Pre-trained Convolutional Neural Network Model Research Articles

Toward a Vision-Based Intelligent System: A Stacked Encoded Deep Learning Framework for Sign Language Recognition.

Sign language recognition, an essential interface between the hearing and deaf-mute communities, faces challenges with high false positive rates and computational costs, even with the use of advanced deep learning techniques. Our proposed solution is a stacked encoded model, combining artificial intelligence (AI) with the Internet of Things (IoT), which refines feature extraction and classification to overcome these challenges. We leverage a lightweight backbone model for preliminary feature extraction and use stacked autoencoders to further refine these features. Our approach harnesses the scalability of big data, showing notable improvement in accuracy, precision, recall, F1-score, and complexity analysis. Our model's effectiveness is demonstrated through testing on the ArSL2018 benchmark dataset, showcasing superior performance compared to state-of-the-art approaches. Additional validation through an ablation study with pre-trained convolutional neural network (CNN) models affirms our model's efficacy across all evaluation metrics. Our work paves the way for the sustainable development of high-performing, IoT-based sign-language-recognition applications.

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.

An efficient hybrid model of CNNs and different kernels of SVM for brain tumor classification

Technological advancements have had a profound impact on various aspects of human existence. The realm of medicine is one major area where technology has made important advances. We will talk specifically about the role technology has had in treating brain tumors, a serious and widespread condition. A large number of people pass away from brain tumors every year. Patients with BTs have a worse likelihood of survival when they receive subpar care and a false diagnosis. The most widely employed technique for detecting brain tumors is magnetic resonance imaging (MRI). Moreover, MRI is extensively utilized in medical imaging and image processing to identify variations in various regions of the body. A convolutional neural network (CNN)-based model was developed in this study to classify brain tumor. Using nine pre-trained CNN models (efficientnetb0, mobilenetv2, nasnetlarge, resnet50, resnet10, googlenet, vgg16, vgg19, and shufflenet), deep features were extracted from the acquired images. Then use a Support Vector Machines (SVM) classifier to classify the deep features. The classification accuracy results obtained from the various kernel functions, namely linear, gaussian, cubic, and quadratic—was then compared. The deep features retrieved from the efficientnetb0 model allowed accurate classification of brain tumors. The classification accuracy achieved using the Gaussian kernel function of SVM was recorded at an impressive 99.78%.

Classification-based Method for Wall Crack Detection System

It is imperative to ensure that building inspectors have adequate resources and tools to conduct their inspections efficiently and effectively. Relying solely on manual labour to check for wall cracks is inconvenient and may prove inefficient and poor use of time and financial recourses. Besides, there are concerns regarding the need for skilled inspectors due to their limited accessibility and the subjective nature of their evaluations. Previously, image processing and artificial intelligence have been independently utilized to identify wall cracks and estimate their width. However, more can be done when integrating these two approaches to produce a comprehensive solution. This study presented a technique to indicate wall cracks utilizing a pre-trained Convolutional Neural Network (CNN) model called Squeezenet. Then, the following image processing can precisely estimate the width of the cracks in pixels. Based on the total models studied, 78% were successfully detected and classified into their respective crack groups. Although 22% of the remaining models were mistakenly classified, the system still managed to detect the presence of cracks in them accurately. This study only considers analyzing projected cracks categorized as minor, moderate and major. Nevertheless, the discussion does not address the translation of pixel approximations into their respective physical measurements.

Exploiting sublimated deep features for image retrieval

Deep learning techniques can be used to describe image content, which is a good way to reduce the semantic gap between low-level and high-level features. However, convolutional neural networks (CNNs) exhibit texture bias and largely ignore global object shape. This does not conform to human perception and can fail to gain the advantages of both low-level features and deep features. To address this problem, in this work, the main focus was shifted from using simple deep features to the use of sublimated deep features, which incorporate global object shape and color features. Along these lines, in this work, a novel image retrieval method named the sublimated deep feature histogram (SDFH) was proposed. The main highlights are: 1) An effect orientation feature was introduced, namely, orientation-selective feature, to mimic the orientation-selection mechanism. This provides a good representation of global object shape and reduces the side-effects of texture bias. 2) A new concept was developed, namely color perceptual feature, to address the shortcoming of deep features—that they discard color features. This includes color cues in the deep features and provides a more discriminating representation. 3) The orientation selective and color perception mechanisms were effectively mimicked to provide a compact yet efficient representation, and propose an effective transfer learning method called gain whitening learning. By carrying out comparative experiments, it was demonstrated that sublimated deep features can provide highly competitive retrieval performance (in terms of mean average precision) using a pre-trained CNN model applied to well-known benchmark datasets. Our results provide new insights into image retrieval based on the mechanisms of the primary visual cortex (V1). Furthermore, the method is more in line with human perception than other methods.

DETECTION OF CHRONIC VENOUS INSUFFICIENCY CONDITION USING TRANSFER LEARNING WITH CONVOLUTIONAL NEURAL NETWORKS BASED ON THERMAL IMAGES

Chronic Venous Insufficiency (CVI) is a venous incompetence condition that leads to improper blood circulation from the lower limbs towards the heart. This occurs as a result of blood pooling in the veins of the leg, resulting in twisted, dilated, and tortuous veins. Aging, obesity, prolonged standing or sitting, and lack of mobility are all important causes of the occurrence of this chronic disease. The cost of CVI diagnosis and treatment is extremely high. Infrared thermographic image analysis is used for early detection and reduces the cost of diagnosis. Deep learning (DL) techniques play an important role in early prediction and may aid clinicians in diagnosing CVI. An automated classification model will assist the physician in making a precise diagnosis of the abnormal vein and treating the patient according to the severity of the condition. There is a need for a model that can perform successful classification without the need for pre-processing when compared to the traditional machine learning (ML) methods that depend on ideal manual feature extraction to achieve optimal outcomes. In this research, we recommend the customized DenseNet-121 architecture for CVI detection and compare it with other advanced DL models to determine its efficacy. DenseNet-121 and other pre-trained convolutional neural network models, including EfficientNetB0 and Inception_v3, were trained using a transfer learning strategy. The experimental findings indicate that the proposed modified DenseNet-121 model outperformed other classical methods. The reported results provide evidence of the robustness of the suggested method in addition to the high accuracy that it possessed, as shown by the overall testing accuracy of 97.4%. Thus, this study can be considered as a non-invasive and cost-effective approach for diagnosing chronic venous insufficiency condition in lower extremity.

Classification of Pneumonia and Covid-19 using Convolutional Neural Network

Purpose: The early and exact classification and identification is necessary for proper treatment which needs excessive time and effort of professionals. This examination is meant to foster a task to recognize Pneumonia and Coronavirus utilizing the idea of the Convolutional Neural Network (CNN) for picture grouping and is centered on building the profound learning model that aids in the characterization utilizing chest X-beam pictures in one of the quick and financially savvy ways. Design/Methodology/Approach: This study uses a wide dataset comprising of chest X-beam pictures accumulated from the Mendeley dataset. Include extraction strategies like picture pre-handling and data augmentation are applied to improve the arrangement execution. The framework utilizes the ResNet-18, which is a sort of CNN model for order. The examination includes assessing the exactness, accuracy, review, F1 score, and area under the receiver working trademark bend (AUC-ROC) for every classification model. Findings/Result: The dataset is separated into preparing and testing subsets to ensure unbiased performance evaluation. For the development and deployment of an accurate and reliable system, factors like data quality, model interpretability, and ethical considerations are considered. We successfully used the pre-trained ResNet-18 CNN model with chest X-ray image data that helped to build a robust classification system with a learning rate of 0.0001 and epoch size 10 having approx. 98.12% train accuracy and 97.70% test accuracy. Since the start of the project, we researched several methodologies to build the system. The other models (e.g., ResNet-50) were too big algorithms for our problem which created a problem of overfitting. Hence performance was not very accurate. So, we planned to go with the ResNet-18 model. As per our plan, we developed a system that operates as expected. Originality/Value: It helps medical professionals in diagnosing and managing these diseases. Paper Type: Research paper

Classification of defects in wooden structures using pre-trained models of convolutional neural network

Wooden structures, over time, are challenged by different types of defects. Due to mechanical and weathering effects, these defects can occur in the form of cracks, live and dead knots, dampness, and others. Because of the risk of damage or complete failure, treatment of these defects is necessary, but doing so necessitates their proper identification and classification (categorization). Crack identification and categorization must be part of the inspection procedure for engineering structures in the built environment. Convolutional neural networks (CNNs), a sub-type of Deep Learning (DL), can automatically classify the images of wooden structures to identify such defects. In this study, ten pre-trained models of CNN, namely ResNet18, ResNet50, ResNet101, ShuffleNet, GoogLeNet, Inception-V3, MobileNet-V2, Xception, Inception-ResNet-V2, and NASNet-Mobile are evaluated for the tasks of classification and prediction of defects in wooden structures. Each pre-trained CNN model is additionally trained and validated on an image dataset of 9000 images, equally divided into three classes: cracks, knots, and intact (undamaged). A smaller dataset of 300 images is separately used for testing purposes. Statistical parameters such as accuracy, precision, recall, and F1-score are computed for each CNN model. The Inception-V3 model proved to be the best CNN model for classifying defects in wooden structures based on the model’s accuracy, processing time and overall performance.

Efficient pollen grain classification using pre-trained Convolutional Neural Networks: a comprehensive study

Pollen identification is necessary for several subfields of geology, ecology, and evolutionary biology. However, the existing methods for pollen identification are laborious, time-consuming, and require highly skilled scientists. Therefore, there is a pressing need for an automated and accurate system for pollen identification, which can be beneficial for both basic research and applied issues such as identifying airborne allergens. In this study, we propose a deep learning (DL) approach to classify pollen grains in the Great Basin Desert, Nevada, USA. Our dataset consisted of 10,000 images of 40 pollen species. To mitigate the limitations imposed by the small volume of our training dataset, we conducted an in-depth comparative analysis of numerous pre-trained Convolutional Neural Network (CNN) architectures utilizing transfer learning methodologies. Simultaneously, we developed and incorporated an innovative CNN model, serving to augment our exploration and optimization of data modeling strategies. We applied different architectures of well-known pre-trained deep CNN models, including AlexNet, VGG-16, MobileNet-V2, ResNet (18, 34, and 50, 101), ResNeSt (50, 101), SE-ResNeXt, and Vision Transformer (ViT), to uncover the most promising modeling approach for the classification of pollen grains in the Great Basin. To evaluate the performance of the pre-trained deep CNN models, we measured accuracy, precision, F1-Score, and recall. Our results showed that the ResNeSt-110 model achieved the best performance, with an accuracy of 97.24%, precision of 97.89%, F1-Score of 96.86%, and recall of 97.13%. Our results also revealed that transfer learning models can deliver better and faster image classification results compared to traditional CNN models built from scratch. The proposed method can potentially benefit various fields that rely on efficient pollen identification. This study demonstrates that DL approaches can improve the accuracy and efficiency of pollen identification, and it provides a foundation for further research in the field.

An Ensemble Classification Method Based on Deep Neural Networks for Breast Cancer Diagnosis

Advances in technology have led to advances in breast cancer screening by detecting symptoms that doctors have overlooked. In this paper, an automatic detection system for breast cancer cases based on Internet of Things (IoT) is proposed. First, using IoT technology, direct medical images are sent to the data repository after the suspicious person's visit through medical equipment equipped with IoT. Then, in order to help radiologists, interpret medical images as best as possible, we use four pre-trained convolutional neural network models including InceptionResNetV2, InceptionV3, VGG19 and ResNet152. These models are combined by an ensemble classifier. Also, these models are used to accurately predict cases with breast cancer, healthy people, and cases with pneumonia by using two datasets of X-RAY and CT-scan in a three-class classification. Finally, the best result obtained for CT-scan images belongs to InceptionResNetV2 architecture with 99.36% accuracy and for X-RAY images belongs to InceptionV3 architecture with 96.94% accuracy. The results show that this method leads to a reduction in daily visits to medical centers and thus reduces the pressure on the medical care system. It also helps radiologists and medical staff to detect breast cancer in its early stages.

An Automatic Non-Destructive External and Internal Quality Evaluation of Mango Fruits based on Color and X-ray Imaging with Machine Learning and Deep Learning Based Classification Models

Quality evaluation of food products, agricultural produce to be specific, has gained momentum from past few decades due to the increased awareness among consumers across the world. This has resulted in the increased emphasis on the development and use of quality assessment techniques in food industry. Moreover, there is a need to automate the quality monitoring of agricultural produce like fruits and vegetables which is otherwise done manually in developing countries hence labor intensive, time consuming and subjective in nature. This paper presents an empirical analysis to build a rapid, robust, real-time, non-destructive computer vision based quality assessment model for mango fruits. The work employs the automatic disease classification of mango fruits based on machine and deep learning models. Firstly, the dataset of colored mango fruits images with 2279 images falling into three classes and another dataset of soft X-ray images of mango fruits with 572 images belonging to two quality classes are developed for detecting external and internal defects, respectively. The multilayer perceptron neural network (MLP NN) with two hidden layers, which may be considered as the starting point for deep learning technique, is proposed as machine learning model to classify the color images of mango fruits into one of three external quality classes with 95.1% accuracy and also to classify the soft X-ray images into two internal quality classes with 97.5% accuracy. In order to step out of feature engineering, actual deep learning convolutional neural network (CNN) models, a customized CNN model and pre-trained CNN models, VGGNet (VGG16) and DenseNet121 were also explored for mango disease classification. The maximum validation accuracy of custom CNN was found to be with 91.52% and 98.7% for color and augmented X-ray images, respectively. The classification accuracy of pre-trained models were found to be reasonably good for the color images but exhibited high variability in results and made it difficult to draw a general conclusion for the proposed datasets. However, the proposed MLP NN model based on few basic intensity and geometric features and also the proposed customized CNN model were found to be the best models and they outperform the state of the art reported in the literature.

Covid 19 image classification using hybrid averaging transfer learning model

The outbreak of Corona Virus 2019(Covid-19) is a great threat to the whole world. It is crucial to early detect patients infected with covid-19 and treat them to mitigate the rapid spread of this disease. It is an immediate priority to overcome the traditional screening and develop an accurate as well as speedy covid-19 automatic diagnosis system. Computer Tomography (CT) and Chest X-Ray imaging coupled with deep learning models to develop and test Computer Aided Screening (CAS) of covid-19 images from the normal images. In this paper classification and screening of covid-19 disease are performed by using pre-trained convolutional neural networks and a proposed hybrid model on an available standard dataset of chest X-Ray images. The proposed hybrid model employs the pre-trained Convolutional Neural Network models and Transfer Learning models. Our proposed model consists of three stages where extraction of features is performed in first stage by using pre-trained machine learning model. Deep features are extracted by using the infusion of the Transfer Learning Technique in the second stage of the model. The third stage uses Flatten and Classification layers to diagnose of Covid-19 patients. In order to assure the consistency of the proposed model, by considering standard dataset X-Ray images. Simulation results of performance metrics of Accuracy, F1 Score, Precision, Recall, ROC, and AUC curve, and training and testing loss are used to evaluate and compare the proposed model with existing models. Experimental result demonstrates that the hybrid model improves the screening process for Covid-19 disease by achieving higher accuracy.

EFF_D_SVM: a robust multi-type brain tumor classification system.

Brain tumors are one of the most threatening diseases to human health. Accurate identification of the type of brain tumor is essential for patients and doctors. An automated brain tumor diagnosis system based on Magnetic Resonance Imaging (MRI) can help doctors to identify the type of tumor and reduce their workload, so it is vital to improve the performance of such systems. Due to the challenge of collecting sufficient data on brain tumors, utilizing pre-trained Convolutional Neural Network (CNN) models for brain tumors classification is a feasible approach. The study proposes a novel brain tumor classification system, called EFF_D_SVM, which is developed on the basic of pre-trained EfficientNetB0 model. Firstly, a new feature extraction module EFF_D was proposed, in which the classification layer of EfficientNetB0 was replaced with two dropout layers and two dense layers. Secondly, the EFF_D model was fine-tuned using Softmax, and then features of brain tumor images were extracted using the fine-tuned EFF_D. Finally, the features were classified using Support Vector Machine (SVM). In order to verify the effectiveness of the proposed brain tumor classification system, a series of comparative experiments were carried out. Moreover, to understand the extracted features of the brain tumor images, Grad-CAM technology was used to visualize the proposed model. Furthermore, cross-validation was conducted to verify the robustness of the proposed model. The evaluation metrics including accuracy, F1-score, recall, and precision were used to evaluate proposed system performance. The experimental results indicate that the proposed model is superior to other state-of-the-art models.

UTILIZING DEEP LEARNING ON WEB APPLICATION FOR COVID-19 PNEUMONIA DETECTION

Quarantining Coronavirus (COVID-19) patients is a critical measure to prevent the spread of the disease, thus it is important to perform the correct diagnosis quickly. Among other diagnosis tools, radiograph scans exist as one of the common screening methods for detecting COVID-19. This study aims to reduce the healthcare workers’ burden by mitigating the impacts of COVID-19 pneumonia by developing an image classification system to assess radiographs in an accurate and timely manner. The proposed work also assists junior radiologists in familiarizing themselves with reading radiographs confidently using localization features through Explainable Artificial Intelligence. This project focuses on the detection of COVID-19 pneumonia in a multi-classification environment by pre-trained convolutional neural network models such as Inception-V3, VGG-16, and VGG-19. The overall system’s idea is to first perform transfer learning based on the binary classification of Normal and Pneumonia. The knowledge is then transferred to the task-specific models to classify COVID-19, bacterial, viral pneumonia, and normal. Out of all the models, VGG-19, the multiclass CNN model outperformed others by achieving a 99.27% accuracy in detecting COVID-19 in X-rays. The model also has the highest overall accuracy of 85.59% compared to multiclass Inception-V3, VGG-16, VGG-19, and the ensemble model. It was then chosen to be deployed as the Web Application where users are able to pass an input of radiograph to get an output with the label classifying whether the patient is infected with COVID-19 pneumonia, viral pneumonia, bacterial pneumonia, or normal together with highlights on the infected areas of lungs.

An Ensemble Classification Method Based on Deep Neural Networks for Breast Cancer Diagnosis

Advances in technology have led to advances in breast cancer screening by detecting symptoms that doctors have overlooked. In this paper, an automatic detection system for breast cancer cases based on Internet of Things (IoT) is proposed. First, using IoT technology, direct medical images are sent to the data repository after the suspicious person's visit through medical equipment equipped with IoT. Then, in order to help radiologists, interpret medical images as best as possible, we use four pre-trained convolutional neural network models including InceptionResNetV2, InceptionV3, VGG19 and ResNet152. These models are combined by an ensemble classifier. Also, these models are used to accurately predict cases with breast cancer, healthy people, and cases with pneumonia by using two datasets of X-RAY and CT-scan in a three-class classification. Finally, the best result obtained for CT-scan images belongs to InceptionResNetV2 architecture with 99.36% accuracy and for X-RAY images belongs to InceptionV3 architecture with 96.94% accuracy. The results show that this method leads to a reduction in daily visits to medical centers and thus reduces the pressure on the medical care system. It also helps radiologists and medical staff to detect breast cancer in its early stages.

Optimized Bayesian convolutional neural networks for invasive breast cancer diagnosis system

Deep learning (DL) has depicted unprecedented results in the diagnosis of invasive ductal carcinoma (IDC). However, most DL methods are not able to infer the uncertainty in their predictions, and are often too confident which may cause undesirable consequences. The ability of DL models to express their uncertainty in their predictions is particularly important in diagnosing diseases. Therefore, this paper proposes an uncertainty-aware approach called optimized Bayesian convolutional neural network (OBCNN) for the automated detection of IDC from histopathology images and is capable of estimating and inferring the uncertainty in its predictions. The proposed OBCNN employs an approximate Bayesian version of a pre-trained convolutional neural network (CNN) model named ResNet101V2 that is fine-tuned to the histopathology images of IDC. The Bayesian approximation is performed by adding Monte Carlo dropout (MC-dropout) to ResNet101V2 architecture. Due to the effect of MC-dropout on the obtained posterior predictive distribution (PPD), the value of MC-dropout should be chosen carefully because if the MC-dropout value is too large, the predictive distribution (PD) will be very diverse. Whereas if the MC-dropout value is very small, the PD will be very similar. Therefore, an optimization algorithm known as the slime mould algorithm is utilized to set the optimal value of the MC-dropout. During inference, the proposed OBCNN runs for T forward passes to configure the PPD. The mean, standard deviation (SD) and entropy are then calculated over the obtained PD to estimate the predictive uncertainty (PU) of the OBCNN. The proposed OBCNN achieved an accuracy of 93.83%, a precision of 96.14%, a G-mean of 93.04%, False Negative Rate of 4.72%, and False Positive Rate of 9.96 %. The proposed OBCNN has also been compared with recent related work, the comparison’s results demonstrated that OBCNN performed better in the diagnosis of IDC.

Intelligent Fault Diagnosis of Rolling Element Bearings Based on Modified AlexNet.

The reliable and safe operation of industrial systems needs to detect and diagnose bearing faults as early as possible. Intelligent fault diagnostic systems that use deep learning convolutional neural network (CNN) techniques have achieved a great deal of success in recent years. In a traditional CNN, the fully connected layer is located in the final three layers, and such a layer consists of multiple layers that are all connected. However, the fully connected layer of the CNN has the disadvantage of too many training parameters, which makes the model training and testing time longer and incurs overfitting. Additionally, because the working load is constantly changing and noise from the place of operation is unavoidable, the efficiency of intelligent fault diagnosis techniques suffers great reductions. In this research, we propose a novel technique that can effectively solve the problem of traditional CNN and accurately identify the bearing fault. Firstly, the best pre-trained CNN model is identified by considering the classification's success rate for bearing fault diagnosis. Secondly, the selected CNN model is modified to effectively reduce the parameter quantities, overfitting, and calculating time of this model. Finally, the best classifier is identified to make a hybrid model concept to achieve the best performance. It is found that the proposed technique performs well under different load conditions, even in noisy environments, with variable signal-to-noise ratio (SNR) values. Our experimental results confirm that this proposed method is highly reliable and efficient in detecting and classifying bearing faults.

TRANSFER LEARNING BASED CNN FOR COVID 19

Covid-19 has significantly impacted individuals and health care systems on a global scale. Finding COVID-19 is difficult because there aren't manytesting kits available. A deep transfer learning-based CNN technique is used to recognize Covid 19 in chest CTimages. Chest CT scans from both Covid and non-Covid sources are included in the dataset used to train themodel. Computed tomography (CT) is a faster and more precise means of infection diagnosis when compared to x-ray imaging. This is because CT scan pictures show more potential infection indicators than x-rays, which explains why. X-rays do, however, release less ionizing energy than a CT scan. Deep transfer learning approaches can both train and fine- tune the weights of networks that have previously been trained using datasets, which is why they are effective for this. Inorder to evaluate pictures, deep learning in computer vision especially uses convolutional neural networks (CNNs). For the classification of CT scan pictures to manage the complex structure, deep architecture is needed. We hypothesize that using radiographical fluctuations of covid-19 in CT images, pre-trained deep transfer learning-based CNN models may be able to extract certain graphical properties of covid-19 and provide a clinical diagnosis. The main advantages oftransfer learning include reduced training time, improved neural network performance, and the absence of a large amount of data. Key Words: Covid 19, CNN, ResNet101, Transfer Learning

Evaluating Performances of Attention-Based Merge Architecture Models for Image Captioning in Indian Languages

Image captioning is a growing topic of research in which numerous advancements have been made in the past few years. Deep learning methods have been used extensively for generating textual descriptions of image data. In addition, attention-based image captioning mechanisms have also been proposed, which give state-ofthe- art results in image captioning. However, many applications and analyses of these methodologies have not been made in the case of languages from the Indian subcontinent. This paper presents attention-based merge architecture models to achieve accurate captions of images in four Indian languages- Marathi, Kannada, Malayalam, and Tamil. The widely known Flickr8K dataset was used for this project. Pre-trained Convolutional Neural Network (CNN) models and language decoder attention models were implemented, which serve as the components of the mergearchitecture proposed here. Finally, the accuracy of the generated captions was compared against the gold captions using Bilingual Evaluation Understudy (BLEU) as an evaluation metric. It was observed that the merge architectures consisting of InceptionV3 give the best results for the languages we test on, the scores discussed in the paper. Highest BLEU-1 scores obtained for each language were: 0.4939 for Marathi, 0.4557 for Kannada, 0.5082 for Malayalam, and 0.5201 for Tamil. Our proposed architectures gave much higher scores than other architectures implemented for these languages.

Image preprocessing and hyperparameter optimization on pretrained model MobileNetV2 in white blood cell image classification

White blood cells play a role in maintaining the immune system which consists of several types such as neutrophils, lymphocytes, monocytes, eosinophils and basophils. MobileNetV2 is one of the pretrained convolutional neural network (CNN) models that provides excellent advantages and performance in classifying images. In this research was conducted to find out how to apply optimization hyperparameters and the impact of image processing on white blood cell image classification using MobileNetV2, so that it is expected to find a combination of preprocessing and combination of hyperparameter values that can produce the highest accuracy value. To maximize the classification process, before classifying the image, several stages of image preprocessing are carried out, namely cropping, grayscale, resizing and augmentation. Hyperparameter tuning was carried out for an experiment to improve model performance. The three main parameters used in hyperparameter tuning are learning rate, batch size, and number of epochs. Performance optimization model performance will be measured using accuracy, sensitivity, specificity and using a confusion matrix. Based on the experimental results in this study, it shows that the best learning rate value is 0.00001, the best batch size value is 32, and the best epoch value is 250.