Accuracy Of Convolutional Neural Network Research Articles

Machine learning-based prediction of soil organic matter via smartphone

Color is an important property of soil that indicates soil composition and fertility. Soil organic matter (SOM) of darker color soils, which have rich humus and minerals, is higher than others such as red soils. The aim of this study is to construct models to predict SOM for a range of colors using a smartphone as an image-capturing device. Random forest of classification (RFC), random forest of logical regression (RFLR), convolutional neural network (CNN) and MobileNet models are compared, which is better for SOM prediction. Soil photos were collected by smartphone in a camera obscura with a steady light. After treatment by OpenCV, photos were separated by SOM content into five groups. The prediction accuracies of RFC, RFLR, CNN and MobileNet were 0.9743, 0.5614, 0.9600 and 0.7915, respectively. Based on these results, the RFC model for SOM has the best performance both in training and validation. The proposed combination of smartphone and machine learning-based prediction models provide a fast, economic, and robust approach to monitor, detect, and predict SOM contents in precision and intelligence agriculture.

Predicting malignancy in breast lesions: enhancing accuracy with fine-tuned convolutional neural network models.

This study aims to explore the accuracy of Convolutional Neural Network (CNN) models in predicting malignancy in Dynamic Contrast-Enhanced Breast Magnetic Resonance Imaging (DCE-BMRI). A total of 273 benign lesions (benign group) and 274 malignant lesions (malignant group) were collected and randomly divided into a training set (246 benign and 245 malignant lesions) and a testing set (28 benign and 28 malignant lesions) in a 9:1 ratio. An additional 53 lesions from 53 patients were designated as the validation set. Five models-VGG16, VGG19, DenseNet201, ResNet50, and MobileNetV2-were evaluated. Model performance was assessed using accuracy (Ac) in the training and testing sets, and precision (Pr), recall (Rc), F1 score (F1), and area under the receiver operating characteristic curve (AUC) in the validation set. The accuracy of VGG19 on the test set (0.96) is higher than that of VGG16 (0.91), DenseNet201 (0.91), ResNet50 (0.67), and MobileNetV2 (0.88). For the validation set, VGG19 achieved higher performance metrics (Pr 0.75, Rc 0.76, F1 0.73, AUC 0.76) compared to the other models, specifically VGG16 (Pr 0.73, Rc 0.75, F1 0.70, AUC 0.73), DenseNet201 (Pr 0.71, Rc 0.74, F1 0.69, AUC 0.71), ResNet50 (Pr 0.65, Rc 0.68, F1 0.60, AUC 0.65), and MobileNetV2 (Pr 0.73, Rc 0.75, F1 0.71, AUC 0.73). S4 model achieved higher performance metrics (Pr 0.89, Rc 0.88, F1 0.87, AUC 0.89) compared to the other four fine-tuned models, specifically S1 (Pr 0.75, Rc 0.76, F1 0.74, AUC 0.75), S2 (Pr 0.77, Rc 0.79, F1 0.75, AUC 0.77), S3 (Pr 0.76, Rc 0.76, F1 0.73, AUC 0.75), and S5 (Pr 0.77, Rc 0.79, F1 0.75, AUC 0.77). Additionally, S4 model showed the lowest loss value in the testing set. Notably, the AUC of S4 for BI-RADS 3 was 0.90 and for BI-RADS 4 was 0.86, both significantly higher than the 0.65 AUC for BI-RADS 5. The S4 model we propose has demonstrated superior performance in predicting the likelihood of malignancy in DCE-BMRI, making it a promising candidate for clinical application in patients with breast diseases. However, further validation is essential, highlighting the need for additional data to confirm its efficacy.

Application of convolutional neural networks in image classification and applications of improved convolutional neural networks

Abstract. This paper reviews the application and improvement of convolutional neural networks (CNNs) in image classification. Firstly, a shallow CNN for interstitial lung disease image classification is presented. This model suppresses overfitting through a unique network architecture and optimisation algorithm. Next, the improved VGG16 architecture and MIDNet18 model are discussed and their superior performance in brain tumour image classification is demonstrated. Subsequently, a CNN-CapsNet model for cervical cancer image classification and its improvement are presented and the customised model is compared with the conventional VGG-16 CNN architecture in the paper. Next, the application of sparse convolutional kernels and hybrid sparse convolutional kernels (HDCs) in solving the problem of computational resource consumption is presented. Subsequently, methods for solving the problem of limited training data through transfer learning and network data augmentation techniques are discussed, as well as GAN-generated datasets for solving the overfitting problem. Finally, the effect of degraded images on the classification effectiveness of CNNs is explored. The results show that the improved CNN architecture and algorithms have significant effects in solving the problems of overfitting and computational resource consumption, and can significantly improve the accuracy and efficiency of image classification. And degraded images do adversely affect the accuracy of CNN for image classification.

Improvement The Accuracy of Convolutional Neural Network with Using Undersampling Method on Unbalanced Credit Card Dataset

In this study, we address the challenge of imbalanced data in credit card fraud detection by proposing a novel approach that leverages Convolutional Neural Networks (CNNs) and undersampling techniques. The imbalance in the dataset, typical of real-world financial transactions, often leads to biased models favoring the majority class. To mitigate this, we employ undersampling to balance the classes, thereby enhancing the CNN's ability to learn from minority instances crucial for fraud detection. Our method is validated on a large unbalanced credit card dataset, demonstrating significant improvements in accuracy compared to traditional CNN models trained on imbalanced data. We evaluate our approach using standard performance metrics, including precision, recall, and F1-score, showcasing its effectiveness in accurately identifying fraudulent transactions while minimizing false positives. Furthermore, we pro-vide insights into the CNN's decision-making process through visualization techniques, shedding light on its ability to discern fraudulent patterns within the data. Our findings highlight the importance of addressing class imbalance in fraud detection tasks and underscore the efficacy of undersampling in enhancing the performance of deep learning models, particularly CNNs, in handling imbalanced datasets.

Prediction Model of Spinal Osteoporosis Using Lumbar Spine X-Ray from Transfer Learning Deep Convolutional Neural Networks

Objective: Osteoporosis is highly prevalent among older adults and women. This condition leads to a deterioration in bone mineral density and microarchitecture, significantly increasing the risk of fractures. Additionally, osteoporosis commonly results in complications such as screw loosening and non-union during spinal surgery. Deep-learning algorithms have now achieved an accuracy comparable to the current human margin of error. Therefore, this study explored the potential of using transfer learning in deep learning algorithms to predict, diagnose, and screen for osteoporosis using commonly obtained sagittal spine X-rays from patients with spinal conditions.Methods: We retrospectively evaluated 2,300 consecutive patients who underwent dual energy X-ray absorptiometry (DXA) and lumbar sagittal plain X-ray exams between 2013 and 2021. The exclusion criteria included: (1) a gap of more than 1 year between the DXA and X-ray exams; (2) vertebrae that had undergone vertebroplasty; (3) lack of spine anterior-posterior DXA; and (4) images that were unassessable. Ultimately, 256 patients (images) were included in the study. Transfer learning was applied using convolutional neural network (CNN) techniques, specifically visual geometry group (VGG) 16, VGG 19, ResNet50, and Xception.Results: The most accurate CNN model in the training group was ResNet50, with an accuracy of 0.95. ResNet50 showed the best performance, with an accuracy of 0.82, precision of 0.80, recall of 0.86, and F1-score of 0.83. Additionally, its area under the curve (0.76) was higher than that other CNN models. The confusion matrix for ResNet50’s performance displayed the outcomes for images predicted as osteoporosis (n=12) among the test data osteoporosis images (n=14)Conclusion: Artificial intelligence (AI) technology employing deep learning techniques is significantly nearing human capabilities in the role of diagnostic assistance. The diagnosis of osteoporosis using bone mineral density is expected to evolve into a comprehensive diagnostic aid or decision-making tool with the integration of AI in the future.

Accelerating Neural Network Inference in Handwritten Digit Recognition — Comparative Study

In deep learning, one of the popular subclasses of multilayer perceptron is the Convolutional Neural Networks (CNNs), optimized for image and semantic segmentation tasks. With a large number of floating-point operations and relatively less data transfer in the training phase, CNNs are well suited to be handled by parallel architectures. The high accuracy of CNNs comes at the cost of consequential compute and memory demands. In this article, we present a comparative analysis of a pre-trained CNN framework for recognizing a handwritten digit on different processing platforms. The performance of the neural network as the function of the parallel processing platforms offers parametric insights and factors that influence the inference on state-of-the-art Graphics Processing Units (GPUs) and systolic arrays such as Eyeriss and Tensor Processing Unit (TPU). Through inference time analysis, we observed that the systolic arrays can outperform upto 58.7 times better than a Turing architecture powered GPU. We show that while the efficiency with which the available resources are utilized is higher in the existing GPUs (upto 32%) than the TPUs, the efficiency of application-specific systolic arrays can be on par with that of GPUs. We present the results obtained from three different customized systolic array-based platforms which can be adopted by the designers to decide the hardware optimization goal.

Modernizing agricultural workflows using artificial intelligence and deep learning: Enhancing plant species identification

Modernizing agricultural processes is imperative due to the labor-intensive nature that impedes crop production. Despite the steps of farm technologies, the manual tasks of identifying plants and weeds remain very time-consuming. The current paper asserts that Convolutional Neural Networks (CNNs), a type of Artificial Intelligence, can enhance plant species recognition efficiency. The paper focuses explicitly on how CNNs can be leveraged to improve the process of identifying different plant species, thereby promoting efficiency in plant species identification. With a dataset of images of 12 species of plants, we have come up with an entirely accurate CNN model for the classification of plant types. Implementing this AI-powered approach can reduce manual work, enhance decision-making processes, and enrich environmental sustainability in agriculture by rendering the workflow smoother. The present study has broad ramifications for agriculture, technology, and environmental science professionals seeking to embed AI solutions into conventional workflows. The proposed system is meant to show how AI can improve productivity, reduce labor costs, and allow environmentally sustainable agriculture.

Automatic Classification of Defective Photovoltaic Module Cells Based on a Novel CNN-PCA-SVM Deep Hybrid Model in Electroluminescence Images

In today’s world, the rapid development of photovoltaic (PV) power plants has facilitated sustainable energy production. Maintenance and fault detection play a crucial role in ensuring the continuity of energy production. Manual inspection of electroluminescence (EL) images of PV modules requires significant human resources and time investment. This study presents a method for automatic fault detection of PV cells in EL images using hybrid deep features optimized with the principal component analysis (PCA) feature selection algorithm. A lightweight and high-performance model that combines the strengths of convolutional neural network (CNN) architectures is proposed. Firstly, data augmentation techniques are employed due to the imbalance between defective and functional classes in the dataset containing EL images. In experimental studies conducted by integrating the PCA algorithm into MobileNetV2, DenseNet201, and InceptionV3 CNN models, accuracy, precision, recall, and F1-score values of 92.19%, 92%, 90% and 91%, respectively, were achieved. When the results were analyzed, it was observed that the proposed method had an effective performance in detecting faults in PV panel cells.

Creating realistic anterior segment optical coherence tomography images using generative adversarial networks

AimsTo develop a generative adversarial network (GAN) capable of generating realistic high-resolution anterior segment optical coherence tomography (AS-OCT) images.MethodsThis study included 142 628 AS-OCT B-scans from the American University of...

Automatic sucker rod pump fault diagnostics by transfer learning using googlenet integrated machine learning classifiers

Oil and gas extraction is vital for meeting the energy needs of a growing global population. Artificial lift (AL) systems play a crucial role in oilfields, especially when reservoir pressure is low. Among these systems, sucker rod pumps (SRPs) are extensively employed for onshore hydrocarbon recovery operations. However, SRPs are susceptible to mechanical failures and operational challenges arising from evolving reservoir conditions. This study proposes an approach for fault diagnosis in SRPs using real-time data from sensors and machine learning (ML) algorithms to detect anomalies and patterns associated with potential problems. The proposed method involves deep learning (DL) for automated feature extraction from dynamometer cards and error-correcting output codes (ECOC) model-based supervised machine learners for efficiently assessing the operational state of the SRP. By forecasting potential failures, preemptive measures can be taken to minimize downtime and reduce maintenance costs. The application of ML in the analysis of SRP provides a potential tool for diagnosing and predicting failures, improving productivity efficiency, and reducing downtime. Also, the accuracy of Convolutional Neural Networks (CNN) models was compared with that of CNN-ECOC model-based learners in this study, and the results demonstrated that integrating algorithms such as Support Vector Machine (SVM) and k-Nearest Neighbor (k-NN) with the network caused significant improvements in the accuracy; however, Decision Tree (DT) and Naive Bayes (NB) did not perform well in the pattern classification task. Specifically, the GoogLeNet model achieved improved accuracy of 96.40 % and 99.40 % when SVM and k-NN were integrated, respectively. Conversely, with DT and NB, the accuracies dropped to 66.30 % and 75.10 %, respectively.

Ultrasonic signal classification for composite materials via deep convolutional neural networks

ABSTRACT Ultrasonic testing (UT) is the most commonly used non-destructive testing method in the aerospace field. However, for the detection of delamination defect in multi-layer composite adhesive materials with metal and non-metal bonding, traditional energy statistics methods are limited and cannot achieve high-precision automatic signal classification. This article proposes a signal classification model based on hybrid neural networks. By studying the classification accuracy of convolutional neural networks (CNN) and long short-term memory networks (LSTM) for different types of signals, a relatively balanced classification is achieved, which effectively improves the classification accuracy. By integrating attention mechanisms, the ability of the detection model to identify key features is further enhanced. Experimental results demonstrate that the proposed model achieves an accuracy of 98.57%. Bayesian Optimisation (BO) can effectively and automatically select the optimal hyperparameters, and achieve global optimisation. In the experiment, the accuracy increased by 0.73% compared to the benchmark value. Comparative experiments show that the signal classification model established in this paper has good performance.

Applying convolutional neural network and Nadam optimization in flower classification

Flowers have a variety of shapes, colors and structures, the images of which need to be classified using guided learning techniques. Several studies classify flowers using machine learning, but their accuracy performance is not good. The thing is, the flowers come in a variety of colors that can sometimes look similar to the background. Therefore, this study aims to classify flowers using a convolutional neural network (CNN) and measure its performance. The method used is mixed methods by collecting existing data from previous studies and connecting it with the realities in the field. The Kozłowski and Steinbrener models were used, while the image data was obtained from the Oxford17 and Oxford102 dataset with 17 and 102 flower types, respectively. The results show 60% and 84% accuracy of CNN using the scratch and transfer learning approach for the Oxford17 dataset. The Oxford102 dataset shows 42% and 64%, respectively, with CNN from baseline and transfer learning.

ERKT-Net: Implementing Efficient and Robust Knowledge Distillation for Remote Sensing Image Classification

The classification of Remote Sensing Images (RSIs) poses a significant challenge due to the presence of clustered ground objects and noisy backgrounds. While many approaches rely on scaling models to enhance accuracy, the deployment of RSI classifiers often requires substantial computational and storage resources, thus necessitating the use of lightweight algorithms. In this paper, we present an efficient and robust knowledge transfer network named ERKT-Net, which is designed to provide a lightweight yet accurate Convolutional Neural Network (CNN) classifier. This method utilizes innovative yet simple concepts to better accommodate the inherent nature of RSIs, thereby significantly improving the efficiency and robustness of traditional Knowledge Distillation (KD) techniques developed on ImageNet-1K. We evaluated ERKT-Net on three benchmark RSI datasets and found that it demonstrated superior accuracy and a very compact volume compared to 40 other advanced methods published between 2020 and 2023. On the most challenging NWPU45 dataset, ERKT-Net outperformed other KD-based methods with a maximum Overall Accuracy (OA) value of 22.4%. Using the same criterion, it also surpassed the first-ranked multi-model method with a minimum OA value of 0.7 but presented at least an 82% reduction in parameters. Furthermore, ablation experiments indicated that our training approach has significantly improved the efficiency and robustness of classic DA techniques. Notably, it can reduce the time expenditure in the distillation phase by at least 80%, with a slight sacrifice in accuracy. This study confirmed that a logit-based KD technique can be more efficient and effective in developing lightweight yet accurate classifiers, especially when the method is tailored to the inherent characteristics of RSIs.

Auxiliary Diagnosis of Dental Calculus Based on Deep Learning and Image Enhancement by Bitewing Radiographs.

In the field of dentistry, the presence of dental calculus is a commonly encountered issue. If not addressed promptly, it has the potential to lead to gum inflammation and eventual tooth loss. Bitewing (BW) images play a crucial role by providing a comprehensive visual representation of the tooth structure, allowing dentists to examine hard-to-reach areas with precision during clinical assessments. This visual aid significantly aids in the early detection of calculus, facilitating timely interventions and improving overall outcomes for patients. This study introduces a system designed for the detection of dental calculus in BW images, leveraging the power of YOLOv8 to identify individual teeth accurately. This system boasts an impressive precision rate of 97.48%, a recall (sensitivity) of 96.81%, and a specificity rate of 98.25%. Furthermore, this study introduces a novel approach to enhancing interdental edges through an advanced image-enhancement algorithm. This algorithm combines the use of a median filter and bilateral filter to refine the accuracy of convolutional neural networks in classifying dental calculus. Before image enhancement, the accuracy achieved using GoogLeNet stands at 75.00%, which significantly improves to 96.11% post-enhancement. These results hold the potential for streamlining dental consultations, enhancing the overall efficiency of dental services.

Research on FBG Tactile Sensing Shape Recognition Based on Convolutional Neural Network.

Shape recognition plays a significant role in the field of robot perception. In view of the low efficiency and few types of shape recognition of the fiber tactile sensor applied to flexible skin, a convolutional-neural-network-based FBG tactile sensing array shape recognition method was proposed. Firstly, a sensing array was fabricated using flexible resin and 3D printing technology. Secondly, a shape recognition system based on the tactile sensing array was constructed to collect shape data. Finally, shape classification recognition was performed using convolutional neural network, random forest, support vector machine, and k-nearest neighbor. The results indicate that the tactile sensing array exhibits good sensitivity and perception capability. The shape recognition accuracy of convolutional neural network is 96.58%, which is 6.11%, 9.44%, and 12.01% higher than that of random forest, k-nearest neighbor, and support vector machine. Its F1 is 96.95%, which is 6.3%, 8.73%, and 11.94% higher than random forest, k-nearest neighbor, and support vector machine. The research of FBG shape sensing array based on convolutional neural network provides an experimental basis for shape perception of flexible tactile sensing.

ASO Visual Abstract: Assessing Endoscopic Response in Locally Advanced Rectal Cancer Treated with Total Neoadjuvant Therapy: Development and Validation of a Highly Accurate Convolutional Neural Network.

ASO Visual Abstract: Assessing Endoscopic Response in Locally Advanced Rectal Cancer Treated with Total Neoadjuvant Therapy: Development and Validation of a Highly Accurate Convolutional Neural Network.

Using convolutional neural networks to automatically score eight TIMSS 2019 graphical response items

International large-scale assessments (ILSAs) have used graphical response-based items to measure student ability for decades, but they have yet to implement automated scoring of these responses and instead rely on human scoring alone. To investigate how scores provided by machine algorithms compare to those provided by human raters, we applied convolutional neural networks (CNNs) to classify image-based responses from eight Timss 2019 items. Our results show that the most accurate CNN models classified over 99% of the image responses into the appropriate scoring category for dichotomous items and almost 98% for one trichotomous item. Additionally, during the modeling process, the CNNs correctly classified numerous image responses that human raters had scored incorrectly. For most items, the number of incorrectly human-scored responses exceeded the average number of responses misclassified by the most accurate models. These results suggest that automated scoring using CNNs is comparable to, and in many cases more accurate, than human raters, even across a wide variety of graphing tasks. This paper argues that the machine learning procedure explored could be implemented in ILSAs as a verification method to improve the accuracy and consistency of graphical response item scores. In lieu of additional human raters, ILSAs could implement CNN-based automated scoring to provide a second set of scores, thus reducing the workload and costs associated with human scoring.

Deep learned features selection algorithm: Removal operation of anomaly feature maps (RO-AFM)

Class/Regression Activation Maps (CAMs/RAMs; AMs) are often embedded into Convolutional Neural Networks (CNNs) for checking activated regions on input images at estimation. CNNs sometime generate unreliable AMs, such as activated regions, are inappropriate. Because AM is calculated by stacking many feature maps generated by the final convolutional layer, when there are Anomaly Feature Maps (AFMs), unreliable AMs can be generated. For example, suppose we have a CNN that evaluates the heart. In this case, the feature maps that focus on regions unrelated to the heart (e.g., shoulders and esophagus) are AFMs. Additionally, we have a hypothesis that the estimation accuracy of CNNs is increased by removing AFMs. However, methods for automatically detecting and removing AFMs have not been sufficiently studied in previous research to improve the performance of CNNs. Therefore, we propose a method named “Removal Operation of Anomaly Feature Maps (RO-AFMs)” to automatically detect and remove AFMs. When applying an RO-AFM to the Global Average Pooling (GAP) feature vectors of a CNN, dimensions of the GAP vector are reduced. Therefore, an RO-AFM is regarded as a deep-feature selection algorithm. From the results of adopting an RO-AFM to a Regression CNN (R-CNN) for estimating pulmonary artery wedge pressure, which is one of the measurement score for representing cardiac anomaly state, improved reliability of AM and estimation accuracy were verified. A comparison of RO-AFM and the existing methods, i.e., Lasso and the Feature Selection Layer (FSL), indicated that RO-AFM performed slightly better on the estimation accuracy. The computation time required for RO-AFM to evaluate all features was 1.833 s on average, confirming that RO-AFM is a lightweight process. Therefore, RO-AFM is useful for constructing a medical CNN that emphasizes explainability (e.g., CNNs for estimating the risk of a disease or a test value from chest X-ray or computed tomography images).

DMAF-NET: Deep Multi-Scale Attention Fusion Network for Hyperspectral Image Classification with Limited Samples

In recent years, deep learning methods have achieved remarkable success in hyperspectral image classification (HSIC), and the utilization of convolutional neural networks (CNNs) has proven to be highly effective. However, there are still several critical issues that need to be addressed in the HSIC task, such as the lack of labeled training samples, which constrains the classification accuracy and generalization ability of CNNs. To address this problem, a deep multi-scale attention fusion network (DMAF-NET) is proposed in this paper. This network is based on multi-scale features and fully exploits the deep features of samples from multiple levels and different perspectives with an aim to enhance HSIC results using limited samples. The innovation of this article is mainly reflected in three aspects: Firstly, a novel baseline network for multi-scale feature extraction is designed with a pyramid structure and densely connected 3D octave convolutional network enabling the extraction of deep-level information from features at different granularities. Secondly, a multi-scale spatial–spectral attention module and a pyramidal multi-scale channel attention module are designed, respectively. This allows modeling of the comprehensive dependencies of coordinates and directions, local and global, in four dimensions. Finally, a multi-attention fusion module is designed to effectively combine feature mappings extracted from multiple branches. Extensive experiments on four popular datasets demonstrate that the proposed method can achieve high classification accuracy even with fewer labeled samples.

Enhancing Brain Tumor Classification: A CNN-Based Approach with InceptionV3 and Xception

Brain tumors are among the most aggressive and deadly diseases, with a very short life expectancy at the highest grade. To combat this, early detection and treatment are crucial.In this approach, MRI images are used to analyze brain abnormalities. The manual investigation of brain tumor classification is a time-consuming task, and there might be possibilities of human errors. Hence, accurate analysis in a short span of time is essential. This approach presents the automatic brain tumor classification algorithm using a highly accurate Convolutional Neural Network (CNN), InceptionV3 and Xception algorithm for classification of Glioma, Meningioma and No tumor. The brain part is initially segmented by a thresholding approach followed by a morphological operation. The brain MRI is classified using CNN, Inceptionv3, and Xceptionv3 algorithms. The system's performance is evaluated using precision, recall, F1 score, and accuracy parameters.