Visual Geometry Group Research Articles

Measuring the binary thickness of buccal bone of anterior maxilla in low-resolution cone-beam computed tomography via a bilinear convolutional neural network.

The thickness of the buccal bone of the anterior maxilla is an important aesthetic-determining factor for dental implant, which is divided into the thick (≥1 mm) and thin type (<1 mm). However, as a micro-scale structure that is evaluated through low-resolution cone-beam computed tomography (CBCT), its thickness measurement is error-prone under the circumstance of enormous patients and relatively inexperienced primary dentists. Further, the challenges of deep learning-based analysis of the binary thickness of buccal bone include the substantial real-world variance caused by pixel error, the extraction of fine-grained features, and burdensome annotations. This study built bilinear convolutional neural network (BCNN) with 2 convolutional neural network (CNN) backbones and a bilinear pooling module to predict the binary thickness of buccal bone (thick or thin) of the anterior maxilla in an end-to-end manner. The methods of 5-fold cross-validation and model ensemble were adopted at the training and testing stages. The visualization methods of Gradient Weighted Class Activation Mapping (Grad-CAM), Guided Grad-CAM, and layer-wise relevance propagation (LRP) were used for revealing the important features on which the model focused. The performance metrics and efficacy were compared between BCNN, dentists of different clinical experience (i.e., dental student, junior dentist, and senior dentist), and the fusion of BCNN and dentists to investigate the clinical feasibility of BCNN. Based on the dataset of 4,000 CBCT images from 1,000 patients (aged 36.15±13.09 years), the BCNN with visual geometry group (VGG)16 backbone achieved an accuracy of 0.870 [95% confidence interval (CI): 0.838-0.902] and an area under the receiver operating characteristic (ROC) curve (AUC) of 0.924 (95% CI: 0.896-0.948). Compared with the conventional CNNs, BCNN precisely located the buccal bone wall over irrelevant regions. The BCNN generally outperformed the expert-level dentists. The clinical diagnostic performance of the dentists was improved with the assistance of BCNN. The application of BCNN to the quantitative analysis of binary buccal bone thickness validated the model's excellent ability of subtle feature extraction and achieved expert-level performance. This work signals the potential of fine-grained image recognition networks to the precise quantitative analysis of micro-scale structures.

Explainable Deep Learning Approach for Multi-Class Brain Magnetic Resonance Imaging Tumor Classification and Localization Using Gradient-Weighted Class Activation Mapping

Brain tumors (BT) present a considerable global health concern because of their high mortality rates across diverse age groups. A delay in diagnosing BT can lead to death. Therefore, a timely and accurate diagnosis through magnetic resonance imaging (MRI) is crucial. A radiologist makes the final decision to identify the tumor through MRI. However, manual assessments are flawed, time-consuming, and rely on experienced radiologists or neurologists to identify and diagnose a BT. Computer-aided classification models often lack performance and explainability for clinical translation, particularly in neuroscience research, resulting in physicians perceiving the model results as inadequate due to the black box model. Explainable deep learning (XDL) can advance neuroscientific research and healthcare tasks. To enhance the explainability of deep learning (DL) and provide diagnostic support, we propose a new classification and localization model, combining existing methods to enhance the explainability of DL and provide diagnostic support. We adopt a pre-trained visual geometry group (pre-trained-VGG-19), scratch-VGG-19, and EfficientNet model that runs a modified form of the class activation mapping (CAM), gradient-weighted class activation mapping (Grad-CAM) and Grad-CAM++ algorithms. These algorithms, introduced into a convolutional neural network (CNN), uncover a crucial part of the classification and can provide an explanatory interface for diagnosing BT. The experimental results demonstrate that the pre-trained-VGG-19 with Grad-CAM provides better classification and visualization results than the scratch-VGG-19, EfficientNet, and cutting-edge DL techniques regarding visual and quantitative evaluations with increased accuracy. The proposed approach may contribute to reducing the diagnostic uncertainty and validating BT classification.

A comparative study of state-of-the-art deep learning architectures for rice grain classification

Accurate and efficient automated rice grain classification systems are vital for rice producers, distributors, and traders, offering improved quality control, cost optimization, and supply chain management. They also hold the potential to aid in the development of rice varieties that are more resistant to disease, pests, and environmental stress. While most existing studies in the rice classification domain rely on traditional machine-learning techniques that necessitate feature extraction engineering processes, our research explores the effectiveness of novel deep-learning models for this task. We evaluated the performance of various contemporary deep-learning models, including Residual Network (ResNet), Visual Geometry Group (VGG) network, EfficientNet, and MobileNet. These models were tested on a dataset comprising 75,000 images, classified into five different rice categories. We assessed each model using established evaluation metrics such as accuracy, F1 score, precision, recall, and per-class accuracy. Our findings showed that the EfficientNet-based model delivered the highest accuracy (99.67%), while the MobileNet-based model excelled in the speed of classification (2556 s). We concluded that, compared to traditional machine learning methods, the models employed in our study are highly scalable and capable of managing large volumes of complex data with millions of features and samples.

Skeletal age evaluation using hand X-rays to determine growth problems

A common clinical method for identifying anomalies in bone growth in infants and newborns is skeletal age estimation with X-ray images. Children’s bone abnormalities can result from several conditions including wounds, infections, or tumors. One of the most frequent reasons for bone issues is that most youngsters are affected by the slow displacement of bones caused by pressure applied to the growth plates as youngsters develop. The growth plate can be harmed by a lack of blood supply, separation from other parts of the bone, or slight misalignment. Problems with the growth plate prevent bones from developing, cause joint distortion, and may cause permanent joint injury. A significant discrepancy between the chronological and assessed ages may indicate a growth problem because determining bone age represents the real level of growth. Therefore, skeletal age estimation is performed to look for endocrine disorders, genetic problems, and growth anomalies. To address the bone age assessment challenge, this study uses the Radiological Society of North America’s Pediatric Bone Age Challenge dataset which contains 12,600 radiological images of the left hand of a patient that includes the gender and bone age information. A bone age evaluation system based on the hand skeleton guidelines is proposed in this study for the detection of hand bone maturation. The proposed approach is based on a customized convolutional neural network. For the calculation of the skeletal age, different data augmentation techniques are used; these techniques not only increase the dataset size but also impact the training of the model. The performance of the model is assessed against the Visual Geometry Group (VGG) model. Results demonstrate that the customized convolutional neural network (CNN) model outperforms the VGG model with 97% accuracy.

Analysis and comparison of various deep learning models to implement suspicious activity recognition in CCTV surveillance

The paper aims to analyze and compare various deep learning (DL) algorithms in order to develop a Suspicious Activity Recognition (SAR) system for closed-circuit television (CCTV) surveillance. Automated systems for detecting and classifying suspicious activities are crucial as technology’s role in safety and security expands. This paper addresses these challenges by creating a robust SAR system using machine learning techniques. It analyzes and compares evaluation metrics such as Precision, Recall, F1 Score, and Accuracy using various deep learning methods (convolutional neural network (CNN), Long short-term memory (LSTM) – Visual Geometry Group 16 (VGG16), LSTM – ResNet50, LSTM – EfficientNetB0, LSTM – InceptionNetV3, LSTM – DenseNet121, and Long-term Recurrent Convolutional Network (LRCN)). The proposed system improves threat identification, vandalism deterrence, fight prevention, and video surveillance. It aids emergency response by accurately classifying suspicious activities from CCTV footage, reducing reliance on human security personnel and addressing limitations in manual monitoring. The objectives of the paper include analyzing existing works, extracting features from CCTV videos, training robust deep learning models, evaluating algorithms, and improving accuracy. The conclusion highlights the superior performance of the LSTM-DenseNet121 algorithm, achieving an overall accuracy of 91.17% in detecting suspicious activities. This enhances security monitoring capabilities and reduces response time. Limitations of the system include subjectivity, contextual understanding, occlusion, false alarms, and privacy concerns. Future improvements involve real-time object tracking, collaboration with law enforcement agencies, and performance optimization. Ongoing research is necessary to overcome limitations and enhance the effectiveness of CCTV surveillance.

MRI-based automated multitask deep learning system to evaluate supraspinatus tendon injuries.

To establish an automated, multitask, MRI-based deep learning system for the detailed evaluation of supraspinatus tendon (SST) injuries. According to arthroscopy findings, 3087 patients were divided into normal, degenerative, and tear groups (groups 0-2). Group 2 was further divided into bursal-side, articular-side, intratendinous, and full-thickness tear groups (groups 2.1-2.4), and external validation was performed with 573 patients. Visual geometry group network 16 (VGG16) was used for preliminary image screening. Then, the rotator cuff multitask learning (RC-MTL) model performed multitask classification (classifiers 1-4). A multistage decision model produced the final output. Model performance was evaluated by receiver operating characteristic (ROC) curve analysis and calculation of related parameters. McNemar's test was used to compare the differences in the diagnostic effects between radiologists and the model. The intraclass correlation coefficient (ICC) was used to assess the radiologists' reliability. p < 0.05 indicated statistical significance. In the in-group dataset, the area under the ROC curve (AUC) of VGG16 was 0.92, and the average AUCs of RC-MTL classifiers 1-4 were 0.99, 0.98, 0.97, and 0.97, respectively. The average AUC of the automated multitask deep learning system for groups 0-2.4 was 0.98 and 0.97 in the in-group and out-group datasets, respectively. The ICCs of the radiologists were 0.97-0.99. The automated multitask deep learning system outperformed the radiologists in classifying groups 0-2.4 in both the in-group and out-group datasets (p < 0.001). The MRI-based automated multitask deep learning system performed well in diagnosing SST injuries and is comparable to experienced radiologists. Our study established an automated multitask deep learning system to evaluate supraspinatus tendon (SST) injuries and further determine the location of SST tears. The model can potentially improve radiologists' diagnostic efficiency, reduce diagnostic variability, and accurately assess SST injuries. • A detailed classification of supraspinatus tendon tears can help clinical decision-making. • Deep learning enables the detailed classification of supraspinatus tendon injuries. • The proposed automated multitask deep learning system is comparable to radiologists.

An Enhanced Spatio-Temporal Human Detected Keyframe Extraction

Due to the immense availability of Closed-Circuit Television surveillance, it is quite difficult for crime investigation due to its huge storage and complex background. Content-based video retrieval is an excellent method to identify the best Keyframes from these surveillance videos. As the crime surveillance reports numerous action scenes, the existing keyframe extraction is not exemplary. At this point, the Spatio-temporal Histogram of Oriented Gradients - Support Vector Machine feature method with the combination of Background Subtraction is appended over the recovered crime video to highlight the human presence in surveillance frames. Additionally, the Visual Geometry Group trains these frames for the classification report of human-detected frames. These detected frames are processed to extract the keyframe by manipulating an inter-frame difference with its threshold value to favor the requisite human-detected keyframes. Thus, the experimental results of HOG-SVM illustrate a compression ratio of 98.54%, which is preferable to the proposed work's compression ratio of 98.71%, which supports the criminal investigation.

An innovative lightweight 1D-CNN model for efficient monitoring of large-scale forest composition: a case study of Heilongjiang Province, China

ABSTRACT Large-scale forest composition mapping and change monitoring are essential for regional and national forest resource management, monitoring, and carbon stock assessment. However, the existing large-scale mapping methods are not effective enough in terms of efficiency and accuracy. To address this limitation, this study proposes a lightweight one-dimensional convolutional neural network (LW-CNN) model for forest composition mapping. The LW-CNN model is developed using Landsat imagery covering 470,700 km2 obtained from Google Earth Engine (GEE) collected during two periods (2007 and 2018). The proposed LW-CNN is compared with a visual geometry group with 16 convolutional layers (VGG16), a residual network with 34 convolutional layers (Resnet34), and a residual network with 50 convolutional layers (Resnet50) in terms of model accuracy and efficiency. The factors influencing forest composition change are analyzed using the structural equation model (SEM). The results show that the proposed LW-CNN model can outperform the other three models in terms of model accuracy, achieving a mean overall accuracy (OA) of: 0.75 and efficiency of 7–22-fold. The changed forest composition from 2007 to 2018 accounts for 29.6% of the total forest area. The SEM results show that the climate factors have the most significant effect on the forest composition change. This study presents an innovative model for large-scale forest composition mapping, which is proven to be both efficient and accurate. This study also provides insights into the factors that affect the forest composition change, which could be valuable for forest resource management, monitoring, and carbon stock assessment.

Localization of early infarction on non-contrast CT images in acute ischemic stroke with deep learning approach

Localization of early infarction on first-line Non-contrast computed tomogram (NCCT) guides prompt treatment to improve stroke outcome. Our previous study has shown a good performance in the identification of ischemic injury on NCCT. In the present study, we developed a deep learning (DL) localization model to help localize the early infarction sign on NCCT. This retrospective study included consecutive 517 ischemic stroke (IS) patients who received NCCT within 12 h after stroke onset. A total of 21,436 infarction patches and 20,391 non-infarction patches were extracted from the slice pool of 1,634 NCCT according to brain symmetricity property. The generated patches were fed into different pretrained convolutional neural network (CNN) models such as Visual Geometry Group 16 (VGG16), GoogleNet, Residual Networks 50 (ResNet50), Inception-ResNet-v2 (IR-v2), Inception-v3 and Inception-v4. The selected VGG16 model could detect the early infarction in both supratentorial and infratentorial regions to achieve an average area under curve (AUC) 0.73 after extensive customization. The properly tuned-VGG16 model could identify the early infarction in the cortical, subcortical and cortical plus subcortical areas of supratentorial region with the mean AUC > 0.70. Further, the model could attain 95.6% of accuracy on recognizing infarction lesion in 494 out of 517 IS patients.

A Novel Profound Learning-Based System for Breast Cancer Location and Classification Utilizing Exchange Learning

Breast cancer is the leading cause of death in Asian women. Breast cancer can be detected early and treated aggressively. In this paper, we propose a new deep learning framework utilizing the index learning concept for breast cancer diagnosis and imaging. Typically, in-depth questions from experts are designed and presented in a specific format. Unlike classical learning paradigms, which are based on parallel development and emergence, the goal is to use the knowledge gained from solving one problem to solve another problem. In this study, we used CNN pretrainers to extract features from images, such as Google Net, VGGNet (Visual Geometry Group Network), and ResNet (Unsupervised Network), which combine the principles of neural networks and social networks. This has already happened. Contains partial information. Use the blunt treasure in the veil. A standard data model was used to evaluate the proposed optical products. Chemical images of different types of breast cancer have been discovered in archaeological finds and have proven useful in educational exploration into the unknown.

VGG-based discrimination for pneumonia images enhanced by Generative Adversarial Network

Due to problems such as blurring of chest X-ray image and double shadow of heart and lung, the accuracy of judging whether there is COVID-19 is not high. This study intends to use Generative Adversarial Network (GAN) to segment images, and then put the segmented image set into the classification algorithm to determine whether it is infected by the pneumonia. The data set is divided into “normal” and “pneumonia”. In order to remove the interference of rib and heart on lung contour in X-ray image, GAN technology is used to segment the original data set. The segmented image only retains the most important lung silhouette, which is convenient for the next step of recognition and information extraction. The classification algorithm uses Visual Geometry Group (VGG)-16. During the training, only fully connected layer 1 and fully connected layer 2 layers were trained, and the rest layers were frozen. When the model loads the original weight, the top level of the model will be replaced by the head model. The head model is placed on the basic model and becomes part of the actual training model to determine the best weight. After the original X-ray image is processed by GAN, the accuracy of the segmented lung image is improved from 84% to 98%. Loss decreased from 17% to 3%, and the accuracy of classification results was greatly improved. Therefore, it is proved that the segmented image can effectively remove irrelevant parts in the recognition process, and the accuracy is improved due to the removal of noise interference and clearer lung contour.

An Innovative Method for Lung Cancer Identification Using Machine Learning Algorithms

Biological community and the healthcare sector have greatly benefited from technological advancements in biomedical imaging. These advantages include early cancer identification and categorization, prognostication of patients' clinical outcomes following cancer surgery, and prognostication of survival for various cancer types. Medical professionals must spend a lot of time and effort gathering, analyzing, and evaluating enormous amounts of wellness data, such as scan results. Although radiologists spend a lot of time carefully reviewing several scans, tiny nodule diagnosis is incredibly prone to inaccuracy. Low dose computed tomography (LDCT) scans are used to categorize benign (Noncancerous) and malignant (Cancerous) nodules in order to study the issue of lung cancer (LC) diagnosis. Machine learning (ML), Deep learning (DL), and Artificial intelligence (AI) applications aid in the rapid identification of a number of infectious and malignant diseases, including lung cancer, using cutting-edge convolutional neural network (CNN) and Deep CNN architectures, we propose three unique detection models in this study: SEQUENTIAL 1 (Model-1), SEQUENTIAL 2 (Model-2), and transfer learning model Visual Geometry Group, VGG 16 (Model-3). The best accuracy model and methodology that are proposedas an effective and non-invasive diagnostic tool, outperforms other models trained with similar labels using lung CT scans to identify malignant nodules. Using a standard LIDC-IDRI data set that is freely available, the deep learning models are verified. The results of the experiment show a decrease in false positives while an increase in accuracy.

OPTIMIZATION OF PREDICTION AND PREVENTION OF DEFECTS ON METAL BASED ON AI USING VGG16 ARCHITECTURE

Manufacturing is one of the most valuable industries in the world, it can be automated without limits but still stuck in traditional manual and slow processes. Industry 4.0 is racing to define a new era in digital manufacturing through the implementation of Machine Learning methods. In this era, Machine learning has been widely applied to various fields and will certainly be very good applied in the manufacturing world. One of them is used to predict and prevent defects in metal. The process of predicting and preventing defects in metal is one of the important efforts in improving and maintaining production quality. Accuracy in predicting and preventing defects in metal can be an innovation and competitiveness in technology, both in production methods, and improving product safety and its users. Human operators and inspectors without digital assistance generally can spend a lot of time researching visual data, especially in high-volume production environments. For this reason, there needs to be research in developing Machine Learning technology in an effort to prevent the occurrence of defects in metal. And one of the development of this technology by using Convolutional Neural Network (CNN) architecture Visual Geometry Group 16 layer (VGG16). As for the metal defect dataset with 10 classes with details for training data as many as 17221, and test dataset as many as 4311, From the use of methods and datasets available, has been done training model used and produce very good accuracy, that is equal to 89% and testing with accuracy equal to 76%. And also done Interpreter process against new input data, to know metal defect type, prediction accuracy and appropriate action to prevent and overcome metal defect type result of Interpreter process application.

Vibration Event Recognition Using SST-Based Φ-OTDR System.

We propose a method based on Synchrosqueezing Transform (SST) for vibration event analysis and identification in Phase Sensitive Optical Time-Domain Reflectometry (Φ-OTDR) systems. SST has high time-frequency resolution and phase information, which can distinguish and enhance different vibration events. We use six tap events with different intensities and six other events as experimental data and test the effect of attenuation. We use Visual Geometry Group (VGG), Vision Transformer (ViT), and Residual Network (ResNet) as deep classifiers for the SST transformed data. The results show that our method outperforms the methods based on Continuous Wavelet Transform (CWT) and Short-Time Fourier Transform (STFT), while ResNet is the best classifier. Our method can achieve high recognition rate under different signal strengths, event types, and attenuation levels, which shows its value for Φ-OTDR system.

A DEEP LEARNING METHODOLOGY FOR PLANT SPECIES RECOGNITION USING MORPHOLOGY OF LEAVES

Plants play a crucial role in supporting all forms of life on Earth, not just humans but every living organism. Understanding the diverse range of plant species that surround us is essential due to their significance in various aspects of human life, including agriculture, the environment, medicine, cosmetics, and more. Advancements in machine learning and computer vision algorithms have opened possibilities for identifying different types of plant species, both within and across classes. Plant species detection typically involves several steps, such as image acquisition, feature extraction, categorization, and pre-processing. In this study, three datasets—namely Flavia, Swedish, and the intelligent computing laboratory (ICL) dataset—were chosen for experimentation purposes. For feature extraction, three different models were employed: k-nearest neighbour (KNN), naive Bayes (NB), and the visual geometry group (VGG)-16 model. These models were used to extract distinctive features such as shape, texture, venation, and margin from the plant images. A multiclass classification task was conducted to categorize the plant species. Among the models tested, the VGG-16 model consistently demonstrated superior performance in terms of accuracy. Specifically, when using the VGG-16 model, the obtained accuracies were 96.68% for the Flavia dataset, 97.65% for the Swedish dataset, and 96.11% for the ICL dataset.

Neural Networks for the Diagnosis of Covid-19 in Chest X-ray Images: A Systematic Review and Meta-Analysis

Introduction: The COVID-19 pandemic has triggered a global crisis with significant repercussions in terms of mortality and an ever-increasing demand for urgent medical care, particularly in emergency care settings. This demand arises mainly from the prevailing need to carry out real-time diagnoses and provide immediate care to patients at high risk of serious complications. With the purpose of addressing this problem in a rigorous manner, we have carried out a systematic review focused on evaluating the effectiveness of models based on neural networks for the diagnosis of COVID-19 from chest x-ray images. Methods: This review has been carried out through an exhaustive search in various renowned electronic bibliographic databases, such as Scopus, IEEE Xplore, PubMed and ScienceDirect. The search period has been extended until September 2023, culminating in the identification of a total of 1,250 relevant articles. Results: The culminating phase of our review involved the inclusion of 37 studies that met rigorously established selection criteria. These studies have been the subject of a thorough analysis, where various performance metrics such as accuracy/precision, sensitivity/recall, specificity and the F1 value (F1-score) have been evaluated. Conclusions: Our results reveal that the VGG16 (Visual Geometry Group 16) model, based on neural networks, has emerged as the most widely adopted, manifesting itself in 13.04% of the total models analyzed and in 16.21% of the models supported by the 37 studies. selected. Notably, this algorithm has exhibited an impressive accuracy of over 99% in predicting the diagnosis of patients with suspected COVID-19.

The study and application of brain tumors classification based on artificial intelligence

The application of AI in the medical field has been increasingly popular due to its ability to enhance the accuracy and efficiency of diagnoses. A plurality of individuals employs diverse techniques and refine existing methods in order to achieve greater precision. This paper provides an overview of the methodology utilized in early research papers for brain tumors classification. This includes the input of datasets, preprocessing, model building, training and testing, evaluation, and application. In addition, this paper presents four models used in early research, including the Artificial neural network (ANN) which mimics the organization and operation of neural networks in living organisms by processing information through interconnected nodes or neurons. Another model is k-Nearest Neighbor (KNN), an instance-based learning algorithm that labels new data points by comparing them to the K closest labeled data points in the training dataset. The third model is Visual Geometry Group-16 (VGG-16), a 16-layer Convolutional Neural Network (CNN) that is highly regarded for its simplicity and effectiveness and is one of the top-performing CNN models used by VGG. Lastly, GN-AlexNet is a hybrid learning mode that combines the GoogleNet structure with the AlexNet model. The remainder of the article explains the utilization of this AI in three application scenarios, namely as an assistant for doctors, patients, and insurance companies. Moreover, the paper highlights potential challenges that may arise in both the present and future, such as patient mistrust and the emergence of new models like Vision Transformer (ViT) that could potentially outperform CNNs.

Research on optimization of forest fire early warning task by neural architecture search

We employed Neural Architecture Search (NAS) technology in this research paper and com-pared it with the classic convolutional neural network structures LeNet-15 and VGG (Visu-al Geometry Group)16. The objective was to optimize the performance of early warning forest fire image classification tasks. We adopted a public dataset "Forest Fire Early Warning 2 Clas-sification" from the Baidu PaddlePaddle platform, which comprises images of fires and no-fires under various environmental conditions. A convolutional neural network (CNN) model was automatically designed through AutoKeras and underwent 20 training epochs. In the ex-perimental results, NAS outperformed with an accuracy rate of 92%, surpassing LeNet-15 (83%) and VGG16 (49%). However, its training time was longer at 33 seconds, and GPU utili-zation was higher, ranging from 28% to 33%. Despite the room for improvement in training time and resource utilization, NAS has proven its superiority in complex image classification tasks due to its high accuracy. Although NAS has room for enhancement in terms of training time and resource utilization, its outstanding performance in the image recognition task of ear-ly forest fire warning shows great potential for future research and application.

Comparison and analysis of deep neural networks in facial expression recognition

Recognizing facial expressions automatically is of importance to interactive computer systems since facial expression is an efficient means of conveying emotions. Over the past few years, many researchers have attempted to use deep learning for expression recognition. The advantage of deep learning lies in its ability to learn features from datasets automatically, without relying on hand-crafted features. The paper analyzes what mechanism is useful for expression recognition in deep learning by comparing the performance of different popular models and algorithms from recent research. The paper trains the models on the Facial Expression Recognition Dataset (FER-2013), which is a relatively small and imbalanced dataset. After that, model performance is assessed on the private test dataset of FER-2013. Specifically, Residual Network (ResNet), Visual Geometry Group Network (VGGNet), and MobileNet are evaluated in the experiment. The evaluation is based on running time, the number of parameters, and the accuracy. Squeeze and Excitation (SE) block is utilized in the ResNet, which enables the models to learn useful features from global information. In the paper, ResNet34 inserted with SE block (SE-ResNet34) get the highest private test accuracy of 70.80%. Experimental results show that the residual learning enables models to go deeper without degradation and the SE block is beneficial for the model to learn global information.

Facial emotion detection based on improved VGG-16

In real life, facial emotion recognition is very important because it can convey information, build relationships, and facilitate communication. Therefore, emotion recognition technology is used in medicine, education, entertainment, security, and other fields. In the emotion detection field, the Facial Emotion Recognition 2013 Dataset (FER-2013) is a dataset that has been used in many places, that contains images of seven emotional expressions. In the area of detecting emotions from facial expressions, the deep learning structures, especially the convolutional neural networks (CNNs), have demonstrated significant potential since they have the ability to extract features and their computational efficiency. In this paper, the author constructs a model named Improved VGG-16 based on Visual Geometry Group Network of 16 weight layers (VGG-16). To be specific, first, the author adds two dense layers to improve the complexity and expressiveness; second, two dropout layers are used in order to reduce overfitting. An accuracy of 68.0% is achieved by this model on the test dataset of FER-2013. The result is better than some previous methods and shows that the improved VGG-16 model can recognize facial expressions effectively. In conclusion, this work aims to increase the accuracy and reliability of facial emotion recognition, providing support for research and application in related fields.