Image Classification Performance Research Articles

The Detection of COVID-19 in Chest X-rays Using Ensemble CNN Techniques

Advances in the field of image classification using convolutional neural networks (CNNs) have greatly improved the accuracy of medical image diagnosis by radiologists. Numerous research groups have applied CNN methods to diagnose respiratory illnesses from chest X-rays and have extended this work to prove the feasibility of rapidly diagnosing COVID-19 with high degrees of accuracy. One issue in previous research has been the use of datasets containing only a few hundred images of chest X-rays containing COVID-19, causing CNNs to overfit the image data. This leads to lower accuracy when the model attempts to classify new images, as would be clinically expected. In this work, we present a model trained on the COVID-QU-Ex dataset containing 33,920 chest X-ray images, with an equal share of COVID-19, Non-COVID pneumonia, and Normal images. The model is an ensemble of pre-trained CNNs (ResNet50, VGG19, and VGG16) and GLCM textural features. The model achieved a 98.34% binary classification accuracy (COVID-19/no COVID-19) on a test dataset of 6581 chest X-rays and 94.68% for distinguishing between COVID-19, Non-COVID pneumonia, and normal chest X-rays. The results also demonstrate that a higher 98.82% three-class test accuracy can be achieved using the model if the training dataset only contains a few thousand images. However, the generalizability of the model suffers due to the smaller dataset size. This study highlights the benefits of both ensemble CNN techniques and larger dataset sizes for medical image classification performance.

ESPT: A Self-Supervised Episodic Spatial Pretext Task for Improving Few-Shot Learning

Self-supervised learning (SSL) techniques have recently been integrated into the few-shot learning (FSL) framework and have shown promising results in improving the few-shot image classification performance. However, existing SSL approaches used in FSL typically seek the supervision signals from the global embedding of every single image. Therefore, during the episodic training of FSL, these methods cannot capture and fully utilize the local visual information in image samples and the data structure information of the whole episode, which are beneficial to FSL. To this end, we propose to augment the few-shot learning objective with a novel self-supervised Episodic Spatial Pretext Task (ESPT). Specifically, for each few-shot episode, we generate its corresponding transformed episode by applying a random geometric transformation to all the images in it. Based on these, our ESPT objective is defined as maximizing the local spatial relationship consistency between the original episode and the transformed one. With this definition, the ESPT-augmented FSL objective promotes learning more transferable feature representations that capture the local spatial features of different images and their inter-relational structural information in each input episode, thus enabling the model to generalize better to new categories with only a few samples. Extensive experiments indicate that our ESPT method achieves new state-of-the-art performance for few-shot image classification on three mainstay benchmark datasets. The source code will be available at: https://github.com/Whut-YiRong/ESPT.

P-075 Artificial intelligence for sperm analysis: A KNIME-Based automated analysis of sperm morphology

Abstract Study question To evaluate an automated sperm morphology assessment method using the KNIME Analytics Platform compared to manual sperm morphology analysis performed as per WHO guidelines. Summary answer Our method emphasizes how artificial intelligence technologies have the potential to foster standardization of sperm morphology assessment with comparable precision and reliability. What is known already Manual sperm morphology assessment is considered the most difficult parameter to standardize due to its subjective nature, strongly linked to the operator's level of expertise. Indeed, there is a high degree of inter and intra-laboratory variability. Manual examination is time-consuming and laborious. There were many attempts to automate sperm morphology analysis, especially with CASA (Computer Assisted Sperm Analysis) systems, but their performance is still disputable. One of the difficulties in this field of study is the lack of publicly available datasets. Besides, the available databases are only focused on the sperm head morphology. Study design, size, duration A total of 37 semen samples from men attending our laboratory for infertility investigation were included, over a period of one year. For each sample, semen smears were fixed and stained by the Spermoscan® kit. Participants/materials, setting, methods A total of 1000 images of individual spermatozoa were obtained using the MMC® CASA system. The number of images per sample depended on its quality. Three experts have classified these spermatozoa according to modified David classification for sperm morphology. The results were then processed and an algorithm created using teh KNIME Analytics Platform, trained and tested to classify spermatozoa. This workflow uses CNN (convolutional neural network) to perform image classification on our dataset. Main results and the role of chance Of the 1000 images analyzed, we counted : 116 Normal sperm morphology, 67 abnormal post-acrosomal region, 128 abnormal acrosomes, 8 elongated heads, 6 thin heads,10 microcephalic,7 multiple heads, 27 coiled tails, 7 cytoplasmic residues, 17 angulated tails, 6 short tails, 4 multiple tails, and 697 associated abnormalities. For each image, a notebook file containing sperm abnormalities as assessed by the three experts was created, in addition to sperm head, mid-piece and tail dimensions obtained by the CASA system. Our dataset was randomly partitioned into 2 groups: 80% of data formed the training set, and The remaining 20% formed the fully independent test set. The best performance with the KNIME-Based algorithm was achieved for post-acrosomal abnormalities (97 % True positive rate), and the worst for multiple tail abnormalities (69 % True positive rate). Our machine learning model classifies sperm morphology at high accuracy (99.5 %). The overall process occurs in less than 10 seconds. Limitations, reasons for caution Our database included various head, midpiece and tail anomalies. However, the number of images in each category is unequal due to the limited occurrence of some morphological abnormalities. Thus, it is important to increase the number of images in these categories to obtain better results. Wider implications of the findings Our goal is to expand our modified David’s classification-based database. We aim to improve the performance of our model, and to put it at the service of learning and routine use in laboratories. Trial registration number not applicable

A Multi-Stage Faster RCNN-Based iSPLInception for Skin Disease Classification Using Novel Optimization.

Nowadays, skin cancer is considered a serious disorder in which early identification and treatment of the disease are essential to ensure the stability of the patients. Several existing skin cancer detection methods are introduced by employing deep learning (DL) to perform skin disease classification. Convolutional neural networks (CNNs) can classify melanoma skin cancer images. But, it suffers from an overfitting problem. Therefore, to overcome this problem and to classify both benign and malignant tumors efficiently, the multi-stage faster RCNN-based iSPLInception (MFRCNN-iSPLI) method is proposed. Then, the test dataset is used for evaluating the proposed model performance. The faster RCNN is employed directly to perform image classification. This may heavily raise computation time and network complications. So, the iSPLInception model is applied in the multi-stage classification. In this, the iSPLInception model is formulated using the Inception-ResNet design. For candidate box deletion, the prairie dog optimization algorithm is utilized. We have utilized two skin disease datasets, namely, ISIC 2019 Skin lesion image classification and the HAM10000 dataset for conducting experimental results. The methods' accuracy, precision, recall, and F1 score values are calculated, and the results are compared with the existing methods such as CNN, hybrid DL, Inception v3, and VGG19. With 95.82% accuracy, 96.85% precision, 96.52% recall, and 0.95% F1 score values, the output analysis of each measure verified the prediction and classification effectiveness of the method.

Training energy-based single-layer Hopfield and oscillatory networks with unsupervised and supervised algorithms for image classification

This paper investigates how to solve image classification with Hopfield neural networks (HNNs) and oscillatory neural networks (ONNs). This is a first attempt to apply ONNs for image classification. State-of-the-art image classification networks are multi-layer models trained with supervised gradient back-propagation, which provide high-fidelity results but require high energy consumption and computational resources to be implemented. On the contrary, HNN and ONN networks are single-layer, requiring less computational resources, however, they necessitate some adaptation as they are not directly applicable for image classification. ONN is a novel brain-inspired computing paradigm that performs low-power computation and is attractive for edge artificial intelligence applications, such as image classification. In this paper, we perform image classification with HNN and ONN by exploiting their auto-associative memory (AAM) properties. We evaluate precision of HNN and ONN trained with state-of-the-art unsupervised learning algorithms. Additionally, we adapt the supervised equilibrium propagation (EP) algorithm to single-layer AAM architectures, proposing the AAM-EP. We test and validate HNN and ONN classification on images of handwritten digits using a simplified MNIST set. We find that using unsupervised learning, HNN reaches 65.2%, and ONN 59.1% precision. Moreover, we show that AAM-EP can increase HNN and ONN precision up to 67.04% for HNN and 62.6% for ONN. While intrinsically HNN and ONN are not meant for classification tasks, to the best of our knowledge, these are the best-reported precisions of HNN and ONN performing classification of images of handwritten digits.

A federated learning algorithm using parallel-ensemble method on non-IID datasets

Traditional federated learning algorithms suffer from considerable performance reduction with non-identically and independently distributed datasets. This paper proposes a federated learning algorithm based on parallel-ensemble learning, which improves performance for image classification on these datasets. The training process of this algorithm includes basic federation learning and meta federation learning. First, several basic models are trained through cross-validation of federated learning, and then the meta-model is trained using the prediction results of the validation sets. In the training process, the training of different basic models is parallel. In prediction, meta-model is used to aggregate the output of the basic models to get the final prediction results. Our algorithm can achieve higher accuracy than traditional federated learning when using non-independent identically distributed datasets for image classification. Our algorithm aggregates different models through federated learning based on parallel-ensemble method, and improves the image classification performance of federated learning on non-independent identically distributed datasets.

Fire Detection with Image Processing

Abstract: Convolutional neural networks (CNNs) have yielded state-of-theart performance in image classification and other computer vision tasks. Their application in fire detection systems will substantially improve detection accuracy, which will eventually minimize fire disasters and reduce the ecological and social ramifications. However, the major concern with CNNbased fire detection systems is their implementation in realworld surveillance networks, due to their high memory and computational requirements for inference. In this paper, we propose an original, energy-friendly, and computationally efficient CNN architecture, inspired by the SqueezeNet architecture for fire detection, localization, and semantic understanding of the scene of the fire. It uses smaller convolutional kernels and contains no dense, fully connected layers, which helps keep the computational requirements to a minimum. Despite its low computational needs, the experimental results demonstrate that our proposed solution achieves accuracies that are comparable to other, more complex models, mainly due to its increased depth. Moreover, this paper shows how a tradeoff can be reached between fire detection accuracy and efficiency, by considering the specific characteristics of the problem of interest and the variety of fire data

Dataset Bias Prediction for Few-Shot Image Classification

Dataset bias is a significant obstacle that negatively affects image classification performance, especially in few-shot learning, where datasets have limited samples per class. However, few studies have focused on this issue. To address this, we propose a bias prediction network that recovers biases such as color from the extracted features of image data, resulting in performance improvement in few-shot image classification. If the network can easily recover the bias, the extracted features may contain the bias. Therefore, the whole framework is trained to extract features that are difficult for the bias prediction network to recover. We evaluate our method by integrating it with several existing few-shot learning models across multiple benchmark datasets. The results show that the proposed network can improve the performance in different scenarios. The proposed approach effectively reduces the negative effect of the dataset bias, resulting in the performance improvements in few-shot image classification. The proposed bias prediction model is easily compatible with other few-shot learning models, and applicable to various real-world applications where biased samples are prevalent, such as VR/AR systems and computer vision applications.

Application of Principal Component Analysis to advancing digital phenotyping of plant disease in the context of limited memory for training data storage

Despite its widespread employment as a highly efficient dimensionality reduction technique, limited research has been carried out on the advantage of Principal Component Analysis (PCA)–based compression/reconstruction of image data to machine learning-based image classification performance and storage space optimization. To address this limitation, we designed a study in which we compared the performances of two Convolutional Neural Network-Random Forest Algorithm (CNN-RF) guava leaf image classification models developed using training data from a number of original guava leaf images contained in a predefined amount of storage space (on the one hand), and a number of PCA compressed/reconstructed guava leaf images contained in the same amount of storage space (on the other hand), on the basis of four criteria – Accuracy, F1-Score, Phi Coefficient and the Fowlkes–Mallows index. Our approach achieved a 1:100 image compression ratio (99.00% image compression) which was comparatively much better than previous results achieved using other algorithms like arithmetic coding (1:1.50), wavelet transform (90.00% image compression), and a combination of three transform-based techniques – Discrete Fourier (DFT), Discrete Wavelet (DWT) and Discrete Cosine (DCT) (1:22.50). From a subjective visual quality perspective, the PCA compressed/reconstructed guava leaf images presented almost no loss of image detail. Finally, the CNN-RF model developed using PCA compressed/reconstructed guava leaf images outperformed the CNN-RF model developed using original guava leaf images by 0.10% accuracy increase, 0.10 F1-Score increase, 0.18 Phi Coefficient increase and 0.09 Fowlkes–Mallows increase.

MLP-based classification of COVID-19 and skin diseases

Recent years have witnessed a growing interest in neural network-based medical image classification methods, which have demonstrated remarkable performance in this field. Typically, convolutional neural network (CNN) architectures have been commonly employed to extract local features. However, the transformer, a newly emerged architecture, has gained popularity due to its ability to explore the relevance of remote elements in an image through a self-attention mechanism. Despite this, it is crucial to establish not only local connectivity but also remote relationships between lesion features and capture the overall image structure to improve image classification accuracy. Therefore, to tackle the aforementioned issues, this paper proposes a network based on multilayer perceptrons (MLPs) that can learn the local features of medical images on the one hand and capture the overall feature information in both spatial and channel dimensions on the other hand, thus utilizing image features effectively. This paper has been extensively validated on COVID19-CT dataset and ISIC 2018 dataset, and the results show that the method in this paper is more competitive and has higher performance in medical image classification compared with existing methods. This shows that the use of MLP to capture image features and establish connections between lesions is expected to provide novel ideas for medical image classification tasks in the future.

Classification of Monkeypox Images Using LIME-Enabled Investigation of Deep Convolutional Neural Network.

In this research, we demonstrate a Deep Convolutional Neural Network-based classification model for the detection of monkeypox. Monkeypox can be difficult to diagnose clinically in its early stages since it resembles both chickenpox and measles in symptoms. The early diagnosis of monkeypox helps doctors cure it more quickly. Therefore, pre-trained models are frequently used in the diagnosis of monkeypox, because the manual analysis of a large number of images is labor-intensive and prone to inaccuracy. Therefore, finding the monkeypox virus requires an automated process. The large layer count of convolutional neural network (CNN) architectures enables them to successfully conceptualize the features on their own, thereby contributing to better performance in image classification. The scientific community has recently articulated significant attention in employing artificial intelligence (AI) to diagnose monkeypox from digital skin images due primarily to AI's success in COVID-19 identification. The VGG16, VGG19, ResNet50, ResNet101, DenseNet201, and AlexNet models were used in our proposed method to classify patients with monkeypox symptoms with other diseases of a similar kind (chickenpox, measles, and normal). The majority of images in our research are collected from publicly available datasets. This study suggests an adaptive k-means clustering image segmentation technique that delivers precise segmentation results with straightforward operation. Our preliminary computational findings reveal that the proposed model could accurately detect patients with monkeypox. The best overall accuracy achieved by ResNet101 is 94.25%, with an AUC of 98.59%. Additionally, we describe the categorization of our model utilizing feature extraction using Local Interpretable Model-Agnostic Explanations (LIME), which provides a more in-depth understanding of particular properties that distinguish the monkeypox virus.

A new weighted extreme learning machine based on elastic net regularization embedded exponential regularized discriminative dictionary learning for image classification

It is well known that discriminative sparse representation can significantly improve the performance of image classification. However, there remain several tricky issues to be addressed due to the unsatisfied performance and high time consumption. In this paper, a novel classification framework called weighted extreme learning machine exponential regularized discriminative dictionary learning (WELM-ERDDL) is proposed to address these issues. The main contributions of this paper include (1) the WELM is embedded with ERDDL via exponential regularized linear discriminative analysis (ERLDA) for feature mappings while enabling nonlinear and diverse feature representation; (2) in the ELM learning process, the elastic net regularization is utilized to optimize more robust and meaningful output weights; (3) an effective weight update rule is designed for WELM. To verify the effectiveness of the proposed method, several experiments are conducted on real-world image classification databases. The results show that the proposed WELM-ERDDL framework is even more efficient than other state-of-the-art algorithms in general.

Grad-CAM을 활용한 EfficientNetV2 기반 웨이퍼 맵 불량분석 연구

In semiconductors, various defect patterns appear on the wafer map due to problems in the design and manufacturing process. Analysis of generated defect patterns will reduce the rate of defects and enable the production of high-quality semiconductors. Considering the number of semiconductors produced, performing defect analysis by human resources is inefficient. Recently, as hardware performance has improved, high-performance deep learning models have been designed. These models show high performance in image classification and have advantages in terms of processing speed. Therefore, this paper used EfficientNetV2 designed to achieve maximum efficiency with few parameters for semiconductor failure analysis. Identifying the location of defects is a critical element of defect analysis, not just classifying defect patterns. Therefore, in this study, we used Grad-CAM to identify the classification of defect patterns and their approximate location. Wafer map dataset is difficult to collect as data includes defects in manufacturing companies

Komparasi Kinerja DenseNet 121 dan MobileNet untuk Klasifikasi Citra Penyakit Daun Kentang

Potato plants are one of the plants that are included in horticultural commodities that are widely cultivated by farmers. Potatoes are the part produced from this plant and are the fourth largest agricultural food crop in the world after corn, wheat and rice. Potato plants are susceptible to attacks by various diseases in the leaf area, resulting in delays in potato production. This disease can be recognized by farmers visually, because the infected leaves have a different color and texture from healthy or fresh leaves. However, it was found that detection using the naked eye by farmers required more processing time and often gave inappropriate results. Methods in the field of image processing can be applied, namely by using pattern recognition or characteristics from the image of diseased potato leaves. Through this technique it is hoped that it can detect diseases on potato leaves correctly and accurately. Based on this description, this study aims to design a Convolution Neural Network (CNN) model and evaluate the performance of two architectures, namely DenseNet 121 and MobileNet. From the results of research conducted by the author on potato leaf disease images, it shows that the MobileNet algorithm is better than the DenseNet 12 algorithm. The MobileNet algorithm produces an accuracy of 98.00%, therefore the MobileNet algorithm has better performance for image classification of potato leaf disease

Hybrid Facial Emotion Recognition Using CNN-Based Features

In computer vision, the convolutional neural network (CNN) is a very popular model used for emotion recognition. It has been successfully applied to detect various objects in digital images with remarkable accuracy. In this paper, we extracted learned features from a pre-trained CNN and evaluated different machine learning (ML) algorithms to perform classification. Our research looks at the impact of replacing the standard SoftMax classifier with other ML algorithms by applying them to the FC6, FC7, and FC8 layers of Deep Convolutional Neural Networks (DCNNs). Experiments were conducted on two well-known CNN architectures, AlexNet and VGG-16, using a dataset of masked facial expressions (MLF-W-FER dataset). The results of our experiments demonstrate that Support Vector Machine (SVM) and Ensemble classifiers outperform the SoftMax classifier on both AlexNet and VGG-16 architectures. These algorithms were able to achieve improved accuracy of between 7% and 9% on each layer, suggesting that replacing the classifier in each layer of a DCNN with SVM or ensemble classifiers can be an efficient method for enhancing image classification performance. Overall, our research demonstrates the potential for combining the strengths of CNNs and other machine learning (ML) algorithms to achieve better results in emotion recognition tasks. By extracting learned features from pre-trained CNNs and applying a variety of classifiers, we provide a framework for investigating alternative methods to improve the accuracy of image classification.

A Cross-Channel Dense Connection and Multi-Scale Dual Aggregated Attention Network for Hyperspectral Image Classification

Hyperspectral image classification (HSIC) is one of the most important research topics in the field of remote sensing. However, it is difficult to label hyperspectral data, which limits the improvement of classification performance of hyperspectral images in the case of small samples. To alleviate this problem, in this paper, a dual-branch network which combines cross-channel dense connection and multi-scale dual aggregated attention (CDC_MDAA) is proposed. On the spatial branch, a cross-channel dense connections (CDC) module is designed. The CDC can effectively combine cross-channel convolution with dense connections to extract the deep spatial features of HSIs. Then, a spatial multi-scale dual aggregated attention module (SPA_MDAA) is constructed. The SPA_MDAA adopts dual autocorrelation for attention modeling to strengthen the differences between features and enhance the ability to pay attention to important features. On the spectral branch, a spectral multi-scale dual aggregated attention module (SPE_MDAA) is designed to capture important spectral features. Finally, the spatial spectral features are fused, and the classification results are obtained. The experimental results show that the classification performance of the proposed method is superior to some state-of-the-art methods in small samples and has good generalization.

An image compression approach for efficient pneumonia recognition

ABSTRACT Increasingly, analytics such as classification and detection suffer from a significant amount of generated visual data. Nonetheless, recent approaches have not given substantial thought to CAD systems within limited capacities at the expense of performance. For that purpose, we proposed an autoencoder-based classification approach for pneumonia recognition, extending the use of the features extracted by autoencoders for compression to enhance efficiency. Thus, we substitute the classification of images with compressed sequences and encoded tensors, representing a more convenient format for managing and storing data. Which significantly minimizes computing costs and enhances transmission bandwidth. We designed CNN model introducing the attention mechanism to the latent space with minimum parameters optimizing the classification complexity. We assess our method's effectiveness on two medical imaging datasets. In addition, we compared latent space classification to multi-resolution image classification performance. Our approach improves state-of-art performance, boosting efficiency; the number of parameters is negligible, reduced by 69%.

Federated Learning for Medical Image Analysis with Deep Neural Networks.

Medical image analysis using deep neural networks (DNN) has demonstrated state-of-the-art performance in image classification and segmentation tasks, aiding disease diagnosis. The accuracy of the DNN is largely governed by the quality and quantity of the data used to train the model. However, for the medical images, the critical security and privacy concerns regarding sharing of local medical data across medical establishments precludes exploiting the full DNN potential for clinical diagnosis. The federated learning (FL) approach enables the use of local model's parameters to train a global model, while ensuring data privacy and security. In this paper, we review the federated learning applications in medical image analysis with DNNs, highlight the security concerns, cover some efforts to improve FL model performance, and describe the challenges and future research directions.

Dynamical memristive neural networks and associative self-learning architectures using biomimetic devices.

While there is an abundance of research on neural networks that are "inspired" by the brain, few mimic the critical temporal compute features that allow the brain to efficiently perform complex computations. Even fewer methods emulate the heterogeneity of learning produced by biological neurons. Memory devices, such as memristors, are also investigated for their potential to implement neuronal functions in electronic hardware. However, memristors in computing architectures typically operate as non-volatile memories, either as storage or as the weights in a multiply-and-accumulate function that requires direct access to manipulate memristance via a costly learning algorithm. Hence, the integration of memristors into architectures as time-dependent computational units is studied, starting with the development of a compact and versatile mathematical model that is capable of emulating flux-linkage controlled analog (FLCA) memristors and their unique temporal characteristics. The proposed model, which is validated against experimental FLCA LixNbO2 intercalation devices, is used to create memristive circuits that mimic neuronal behavior such as desensitization, paired-pulse facilitation, and spike-timing-dependent plasticity. The model is used to demonstrate building blocks of biomimetic learning via dynamical memristive circuits that implement biomimetic learning rules in a self-training neural network, with dynamical memristive weights that are capable of associative lifelong learning. Successful training of the dynamical memristive neural network to perform image classification of handwritten digits is shown, including lifelong learning by having the dynamical memristive network relearn different characters in succession. An analog computing architecture that learns to associate input-to-input correlations is also introduced, with examples demonstrating image classification and pattern recognition without convolution. The biomimetic functions shown in this paper result from fully ion-driven memristive circuits devoid of integrating capacitors and thus are instructive for exploiting the immense potential of memristive technology for neuromorphic computation in hardware and allowing a common architecture to be applied to a wide range of learning rules, including STDP, magnitude, frequency, and pulse shape among others, to enable an inorganic implementation of the complex heterogeneity of biological neural systems.

Lightweight Parallel Octave Convolutional Neural Network for Hyperspectral Image Classification

Although most deep learning-based methods have achieved excellent performance for hyperspectral image (HSI) classification, they are often limited by complex networks and require massive training samples in practical applications. Therefore, designing an efficient, lightweight model to obtain better classification results under small samples situations remains a challenging task. To alleviate this problem, a novel, lightweight parallel octave convolutional neural network (LPOCNN) for HSI classification is proposed in this paper. First, the HSI data is preprocessed to construct two three-dimensional (3D) patch cubes with different spatial and spectral scales for each central pixel, removing redundancy and focusing on extracting spatial features and spectral features, respectively. Next, two non- deep parallel branches are created for the two inputs, which design octave convolution rather than classical 3D convolution to facilitate light weighting of the model. Then two-dimensional convolutional neural network is used to extract deeper spectral-spatial features when fusing spectral-spatial features from different parallel layers. Moreover, the spectral-spatial attention is designed to promote the classification performance even further by adaptively adjusting the weights of different spectral-spatial features according to their contribution to classification. Experiments show that our suggested LPOCNN acquires a significant advantage on classification performance over other competitive methods under small sample situations.