Visual Geometry Group Research Articles

Deep learning-based phenotype classification of three ark shells: Anadara kagoshimensis, Tegillarca granosa, and Anadara broughtonii

The rapid and accurate classification of aquatic products is crucial for ensuring food safety, production efficiency, and economic benefits. However, traditional manual methods for classifying ark shell species based on phenotype are time-consuming and inefficient, especially during peak seasons when the demand is high and labor is scarce. This study aimed to develop a deep learning model for the automated identification and classification of commercially important three ark shells (Tegillarca granosa, Anadara broughtonii, and Anadara kagoshimensis) from images. The ark shells were collected and identified using a polymerase chain reaction method developed in a previous study, and a total of 1,400 images were categorized into three species. Three convolutional neural network (CNN) models, Visual Geometry Group Network (VGGnet), Inception-Residual Network (ResNet), and SqueezeNet, were then applied to two different classification sets, Set-1 (four bivalve species) and Set-2 (three ark shell species). Our results showed that SqueezeNet demonstrated the highest accuracy during the training phase for both classification sets, whereas Inception-ResNet exhibited superior accuracy during the validation phase. Similar results were obtained after developing a third classification set (Set-3) to classify six categories by combining Set-1 and Set-2. Overall, the developed CNN-based classification model exhibited a performance comparable or superior to that presented in previous studies and can provide a theoretical basis for bivalve classification, thereby contributing to improved food safety, production efficiency, and economic benefits in the aquatic products industry.

Development of a deep-learning algorithm for age estimation on CT images of the vertebral column

PurposeThe accurate age estimation of cadavers is essential for their identification. However, conventional methods fail to yield adequate age estimation especially in elderly cadavers. We developed a deep learning algorithm for age estimation on CT images of the vertebral column and checked its accuracy. MethodFor the development of our deep learning algorithm, we included 1,120 CT data of the vertebral column of 140 patients for each of 8 age decades. The deep learning model of regression analysis based on Visual Geometry Group-16 (VGG16) was improved in its estimation accuracy by bagging. To verify its accuracy, we applied our deep learning algorithm to estimate the age of 219 cadavers who had undergone postmortem CT (PMCT). The mean difference and the mean absolute error (MAE), the standard error of the estimate (SEE) between the known- and the estimated age, were calculated. Correlation analysis using the intraclass correlation coefficient (ICC) and Bland-Altman analysis were performed to assess differences between the known- and the estimated age. ResultsFor the 219 cadavers, the mean difference between the known- and the estimated age was 0.30 years; it was 4.36 years for the MAE, and 5.48 years for the SEE. The ICC (2,1) was 0.96 (95 % confidence interval: 0.95–0.97, p < 0.001). Bland-Altman analysis showed that there were no proportional or fixed errors (p = 0.08 and 0.41). ConclusionsOur deep learning algorithm for estimating the age of 219 cadavers on CT images of the vertebral column was more accurate than conventional methods and highly useful.

Convolutional neural network for estimation of harvest time of forage sorghum (sorghum bicolor) cultivar samurai-1

One of the economic alternatives to improve the quality of ruminant feed is combining grass as the main feed with high-protein forages such as sorghum. To get a quality sorghum harvest during the period, it must be right when it has good biomass content, nutrients, and digestibility. The problem is that measuring quality in the laboratory has additional costs and time, which is not short, causing delays. An approach with machine learning using a convolutional neural network can be a better solution. This research uses a convolutional neural network algorithm with the right architecture to estimate sorghum harvest time from imaging results of unmanned aerial vehicles. The stages of this research include data collection, pre-processing, modeling, and finally, the evaluation stage. This research compares the results of several convolutional neural network (CNN) algorithm architectural models: simple CNN, ResNet50 V2, visual geometry group-16 (VGG-16), MobileNet V2, and Inception V3. The result is determining the CNN algorithm architectural model that can estimate sorghum harvest time with maximum accuracy. The best result is the simple CNN architectural model with an accuracy of 0.95. This research shows that the classification model obtained from the CNN algorithm with a simple CNN architecture is the choice model for estimating sorghum harvest time.

Computer-Aided Diagnosis of Duchenne Muscular Dystrophy Based on Texture Pattern Recognition on Ultrasound Images Using Unsupervised Clustering Algorithms and Deep Learning

ObjectiveThe feasibility of using deep learning in ultrasound imaging to predict the ambulatory status of patients with Duchenne muscular dystrophy (DMD) was previously explored for the first time. The present study further used clustering algorithms for the texture reconstruction of ultrasound images of DMD data sets and analyzed the difference in echo intensity between disease stages. Methodsk-means (Kms) and fuzzy c-means (FCM) clustering algorithms were used to reconstruct the DMD data-set textures. Each image was reconstructed using seven texture-feature categories, six of which were used as the primary analysis items. The task of automatically identifying the ambulatory function and DMD severity was performed by establishing a machine-learning model. ResultsThe experimental results indicated that the Gaussian Naïve Bayes and k-nearest neighbors classification models achieved an accuracy of 86.78% in ambulatory function classification. The decision-tree model achieved an identification accuracy of 83.80% in severity classification. A deep convolutional neural network model was established as the main structure of the deep-learning model while automatic auxiliary interpretation tasks of ambulatory function and severity were performed, and data augmentation was used to improve the recognition performance of the trained model. Both the visual geometry group (VGG)-16 and VGG-19 models achieved 98.53% accuracy in ambulatory-function classification. The VGG-19 model achieved 92.64% accuracy in severity classification. ConclusionRegarding the overall results, the Kms and FCM clustering algorithms were used in this study to reconstruct the characteristic texture of the gastrocnemius muscle group in DMD, which was indeed helpful in quantitatively analyzing the deterioration of the gastrocnemius muscle group in patients with DMD at different stages. Subsequent combination of machine-learning and deep-learning technologies can automatically and accurately assist in identifying DMD symptoms and tracking DMD deterioration for long-term observation.

One-shot font generation via local style self-supervision using Region-Aware Contrastive Loss

Compositional scripts like Hangeul (Korean characters) and Chinese characters involve numerous characters, making manual font design labor-intensive and cost-ineffective work. Although many few-shot font generation methods have been introduced, they have at least one of the limitations, i.e., lacking local styles of font, additional component labeling, and high complexity in network structure and training. To solve these limitations, given our observation that font style can be perceived at a patch-level rather than a component-level, we propose Region-Aware Contrastive loss (RAC-loss) so that the generator can capture the local style by self-supervision. The proposed loss maximizes the style information between patches of the generated image and the style reference image. And we introduce an attention mechanism to the patch-level contrastive loss to handle multiple patch correspondences. This attention learns style similarity between two glyph images, which serves as a patch-correspondence map. RAC-loss gives more fine-grained feedback to the generator than component-level loss, allowing it to incorporate local styles, even in a straightforward structure like a visual geometry group network (VGGNet). This results in a fast inference latency (3.02 ms), and the proposed method achieved 43.18 mean Fréchet Inception Distance (mFID) on the test dataset, a notable decrease of 5.42 compared to the previous method.

A COMPREHENSIVE STUDY: INTRACRANIAL ANEURYSM DETECTION VIA VGG16-DENSENET HYBRID DEEP LEARNING ON DSA IMAGES

An intracranial aneurysm is a swelling in a weak area of a brain artery. The main cause of aneurysm is high blood pressure, smoking, and head injury. A ruptured aneurysm is a serious medical emergency that can lead to coma and then death. A digital subtraction angiogram (DSA) is used to detect a brain aneurysm. A neurosurgeon carefully examines the scan to find the exact location of the aneurysm. A hybrid model has been proposed to detect these aneurysms accurately and quickly. Visual Geometry Group 16 (VGG16) and DenseNet are two deep-learning architectures used for image classification. Ensembling both models opens the possibility of using diversity in a robust and stable feature extraction. The model results assist in identifying the location of aneurysms, which are much less prone to false positives or false negatives. This integration of a deep learning-based architecture into medical practice holds great promise for the timely and accurate detection of aneurysms. The study encompasses 1654 DSA images from distinct patients, partitioned into 70% for training (1157 images) and 30% for testing (496 images). The ensembled model manifests an impressive accuracy of 95.38%, outperforming the respective accuracies of VGG16 (94.38%) and DenseNet (93.57%). Additionally, the ensembled model achieves a recall value of 0.8657, indicating its ability to correctly identify approximately 86.57% of true aneurysm cases out of all actual positive cases present in the dataset. Furthermore, when considering DenseNet individually, it attains a recall value of 0.8209, while VGG16 attains a recall value of 0.8642. These values demonstrate the sensitivity of each model to detecting aneurysms, with the ensemble model showcasing superior performance compared to its individual components.

Preliminary Study Based on Myocardial Infarction Classification of 12-Lead Electrocardiography Images with Deep Learning Methods

In contemporary medicine, the development of computer-aided diagnostic systems using Electrocardiography (ECG) signals has gained significance for the diagnosis of heart diseases. Myocardial infarction (MI) is recognized as the condition where blood flow to the heart muscle is obstructed due to blockages in coronary vessels. In this study, four deep learning approaches were employed to automatically identify different MI conditions (STEMI, NSTEMI, USAP) using images generated from 12-lead ECG signals. The utilized architectures include deep neural networks such as Visual Geometry Group-16 (VGG-16), AlexNet, Residual Neural Network (ResNet), SqueezeNet and an ensemble model composed of these networks. With the proposed method, classification was performed based on 10-second grayscale images of 12-lead ECG signals for HC-STEMI, HC-NSTEMI, HC-USAP, and NSTEMI-STEMI conditions. According to the obtained results, the HC-STEMI group achieved the highest performance with a cross-validated 0.8237 F1 score using the AlexNet architecture. Among the novel contributions of this study is the image-based ECG classification method that can be more easily adapted to clinical applications and the analysis of the potential use of detecting different MI conditions in clinical practices. In conclusion, this study sheds light on future research by demonstrating the significant potential of using multi-channel ECG signals in image format for MI diagnosis, paving the way for advancements in this field.

Examining the classification performance of pre‐trained capsule networks on imbalanced bone marrow cell dataset

AbstractThe automatic detection of bone marrow (BM) cell diseases plays a vital role in the medical field; it helps to make diagnoses more precise and effective, which leads to early detection and can significantly improve patient outcomes and increase the chances of successful intervention. This study proposed a fully automated intelligent system for BM classification by developing and enhancing Capsule Neural Network (CapsNet) architecture. Although CapsNet has demonstrated success in many classification tasks, it still has some limitations and challenges associated with using Convolutional Neural Networks (CNNs), which suffer from information loss during the pooling and discarding of detailed spatial information, resulting in the loss of fine‐grained features. Additionally, CNNs must help capture hierarchical feature relationships and often learn them implicitly by stacking convolutional layers. In contrast, CapsNets are designed to capture hierarchical features through dynamic routing and relationships between capsules, resulting in a more explicit representation of spatial hierarchy. CapsNets manage transformations and offer equivariance, preserving spatial information through capsule routing mechanisms. Further, to improve how features are represented, pre‐trained models such as Residual Capsule Network (RES‐CapsNet), Visual Geometry Group Capsule Network (VGG‐CapsNet), and Google Network (Inception V3) (GN‐CapsNet) have been used. This helps the network obtain the low‐ and mid‐level features and information it has previously learned so that subsequent capsule layers receive better initial information. Additionally, the Synthetic Minority Over‐Sampling Technique (SMOTE) was implemented to mitigate class imbalance. It generates synthetic samples in feature space by over‐sampling the minority class, leading to improving model performance in accurately classifying rare instances. Fine‐tuning the hyperparameters and implementing these improvements resulted in remarkable accuracy rates on a large BM dataset, with reduced training time and trainable parameters. CapsNet achieved 96.99%, VGG‐CapsNet achieved 98.95%, RES‐CapsNet achieved 99.24%, and the GN‐CapsNet model demonstrated superior accuracy at 99.45%. GN‐Caps Net was the best because it requires a small number of epochs and has an effective deep inception architecture that efficiently extracts features at different scales to form a robust representation of the input. Our proposed models were compared with existing state‐of‐the‐art models using the BM dataset; the results showed that our models outperformed the existing approaches and demonstrated excellent performance. Further, this automated system can analyze large amounts of data and complex cells in images of the BM dataset. Thus, it gives healthcare professionals a detailed understanding of different diseases, which may take time to achieve manually.

A Novel Transfer Learning Approach to Improve Breast Cancer Diagnosing on Screening Mammography

Segmentation is a technique for separating an image into discrete areas in order to separate objects of interest from their surroundings. In image analysis, segmentation—which encompasses detection, feature extraction, classification, and treatment—is crucial. In order to plan treatments, segmentation aids doctors in measuring the amount of tissue in the breast. Categorizing the input data into two groups that are mutually exclusive is the aim of a binary classification problem. In this case, the training data is labeled in a binary format based on the problem being solved. Identifying breast lumps accurately in mammography pictures is essential for the purpose of prenatal testing for breast cancer. The proposed TLA (Transfer Learning Approach) based CNN (Convolution Neural Network) –TLA based CNN aims to offer binary classification for rapid and precise breast cancer diagnosis (benign and malignant). In order to predict the sub-type of cancer, this exploration as used Deep Learning techniques on the Histogram of Oriented Gradient (HOG) - Feature extraction technique that creates a local histogram of the image to extract features from each place in the image with CNN classifier. This research work employs two well-known pre-trained models, ResNet-50 and VGG16, to extract characteristics from mammography images. The high-level features from the Mammogram dataset are extracted using a transfer learning model based on Visual Geometry Group (VGG) with 16-layer and Residual Neural Network with 50-layers deep model architecture (ResNet-50). The proposed model TLA based CNN has achieved 96.49% and 95.48% accuracy as compared to ResNet50 and VGG16 in the breast cancer classification and segmentation.

Exploration of hyperparameter efficiency for image style transfer

Image style transfer is a popular computer vision technique that aims to merge the content of one image with the style of another to generate a unique, original image with a different aesthetic feel. Numerous models have been developed for various applications in this field, including portrait painting, art creation, and medical image processing, where additional information or annotations could be added to medical images, making them easier to read and understand. This study focuses on optimizing parameters within the pre-trained Visual Geometry Group (VGG19) network architecture, building on Google Brains 2017 work on Arbitrary style transfer in one model. The goal is to improve the quality and realism of the generated images by exploring different parameter combinations and fine-tuning weights and learning rates. This work carefully selects a range of styles and content to compare their effects during the optimization process. Finally, this fine-tuning process strikes a balance between content loss and style loss, which results in high-quality and more realistic images.

A novel approach to texture recognition combining deep learning orthogonal convolution with regional input features.

Textures provide a powerful segmentation and object detection cue. Recent research has shown that deep convolutional nets like Visual Geometry Group (VGG) and ResNet perform well in non-stationary texture datasets. Non-stationary textures have local structures that change from one region of the image to the other. This is consistent with the view that deep convolutional networks are good at detecting local microstructures disguised as textures. However, stationary textures are textures that have statistical properties that are constant or slow varying over the entire region are not well detected by deep convolutional networks. This research demonstrates that simple seven-layer convolutional networks can obtain better results than deep networks using a novel convolutional technique called orthogonal convolution with pre-calculated regional features using grey level co-occurrence matrix. We obtained an average of 8.5% improvement in accuracy in texture recognition on the Outex dataset over GoogleNet, ResNet, VGG and AlexNet.

Image Caption using VGG model and LSTM

Deep convolutional networks and recurrent neural networks have gained significant popularity in the field of image captioning tasks in recent times. As we all know the performance and the architecture of models are still eternal topic. We constructed the model using a new method to enhance its performance and accuracy. In our model, we make use of pretrained CNN model VGG (Visual Geometry Group) to extract image features, and learn caption sentence features using bidirectional LSTM(Long-Short-Term-Memory) which can better understand the meaning of sentences in the text. Then we combine the image features and caption features to predict captions for images. The dataset Flickr8K is used to train and test the model. Additionally, the model has the ability to produce captions that are shorter than a specified caption length. We evaluated our model with Bilingual Evaluation Understudy (BLEU) score which measures the similarity of predicted text and to the real text. After evaluation and comparison, our model is proved to be well-done on some conditions.

Fast style transfer for ethnic patterns innovation

Minority ethnic patterns in China, reflecting diverse cultures, serve as symbols of ethnic identity and cultural heritage. However, the current artistic domain of minority ethnic patterns faces dual challenges: limited artistic styles and insufficient innovation efficiency. In addressing these issues, this study proposes a novel approach for ethnic pattern innovation, named Fast Style Transfer for Ethnic Pattern Innovation (FST-EPI), aiming to enhance innovation efficiency and diversify artistic styles. FST-EPI comprises three key modules: background generation, background fusion, and fast transfer. Firstly, the background generation module utilizes the Visual Geometry Group network (VGGNet) to extract features from images, generating a background image with the desired style while preserving the necessary stylistic attributes. Subsequently, textile patterns are fused with the stylized background image, forming a content-background fusion image. Finally, the selected stylistic patterns, along with the content-background fusion image, are input into the fast transfer module to produce the ultimate innovative pattern. Experimental results demonstrate that the generated images not only fully retain the core features and cultural elements of minority ethnic patterns but also successfully integrate the unique styles from one or more reference images. This proposed method exhibits outstanding efficiency and feasibility in a broad range of experiments related to ethnic pattern style transfer, as evidenced by thorough comparisons with state-of-the-art methods. Ultimately, the processing speed of this method achieves a real-time output at the level of seconds, reducing processing time by nearly 50% compared to traditional approaches.

Design of Deep Learning Techniques for Side-Channel Attacks on Masked 128-bit AES Implementations

Researchers are exploring the use of convolutional neural networks (CNNs) in side-channel attacks to understand the weaknesses in cryptographic implementation. CNNs can learn hierarchical characteristics automatically from electromagnetic radiation or power usage during cryptographic processes. Researchers train CNNs on side-channel data to extract meaningful representations and deduce secret keys. Deep learning algorithms are helpful in evaluating the security of embedded systems, and CNNs are a feasible paradigm for profiling side-channel analysis attacks. In this paper, it has been introduced a VGG (Visual Geometry Group)-Net architecture, which is a typical deep convolutional neural network design with numerous layers. It uses the ASCAD dataset to conduct experiments. They found that VGG-Net architecture Side Channel Attacks (SCA) provides better results than the previously optimized CNN model by significantly reducing the number of side-channel traces required for successful attacks on desynchronized datasets. The researchers also discovered that synchronous traces serve as the pre-training source for VGG-Net architecture, functioning successfully in terms of jittering with minimal fine-adjusting after training

Optimizing Intracranial Hemorrhage Detection: A Comparative Study of VGG Deep Learning Architectures with Transfer Learning on CT Brain scans

Intracranial Hemorrhage (ICH) is a life-threatening condition that requires rapid and accurate diagnosis. This study applied Visual Geometry Group (VGG) deep learning model to binary classification of cerebral hemorrhage in Computerized Tomography (CT) brain scans, thereby improving the diagnostic process. Using a Kaggle dataset containing 6703 images, this study investigated the validity of VGG11, VGG13, VGG16, and VGG19 architectures, using transfer learning of ImageNet pre-trained weights to improve model performance. This study showed that VGG13 with transfer learning outperformed other models with 99.70% accuracy, indicating the profound impact of pre-trained weights on diagnostic accuracy. VGG models with untrained weights performed differently, with VGG11 achieving 98.61% accuracy, suggesting that less complex models were better suited to specific tasks. Comparative analysis shows that the depth of neural networks and the initialization of pre-training weights are the key factors to optimize performance. The discoveries present encouraging possibilities for integrating Artificial Intelligence (AI) into emergency medical services, with the potential to enhance the detection of cerebral hemorrhage and subsequently improve patient outcomes. While acknowledging limitations associated with dataset diversity and model generalization, the outcomes underscore the importance of tailoring models to specific task requirements. These findings also advocate for additional empirical research to fine-tune and advance the application of AI in healthcare.

Improving Performance Convolutional Neural Networks Using Modified Pooling Function

The Visual Geometry Group-16 (VGG16) network architecture, as part of the development of convolutional neural networks, has been popular among researchers in solving classification tasks, so in this paper, we investigated the number of layers to find better performance. In addition, we also proposed two pooling function techniques inspired by existing research on mixed pooling functions, namely Qmax and Qavg. The purpose of the research was to see the advantages of our method; we conducted several test scenarios, including comparing several modified network configurations based on VGG16 as a baseline and involving our pooling technique and existing pooling functions. Then, the results of the first scenario, we selected a network that can adapt well to our pooling technique, whichwas then carried out several tests involving the Cifar10, Cifar100, TinyImageNet, and Street View House Numbers (SVHN) datasets as benchmarks. In addition, we were also involved in several existing methods. The experiment results showed that Net-E has the highest performance, with 93.90% accuracy for Cifar10, 71.17% for Cifar100, and 52.84% for TinyImageNet. Still, the accuracy was low when the SVHN dataset was used. In addition, in comparison tests with several optimization algorithms using the Qavg pooling function, it can be seen that the best accuracy results lie in the SGD optimization algorithm, with 89.76% for Cifar10 and 89.06% for Cifar100.

A novel data-driven integrated detection method for network intrusion classification based on multi-feature imbalanced data

The multi-feature and imbalanced nature of network data has always been a challenge to be overcome in the field of network intrusion detection. The redundant features in data could reduce the overall quality of network data and the accuracy of detection models, because imbalance could lead to a decrease in the detection rate for minority classes. To improve the detection accuracy for imbalanced intrusion data, we develop a data-driven integrated detection method, which utilizes Recursive Feature Elimination (RFE) for feature selection, and screens out features that are conducive to model recognition for improving the overall quality of data analysis. In this work, we also apply the Adaptive Synthetic Sampling (ADASYN) method to generate the input data close to the original dataset, which aims to eliminate the data imbalance in the studied intrusion detection model. Besides, a novel VGG-ResNet classification algorithm is also proposed via integrating the convolutional block with the output feature map size of 128 from the Visual Geometry Group 16 (VGG16) of the deep learning algorithm and the residual block with output feature map size of 256 from the Residual Network 18 (ResNet18). Based on the numerical results conducted on the well-known NSL-KDD dataset and UNSW-NB15 dataset, it illustrates that our method can achieve the accuracy rates of 86.31% and 82.56% in those two test datasets, respectively. Moreover, it can be found that the present algorithm can achieve a better accuracy and performance in the experiments of comparing our method with several existing algorithms proposed in the recent three years.

Pneumonia prediction on chest x-ray images using deep learning approach

&lt;span lang="EN-US"&gt;Coronavirus disease 2019 (COVID-19) is an infectious disease with first symptoms similar to the flu. In many cases, this disease causes pneumonia. Since pulmonary infections can be observed through radiography images, this paper investigates deep learning methods for automatically analyzing query chest x-ray images. In deep learning, computers can automatically identify useful features for the model, directly from the raw data, bypassing the difficult step of manual information refinement. The main part of the deep learning method is the focus on automatically learning data representations. Visual geometry group 16 (VGG16) and DenseNet121 are methods in deep learning. The data used is a chest x-ray of pneumonia. Data is divided into training, testing, and validation. The best method for this research case is VGG16 with 93% accuracy training and 90% accuracy testing. In this study, DenseNet121 obtained accuracy below VGG16, with 92% accuracy in training and 88% for accuracy testing. Parameters have a significant influence on the accuracy of each model, and with the parameters that have been used, the VGG16 is a method that has high accuracy and can be used to predict chest x-ray images aimed at checking pneumonia in patients.&lt;/span&gt;

Hybrid model for detection of brain tumor using convolution neural networks

The development of aberrant brain cells, some of which may turn cancerous, is known as a brain tumor. Magnetic resonance imaging (MRI) scans are the most common technique for finding brain tumors. Information about the aberrant tissue growth in the brain is discernible from the MRI scans. In numerous research papers, machine learning, and deep learning algorithms are used to detect brain tumors. It takes extremely little time to forecast a brain tumor when these algorithms are applied to MRI pictures, and better accuracy makes it easier to treat patients. The radiologist can make speedy decisions because of this forecast. The proposed work creates a hybrid convolution neural networks (CNN) model using CNN for feature extraction and logistic regression (LR). The pre-trained model visual geometry group 16 (VGG16) is used for the extraction of features. To reduce the complexity and parameters to train we eliminated the last eight layers of VGG16. From this transformed model the features are extracted in the form of a vector array. These features fed into different machine learning classifiers like support vector machine (SVM), naïve bayes (NB), LR, extreme gradient boosting (XGBoost), AdaBoost, and random forest for training and testing. The performance of different classifiers is compared. The CNN-LR hybrid combination outperformed the remaining classifiers. The evaluation measures such as recall, precision, F1-score, and accuracy of the proposed CNN-LR model are 94%, 94%, 94%, and 91% respectively.

Vision Transformers, Ensemble Model, and Transfer Learning Leveraging Explainable AI for Brain Tumor Detection and Classification.

The abnormal growth of malignant or nonmalignant tissues in the brain causes long-term damage to the brain. Magnetic resonance imaging (MRI) is one of the most common methods of detecting brain tumors. To determine whether a patient has a brain tumor, MRI filters are physically examined by experts after they are received. It is possible for MRI images examined by different specialists to produce inconsistent results since professionals formulate evaluations differently. Furthermore, merely identifying a tumor is not enough. To begin treatment as soon as possible, it is equally important to determine the type of tumor the patient has. In this paper, we consider the multiclass classification of brain tumors since significant work has been done on binary classification. In order to detect tumors faster, more unbiased, and reliably, we investigated the performance of several deep learning (DL) architectures including Visual Geometry Group 16 (VGG16), InceptionV3, VGG19, ResNet50, InceptionResNetV2, and Xception. Following this, we propose a transfer learning(TL) based multiclass classification model called IVX16 based on the three best-performing TL models. We use a dataset consisting of a total of 3264 images. Through extensive experiments, we achieve peak accuracy of 95.11%, 93.88%, 94.19%, 93.88%, 93.58%, 94.5%, and 96.94% for VGG16, InceptionV3, VGG19, ResNet50, InceptionResNetV2, Xception, and IVX16, respectively. Furthermore, we use Explainable AI to evaluate the performance and validity of each DL model and implement recently introduced Vison Transformer (ViT) models and compare their obtained output with the TL and ensemble model.