Accuracy Of Image Classification Research Articles

Exploring the Differences in Tree Species Classification between Typical Forest Regions in Northern and Southern China

Focusing on the trend of continuously seeking high-precision tree species classification results in small areas from the perspectives of sensors and classification algorithms. This study aimed to explore the effects of data sources, classifiers, and seasons on classification accuracy in regions with significant environmental variation, examining patterns of tree species classification to enhance the transferability of classification. Considering two typical forest distribution regions in the north and south of China, this study utilized the revisitation cycle and open-source advantages of Sentinel-2 and Landsat-8. Leveraging the Google Earth Engine (GEE) platform, this study captured spectral features, vegetation indices, and texture features for single seasonal and seasonal combination images. With the assistance of Sentinel-1A and SRTM (Shuttle Radar Topography Mission) DEM (Digital Elevation Model), backscattering coefficient features and topographical features were extracted and input with features captured from Sentinel-2 and Landsat-8 into three types of classifiers: random forest (RF), support vector machine (SVM), and gradient tree boosting (GTB) for major tree species classification. In this research, we discovered that the best classification for single season in the northern study area was spring, whereas, for the southern study area, it was winter. Seasonal combination images effectively improved the classification accuracy of single seasonal images, with Sentinel-2 imagery displaying better classification performance compared to Landsat-8, and the optimal classifier differing between the north and the south. The inclusion of topographical or backscattering coefficient features in the four-season combination imagery contributed to improvements in classification accuracy, with topographical features significantly enhancing the classification performance in the topographically varied southern study area. The evaluation of feature importance indicated that elevation was the most critical feature for classification, while spectral features and vegetation indices were also significant. In the southern study area with large topographical discrepancies, subdividing into different terrain units led to improved tree species classification accuracy in medium-altitude, gentle slope areas. These findings provide insights into the regularity of enhancing tree species classification accuracy in environmentally diverse areas through the use of multi-source remote sensing data and multi-seasonal imagery. Consequently, the results offer a reference for the identification of tree species across large areas and the creation of spatial distribution maps.

Identification of Damaged Canopies in Farmland Artificial Shelterbelts Based on Fusion of Unmanned Aerial Vehicle LiDAR and Multispectral Features

With the decline in the protective function for agricultural ecosystems of farmland shelterbelts due to tree withering and dying caused by pest and disease, quickly and accurately identifying the distribution of canopy damage is of great significance for forestry management departments to implement dynamic monitoring. This study focused on Populus bolleana and utilized an unmanned aerial vehicle (UAV) multispectral camera to acquire red–green–blue (RGB) images and multispectral images (MSIs), which were fused with a digital surface model (DSM) generated by UAV LiDAR for feature fusion to obtain DSM + RGB and DSM + MSI images, and random forest (RF), support vector machine (SVM), maximum likelihood classification (MLC), and a deep learning U-Net model were employed to build classification models for forest stand canopy recognition for the four image types. The model results indicate that the recognition performance of RF is superior to that of U-Net, and U-Net performs better overall than SVM and MLC. The classification accuracy of different feature fusion images shows a trend of DSM + MSI images (Kappa = 0.8656, OA = 91.55%) > MSI images > DSM + RGB images > RGB images. DSM + MSI images exhibit the highest producer’s accuracy for identifying healthy and withered canopies, with values of 95.91% and 91.15%, respectively, while RGB images show the lowest accuracy, with producer’s accuracy values of 79.3% and 78.91% for healthy and withered canopies, respectively. This study presents a method for identifying the distribution of Populus bolleana canopies damaged by Anoplophora glabripennis and healthy canopies using the feature fusion of multi-source remote sensing data, providing a valuable data reference for the precise monitoring and management of farmland shelterbelts.

Res2Net-based multi-scale and multi-attention model for traffic scene image classification.

With the increasing applications of traffic scene image classification in intelligent transportation systems, there is a growing demand for improved accuracy and robustness in this classification task. However, due to weather conditions, time, lighting variations, and annotation costs, traditional deep learning methods still have limitations in extracting complex traffic scene features and achieving higher recognition accuracy. The previous classification methods for traffic scene images had gaps in multi-scale feature extraction and the combination of frequency domain, spatial, and channel attention. To address these issues, this paper proposes a multi-scale and multi-attention model based on Res2Net. Our proposed framework introduces an Adaptive Feature Refinement Pyramid Module (AFRPM) to enhance multi-scale feature extraction, thus improving the accuracy of traffic scene image classification. Additionally, we integrate frequency domain and spatial-channel attention mechanisms to develop recognition capabilities for complex backgrounds, objects of different scales, and local details in traffic scene images. The paper conducts the task of classifying traffic scene images using the Traffic-Net dataset. The experimental results demonstrate that our model achieves an accuracy of 96.88% on this dataset, which is an improvement of approximately 2% compared to the baseline Res2Net network. Furthermore, we validate the effectiveness of the proposed modules through ablation experiments.

Bi-sigmoid spike-timing dependent plasticity learning rule for magnetic tunnel junction-based SNN

In this study, we explore spintronic synapses composed of several Magnetic Tunnel Junctions (MTJs), leveraging their attractive characteristics such as endurance, nonvolatility, stochasticity, and energy efficiency for hardware implementation of unsupervised neuromorphic systems. Spiking Neural Networks (SNNs) running on dedicated hardware are suitable for edge computing and IoT devices where continuous online learning and energy efficiency are important characteristics. We focus in this work on synaptic plasticity by conducting comprehensive electrical simulations to optimize the MTJ-based synapse design and find the accurate neuronal pulses that are responsible for the Spike Timing Dependent Plasticity (STDP) behavior. Most proposals in the literature are based on hardware-independent algorithms that require the network to store the spiking history to be able to update the weights accordingly. In this work, we developed a new learning rule, the Bi-Sigmoid STDP (B2STDP), which originates from the physical properties of MTJs. This rule enables immediate synaptic plasticity based on neuronal activity, leveraging in-memory computing. Finally, the integration of this learning approach within an SNN framework leads to a 91.71% accuracy in unsupervised image classification, demonstrating the potential of MTJ-based synapses for effective online learning in hardware-implemented SNNs.

Boundary sample-based class-weighted semi-supervised learning for malignant tumor classification of medical imaging.

Medical image classification plays a pivotal role within the field of medicine. Existing models predominantly rely on supervised learning methods, which necessitate large volumes of labeled data for effective training. However, acquiring and annotating medical image data is both an expensive and time-consuming endeavor. In contrast, semi-supervised learning methods offer a promising approach by harnessing limited labeled data alongside abundant unlabeled data to enhance the performance of medical image classification. Nonetheless, current methods often encounter confirmation bias due to noise inherent in self-generated pseudo-labels and the presence of boundary samples from different classes. To overcome these challenges, this study introduces a novel framework known as boundary sample-based class-weighted semi-supervised learning (BSCSSL) for medical image classification. Our method aims to alleviate the impact of intra- and inter-class boundary samples derived from unlabeled data. Specifically, we address reliable confidential data and inter-class boundary samples separately through the utilization of an inter-class boundary sample mining module. Additionally, we implement an intra-class boundary sample weighting mechanism to extract class-aware features specific to intra-class boundary samples. Rather than discarding such intra-class boundary samples outright, our approach acknowledges their intrinsic value despite the difficulty associated with accurate classification, as they contribute significantly to model prediction. Experimental results on widely recognized medical image datasets demonstrate the superiority of our proposed BSCSSL method over existing semi-supervised learning approaches. By enhancing the accuracy and robustness of medical image classification, our BSCSSL approach yields considerable implications for advancing medical diagnosis and future research endeavors.

Construction of a painting image classification model based on AI stroke feature extraction

Abstract A large number of digital painting image resources cannot be directly converted into electronic form due to their differences in painting techniques and poor preservation of paintings. Moreover, the difficulty of extracting classification features can also lead to the consumption of human time and misclassification problems. The aim of this research is to address the challenges of converting various digital painting image resources into electronic form and the difficulties of accurately extracting classification features. The goal is to improve the usefulness and accuracy of painting image classification. Converting various digital painting image resources directly into electronic format and accurately extracting classification features are challenging due to differences in painting techniques and painting preservation, as well as the complexity of accurately extracting classification features. Overcoming these adjustments and improving the classification of painting features with the help of artificial intelligence (AI) techniques is crucial. The existing classification methods have good applications in different fields. But their research on painting classification is relatively limited. In order to better manage the painting system, advanced intelligent algorithms need to be introduced for corresponding work, such as feature recognition, image classification, etc. Through these studies, unlabeled classification of massive painting images can be carried out, while guiding future research directions. This study proposes an image classification model based on AI stroke features, which utilizes edge detection and grayscale image feature extraction to extract stroke features; and the convolutional neural network (CNN) and support vector machine are introduced into image classification, and an improved LeNet-5 CNN is proposed to achieve comprehensive assurance of image feature extraction. Considering the diversity of painting image features, the study combines color features with stroke features, and uses weighted K-means clustering algorithm to extract sample features. The experiment illustrates that the K-CNN hybrid model proposed in the study achieved an accuracy of 94.37% in extracting image information, which is higher than 78.24, 85.69, and 86.78% of C4.5, K-Nearest Neighbor (KNN), and Bi directional Long short-term Memory (BiLSTM) algorithms. In terms of image classification information recognition, the algorithms with better performance from good to poor are: the mixed model > BiLSTM > KNN > C4.5 model, with corresponding accuracy values of 0.938, 0.897, 0.872, and 0.851, respectively. And the number of fluctuation nodes in the mixed model is relatively small. And the sample search time is significantly shorter than other comparison algorithms, with a maximum recognition accuracy of 92.64% for the style, content, color, texture, and direction features of the image, which can effectively recognize the contrast and discrimination of the image. This method effectively provides a new technical means and research direction for digitizing image information.

Neural Technologies for Objects Classification with Mobile Applications

This paper is related to the study of the features of the neural technologies’ application, in particular, ResNet neural networks for the classification of objects in photographs. The work aims to increase the accuracy of recognition and classification of objects in photographs by using various models of the ResNet neural network. The paper analyzes the features of the application of the corresponding models in comparison with other architectures of deep neural networks and evaluates their efficiency and accuracy in the classification of objects in photographs. The process of data formation for training neural networks, their processing and sorting is described. A web application and a mobile application for recognizing and classifying objects in a photo were also developed. A system for classifying objects, in particular airplanes in photographs, was developed using neural network technologies. It gives a recognition and classification accuracy of about 95%. Research results of ResNet models are of great practical importance, as they can improve the classification accuracy of various images. Features of ResNet, such as the use of skip connections or residual connections, make it effective in the relevant tasks. The results of the study will help to implement ResNet in various fields, including medicine, automatic pattern recognition and other areas where the classification of objects in photographs is an important task.

Convolutional neural network with parallel convolution scale attention module and ResCBAM for breast histology image classification

Breast cancer is the most common cause of female morbidity and death worldwide. Compared with other cancers, early detection of breast cancer is more helpful to improve the prognosis of patients. In order to achieve early diagnosis and treatment, clinical treatment requires rapid and accurate diagnosis. Therefore, the development of an automatic detection system for breast cancer suitable for patient imaging is of great significance for assisting clinical treatment. Accurate classification of pathological images plays a key role in computer-aided medical diagnosis and prognosis. However, in the automatic recognition and classification methods of breast cancer pathological images, the scale information, the loss of image information caused by insufficient feature fusion, and the enormous structure of the model may lead to inaccurate or inefficient classification. To minimize the impact, we proposed a lightweight PCSAM-ResCBAM model based on two-stage convolutional neural network. The model included a Parallel Convolution Scale Attention Module network (PCSAM-Net) and a Residual Convolutional Block Attention Module network (ResCBAM-Net). The first-level convolutional network was built through a 4-layer PCSAM module to achieve prediction and classification of patches extracted from images. To optimize the network's ability to represent global features of images, we proposed a tiled feature fusion method to fuse patch features from the same image, and proposed a residual convolutional attention module. Based on the above, the second-level convolutional network was constructed to achieve predictive classification of images. We evaluated the performance of our proposed model on the ICIAR2018 dataset and the BreakHis dataset, respectively. Furthermore, through model ablation studies, we found that scale attention and dilated convolution play an important role in improving model performance. Our proposed model outperforms the existing state-of-the-art models on 200 × and 400 × magnification datasets with a maximum accuracy of 98.74 %.

LitefusionNet: Boosting the performance for medical image classification with an intelligent and lightweight feature fusion network

Medical image analysis plays a crucial role in modern healthcare for accurate diagnosis and treatment. However, the inherent challenges and limitations posed by the complexity and variability of medical images, coupled with the shortcomings of existing methods, necessitate the development of novel approaches. In this study, we propose LiteFusionNet (Lightweight Fusion Network), a lightweight model that effectively addresses these challenges, offering the advantage of accurate and efficient medical image classification while mitigating the computational demands. The LitefusionNet leverages the power of deep convolutional neural networks (DCNNs) and feature fusion techniques to achieve improved performance in medical image classification. LitefusionNet combines the strengths of MobileNet and MobileNetV2 architectures to extract robust features from medical images. These features capture discriminative information from different levels of abstraction, enhancing the model's ability to capture fine-grained patterns. The fusion process employs a concatenation method to combine the extracted features, resulting in a more comprehensive representation that improves the model's classification accuracy. To evaluate the effectiveness of LitefusionNet, extensive experiments are conducted on a diverse set of publicly available medical image datasets, including brain MRI, skin, CT, X-ray, and histology. The results demonstrate that LitefusionNet outperforms several existing models in terms of classification accuracy, showcasing its efficacy in different medical imaging modalities. Furthermore, we provide interpretability to the model's predictions by performing Grad-CAM analysis, enabling insights into the important regions in the medical images that contribute to the classification decision. In addition, we compare LitefusionNet with five pre-trained models. LiteFusionNet excels in medical image classification, boasting impressive accuracies across diverse datasets: 97.33% for brain MRI, 91.11% for skin, 99.00% for CT, 98.15% for X-ray, and 92.11% for histology. These results underscore LiteFusionNet's robust and versatile performance, making it a compelling solution for accurate and efficient medical image analysis. Overall, LitefusionNet strikes a balance between accuracy, efficiency, and real-time performance. Our findings demonstrate its potential as a promising solution for accurate and efficient medical image analysis, with applications in diagnostic support systems and clinical decision-making.

A new optimization approach based on neural architecture search to enhance deep U-Net for efficient road segmentation

Neural Architecture Search (NAS) has significantly improved the accuracy of image classification and segmentation. However, these methods concentrate on finding segmentation structures for natural or medical applications. In this study, we introduce a NAS approach based on gradient optimization to identify ideal cell designs for road segmentation. To the best of our knowledge, this work represents the first application of gradient-based NAS to road extraction. Taking insight from the U-Net model and its successful variations in different image segmentation tasks, we propose NAS-enhanced U-Net, illustrated by an equal number of cells in both encoder and decoder levels. While cross-entropy combined with dice loss is commonly used in many segmentation tasks, road extraction brings up a unique challenge due to class imbalance. To address this, we introduce a combination of loss function. This function merges cross-entropy with weighted Dice loss, focusing on elevating the importance of the road class by assigning it a weight (⍵), while background Dice values are disregarded. The results indicate that the optimal weight for the proposed model equals 2. Additionally, our work challenges the assumption that increased model parameters or depth inherently leads to improved performance. Therefore, we establish search spaces 2,3,4,5,6,7 and 8 to automatically choose the optimal depth for model. We present promising segmentation results for our proposed method, achieved without any pretraining on the Massachusetts road dataset. Furthermore, these results are compared with those of 14 models categorized into four groups: U-Net, Segnet, FCN8, and Nas-U-Net.

Defect Detection of Tomato Fruit Using Image Dataset

Abstract: The tomato crop is more susceptible to disease than any other vegetable, and it can be infected with over 200 diseases caused by different pathogens worldwide. Tomato plant diseases have become a challenge to food security globally. Currently, diagnosing and preventing tomato plant diseases is a challenge due to the lack of essential methods or tools.The traditional techniques of detecting plant disease are arduous and error-prone. Utilizing precise or automatic detection methods in spotting early plant disease can improve the quality of food production and reduce adverse effects. Deep learning has significantly increased the recognition accuracy of image classification and object detection systems in recent years. In this study, a 15-layer convolutional neural network is proposed as the backbone for single shot detector (SSD) to improve the detection of healthy, and three classes of tomato fruit diseases. The proposed model performance is compared with ResNet-50, AlexNet, VGG 16, and VGG19 as the backbone for Single shot detector. The findings of the experiment showed that the proposed CNN-SDD achieved 98.87% higher detection accuracy, which outperformed state-of-the-art models.

Study of Aerial Image Classification

Abstract: Classifying images captured through remote sensing is necessary for various reasons, including monitoring the environment and assessing agriculture. The progress in thisfield has been greatly enhanced by the integration of advancements in machine learning, especially deep learning, and also accessibility of high-definition images obtained through remote sensing. CNNs are now necessary instruments for precise recognition and classificationof objects in aerial imaging. These models use convolution and pooling operations to capture local patterns and relationships, enhancing image classification accuracy. This review paper presents a comprehensive look at the most recent techniques for categorising remote sensing images. It emphasises the significance of accurate classification in the constantly evolving contemporary society and investigates new progress made possible by deep learning methods.The article explores problems in pinpointing unique and important areas in remote sensing images and illustrates how deep learning models tackle these problems through comprehendingrelationships among objects and other elements in the scene. The article assesses the advantagesand disadvantages of various techniques in relation to accuracy, computational efficiency, andscalability through a comparison between traditional methods and deep learning-based methods.

Review of Convolutional Neural Network Models and Image Classification

With the arrival of the era of big data and the improvement of computing power, deep learning has swept the world. Traditional image classification methods are difficult to deal with the huge image data, and cannot meet the requirements of people on the accuracy and speed of image classification, the image classification method based on convolutional neural network breaks through the bottleneck of the traditional image classification method, and becomes the mainstream algorithm of image classification, how to effectively use convolutional neural network to classify images has become a hot spot of research in the field of computer vision at home and abroad. In this paper, we review the research background, significance and current research status of convolutional neural network model and image classification, study two image classification methods based on ResNet and ShuffleNet, and provide a comprehensive review of the construction methods and characteristics of the two deep convolutional neural network model structures, and finally compare and analyse the performance of the two classification models.

Wavelet Transform-based Methods for Forensic Analysis of Digital Images

In recent years, Generative Adversarial Networks (GANs) have been utilized in many applications of our daily lives to create digital media that was previously impossible. This paper utilizes GANs in multimedia forensics, where the precision and accuracy of image classification are crucial. The objective of the study is to detect sophisticated manipulated images generated by advanced deep learning methods using classical transformation methods. The method combines the classical wavelet transform for feature extraction with a classifier model to distinguish between real and fake images. The proposed method extracts unique features that differentiate between fake and real images, which are then fed to several gradient-boosting-based classifiers. The proposed method was tested using the FaceForensics++ dataset, which contains low-resolution video sequences of faces. The proposed model achieves an accuracy of 95% in detecting manipulated videos compared with 87% accuracy of other classical techniques.

Enhancing Weighted Averaging for CNN Model Ensemble in Plant Diseases Image Classification

Deep learning, especially convolutional neural networks (CNN), has gained traction in the field of image classification. In the specific case of plant disease classification, improving the accuracy and reliability of image classification is paramount. This paper delves into the ensemble prediction technique using a weighted soft-voting method. Instead of assigning a generalized weight to each CNN model, our approach emphasizes giving weights to each label's prediction within every individual model. We employed three esteemed CNN architectures for our experiments: DenseNet201, InceptionV3, and Xception focusing on classifying various diseases affecting grapes. By harnessing transfer learning coupled with end-to-end fine-tuning, we achieved a streamlined and efficient training process. Notably, the f1-score for each grape disease class was used as a parameter for weight determination and as a metric for the final evaluation. In our study, the newly proposed method was tested across various datasets and ensemble scenarios, demonstrating its effectiveness by not only outperforming the conventional soft-voting and prevalent weighted soft-voting methods, which achieved best scores of 95.68% and 95.81% respectively, but also by achieving a remarkable accuracy of 96.56%. This method's efficacy is heightened when ensemble models exhibit distinct characteristics; the more varied the model characteristics, the more enhanced the ensemble results.

Characterizing land use/land cover change dynamics by an enhanced random forest machine learning model: a Google Earth Engine implementation

Land use and land cover (LULC) analysis is crucial for understanding societal development and assessing changes during the Anthropocene era. Conventional LULC mapping faces challenges in capturing changes under cloud cover and limited ground truth data. To enhance the accuracy and comprehensiveness of the descriptions of LULC changes, this investigation employed a combination of advanced techniques. Specifically, multitemporal 30 m resolution Landsat-8 satellite imagery was utilized, in addition to the cloud computing capabilities of the Google Earth Engine (GEE) platform. Additionally, the study incorporated the random forest (RF) algorithm. This study aimed to generate continuous LULC maps for 2014 and 2020 for the Shrirampur area of Maharashtra, India. A novel multiple composite RF approach based on LULC classification was utilized to generate the final LULC classification maps utilizing the RF-50 and RF-100 tree models. Both RF models utilized seven input bands (B1 to B7) as the dataset for LULC classification. By incorporating these bands, the models were able to influence the spectral information captured by each band to classify the LULC categories accurately. The inclusion of multiple bands enhanced the discrimination capabilities of the classifiers, increasing the comprehensiveness of the assessment of the LULC classes. The analysis indicated that RF-100 exhibited higher training and validation/testing accuracy for 2014 and 2020 (0.99 and 0.79/0.80, respectively). The study further revealed that agricultural land, built-up land, and water bodies have changed adequately and have undergone substantial variation among the LULC classes in the study area. Overall, this research provides novel insights into the application of machine learning (ML) models for LULC mapping and emphasizes the importance of selecting the optimal tree combination for enhancing the accuracy and reliability of LULC maps based on the GEE and different RF tree models. The present investigation further enabled the interpretation of pixel-level LULC interactions while improving image classification accuracy and suggested the best models for the classification of LULC maps through the identification of changes in LULC classes.

DenseHillNet: a lightweight CNN for accurate classification of natural images.

The detection of natural images, such as glaciers and mountains, holds practical applications in transportation automation and outdoor activities. Convolutional neural networks (CNNs) have been widely employed for image recognition and classification tasks. While previous studies have focused on fruits, land sliding, and medical images, there is a need for further research on the detection of natural images, particularly glaciers and mountains. To address the limitations of traditional CNNs, such as vanishing gradients and the need for many layers, the proposed work introduces a novel model called DenseHillNet. The model utilizes a DenseHillNet architecture, a type of CNN with densely connected layers, to accurately classify images as glaciers or mountains. The model contributes to the development of automation technologies in transportation and outdoor activities. The dataset used in this study comprises 3,096 images of each of the "glacier" and "mountain" categories. Rigorous methodology was employed for dataset preparation and model training, ensuring the validity of the results. A comparison with a previous work revealed that the proposed DenseHillNet model, trained on both glacier and mountain images, achieved higher accuracy (86%) compared to a CNN model that only utilized glacier images (72%). Researchers and graduate students are the audience of our article.

A Multi-Hyperspectral Image Collaborative Mapping Model Based on Adaptive Learning for Fine Classification

Hyperspectral (HS) data, encompassing hundreds of spectral channels for the same area, offer a wealth of spectral information and are increasingly utilized across various fields. However, their limitations in spatial resolution and imaging width pose challenges for precise recognition and fine classification in large scenes. Conversely, multispectral (MS) data excel in providing spatial details for vast landscapes but lack spectral precision. In this article, we proposed an adaptive learning-based mapping model, including an image fusion module, spectral super-resolution network, and adaptive learning network. Spectral super-resolution networks learn the mapping between multispectral and hyperspectral images based on the attention mechanism. The image fusion module leverages spatial and spectral consistency in training data, providing pseudo labels for spectral super-resolution training. And the adaptive learning network incorporates spectral response priors via unsupervised learning, adjusting the output of the super-resolution network to preserve spectral information in reconstructed data. Through the experiment, the model eliminates the need for the manual setting of image prior information and complex parameter selection, and can adjust the network structure and parameters dynamically, eventually enhancing the reconstructed image quality, and enabling the fine classification of large-scale scenes with high spatial resolution. Compared with the recent dictionary learning and deep learning spectral super-resolution methods, our approach exhibits superior performance in terms of both image similarity and classification accuracy.

A new piracy-resistant DNN watermarking method based on secret key and block-wise image transformations

This paper proposes a watermarking method that can be used for the copyright protection of DNN models, utilizing learnable block-wise image transformation techniques and a secret key to embed a watermark. A black-box watermarking approach is used, which does not require a specific predefined training or trigger set, allowing for the remote verification of model ownership. As a result, this method can achieve copyright protection using authentication methods for DNN models. Results of experiments on established datasets [1, 2] indicate that the original watermark is not easily overwritten by pirated watermarks. Moreover, its performance in pruning attack experiments is similar to that observed in the studies cited above. However, our approach demonstrates stronger robustness against fine-tuning attacks, while also achieving higher image classification accuracy.

Ensemble Transfer Learning for Hand-sign Digit Image Classification

Hand sign classification is a challenging task in image processing and machine learning. Robust learning algorithms are essential to achieve optimal performance. Ensemble transfer learning, a technique that combines ensemble learning and transfer learning, is a promising approach to improve classification model performance. This study investigates the use of ensemble transfer learning for hand sign classification. The Sign Language Digits Dataset, which contains ten distinct handwritten image types, was used to evaluate the performance of three architectures: ResNet-50, VGG-19, and Ensemble Transfer Learning (a fusion of ResNet-50 and VGG-19). The results showed that all three architectures performed well, but Ensemble Transfer Learning achieved the best performance with an accuracy of 96.12%, precision of 96.06%, recall of 96.21%, and F1 score of 96.07%. This suggests that ensemble transfer learning can effectively enhance model performance in image hand sign classification. The study also found that combining ResNet-50 and VGG-19 in Ensemble Transfer Learning yielded superior results compared to individual models. This is because ensemble transfer learning can leverage the strengths of both models to improve the overall performance. The findings of this study highlight the significance of employing ensemble transfer learning techniques to enhance accuracy and reliability in hand sign image classification.