzheimer’s disease hence facilitating the administration of suitable medicinal interventions.

Brain Tumor Detection and Classification Using VGG16 Deep Learning Algorithm and Python Imaging Library

Early diagnosis and treatment of brain cancer depend on the detection and categorization of brain tumors. Deep learning algorithms have produced amazing results in medical imaging applications including tumor identification. Most of this field's research has concentrated on applying CNN algorithms like VGG16, DNN, and ANN to this problem. This work describes the identification and classification of brain tumors using the Python Imaging Library (PIL) and the VGG16 deep learning algorithm. A dataset of 7000 MRI pictures categorized by tumor type served as the foundation for the research. The main objective of this study was to develop a high-efficiency, high-accuracy model. We suggested utilizing the VGG16 architecture and preprocessing images with PIL to ensure consistent images for training on a sizable dataset of brain magnetic resonance imaging (MRI) images. A novel technique we have used in our work is one that can analyze a single image and predict the presence of a tumor from the results. The research's methods produced robust tumor detection across the dataset with 96, 9% accuracy, indicating the value of the method in helping medical professionals make informed decisions when diagnosing the presence of tumors.

SDFPoseGraphNet: Spatial Deep Feature Pose Graph Network for 2D Hand Pose Estimation.

In the field of computer vision, hand pose estimation (HPE) has attracted significant attention from researchers, especially in the fields of human-computer interaction (HCI) and virtual reality (VR). Despite advancements in 2D HPE, challenges persist due to hand dynamics and occlusions. Accurate extraction of hand features, such as edges, textures, and unique patterns, is crucial for enhancing HPE. To address these challenges, we propose SDFPoseGraphNet, a novel framework that combines the strengths of the VGG-19 architecture with spatial attention (SA), enabling a more refined extraction of deep feature maps from hand images. By incorporating the Pose Graph Model (PGM), the network adaptively processes these feature maps to provide tailored pose estimations. First Inference Module (FIM) potentials, alongside adaptively learned parameters, contribute to the PGM's final pose estimation. The SDFPoseGraphNet, with its end-to-end trainable design, optimizes across all components, ensuring enhanced precision in hand pose estimation. Our proposed model outperforms existing state-of-the-art methods, achieving an average precision of 7.49% against the Convolution Pose Machine (CPM) and 3.84% in comparison to the Adaptive Graphical Model Network (AGMN).