Image Classification Research Articles

Geographical Area Classification on Satellite Images Using CNN Architecture

This paper presents a novel approach for classi- fying geographical areas in satellite imagery using a modified Convolutional Neural Network (CNN) architecture.The archi- tecture enhances feature extraction and classification accuracy by combining specialized layers like fully connected, pooling, and conventional layers. Our modified CNN performs better at accurately classifying a variety of geographic locations, according to our testing results. By efficiently capturing and analyzing complex spatial patterns, the use of customized layers enhances classification results in satellite-based geographical area classifi- cation. Index Terms—Geographical area classification, Convolutional Neural Network (CNN), Satellite imagery, Image classification, Fully connected layers, Pooling layers, Conventional layers

Almond (Prunus dulcis) varieties classification with genetic designed lightweight CNN architecture

AbstractAlmond (Prunus dulcis) is a nutritious food with a rich content. In addition to consuming as food, it is also used for various purposes in sectors such as medicine, cosmetics and bioenergy. With all these usages, almond has become a globally demanded product. Accurately determining almond variety is crucial for quality assessment and market value. Convolutional Neural Network (CNN) has a great performance in image classification. In this study, a public dataset containing images of four different almond varieties was created. Five well-known and light-weight CNN models (DenseNet121, EfficientNetB0, MobileNet, MobileNet V2, NASNetMobile) were used to classify almond images. Additionally, a model called 'Genetic CNN', which has its hyperparameters determined by Genetic Algorithm, was proposed. Among the well-known and light-weight CNN models, NASNetMobile achieved the most successful result with an accuracy rate of 99.20%, precision of 99.21%, recall of 99.20% and f1-score of 99.19%. Genetic CNN outperformed well-known models with an accuracy rate of 99.55%, precision of 99.56%, recall of 99.55% and f1-score of 99.55%. Furthermore, the Genetic CNN model has a relatively small size and low test time in comparison to other models, with a parameter count of only 1.1 million. Genetic CNN is suitable for embedded and mobile systems and can be used in real-life solutions.

DeepGAN: Utilizing generative adversarial networks for improved deep learning

In the realm of deep learning, Generative Adversarial Networks (GANs) have emerged as a topic of significant interest for their potential to enhance model performance and enable effective data augmentation. This paper addresses the existing challenges in synthesizing high-quality data and harnessing the capabilities of GANs for improved deep learning outcomes. Unlike traditional approaches that heavily rely on manually engineered data augmentation techniques, our work introduces a novel framework that leverages DeepGANs to autonomously generate diverse and high-fidelity data. Our experiments encompass a diverse spectrum of datasets, including images, text, and time series data. In the context of image classification tasks, we conduct experiments on the widely recognized CIFAR-10 dataset, which consists of 50,000 image samples. Our results demonstrate the remarkable efficacy of DeepGANs in enhancing model performance across various data domains. Notably, in image classification using the CIFAR-10 dataset, our innovative approach achieves an impressive accuracy of 97.2%. This represents a substantial advancement beyond conventional CNN models, underscoring the profound impact of DeepGANs in the realm of deep learning. In summary, this research sheds light on DeepGANs as a fundamental component in the pursuit of enhanced deep learning performance. Our framework not only overcomes existing limitations but also heralds a new era of data augmentation, with generative adversarial networks leading the way. The attainment of an accuracy rate of 97.2% on CIFAR-10 serves as a compelling testament to the transformative potential of DeepGANs, solidifying their pivotal role in the future of deep learning. This promises the development of more robust, adaptive, and accurate models across a myriad of applications, marking a significant contribution to the field.

ENHANCED TERRAIN RECOGNITION WITH DEEP LEARNING

Terrain recognition is vital for numerous real-world applications, from autonomous navigation to disaster management. Convolutional Neural Networks (CNNs) have emerged as potent tools for addressing terrain recognition challenges. In this paper, we propose an innovative approach to significantly improve terrain recognition accuracy using CNNs. We meticulously curate a dataset from Kaggle, comprising 1989 high-resolution images categorized into four terrain classes: Grassy, Sandy, Rocky, and Marshy. Our methodology revolves around the systematic design and implementation of deep learning techniques, primarily focusing on CNN architectures. Additionally, we contribute by training a CNN model tailored for classifying images into the four terrain classes. Leveraging the computational resources of Google Colab, we conduct extensive experimentation and analysis to evaluate the performance of our CNN-based terrain recognition system. Empirical results demonstrate substantial advancements in terrain recognition accuracy, underscoring the transformative role of CNNs in enhancing the efficiency and precision of terrain classification systems. Furthermore, we delve into the intricacies of our CNN model's architecture, exploring key design choices and optimization strategies. These insights deepen our understanding of CNN- based terrain recognition systems and provide valuable guidance for future research endeavors. Overall, our study highlights the practical relevance and transformative potential of CNNs in elevating terrain recognition accuracy. Keywords: Terrain recognition, Convolutional Neural Networks (CNNs), Image classification, Real-world applications, Dataset, Performance evaluation, Optimization strategies, Deep learning

MRI of pelvic endometriosis: evaluation of the mr#Enzian classification and the importance of adenomyosis subtypes.

This study aimed to investigate the utility of the #Enzian classification in magnetic resonance imaging (MRI) for endometriosis assessment, focusing on inter-reader agreement, diagnostic accuracy, and the correlation of adenomyosis with deep endometriosis (DE). This IRB- approved retrospective single-center study included 412 women who underwent MRI evaluation for endometriosis between February 2017 and June 2022. Two experienced radiologists independently analyzed MRI images using the #Enzian classification and assessed the type of adenomyosis, if any. The surgical #Enzian classification served as the gold standard for evaluating preoperative MRI results of 45 patients. Statistical analysis was performed to assess inter-reader agreement and diagnostic accuracy. Inter-reader agreement was substantial to excellent (Cohen's kappa 0.75-0.96) for most compartments except peritoneal involvement (0.39). The preoperative MRI showed mostly substantial to excellent accuracy (0.84-0.98), sensitivity (0.62-1.00), specificity (0.87-1.00), positive (0.58-1.00) and negative predictive values (0.86-1.00) for most compartments, except for peritoneal lesions (0.36, 0.17, 1.00, 1.00, 0.26 respectively). A trend with a higher prevalence of concordant DE in women with MR features of external adenomyosis compared to those with internal adenomyosis was visible (p = 0.067). The mr#Enzian showed mostly high inter-reader agreement and good diagnostic accuracy for various endometriosis compartments. MRI's role is particularly significant in the context of the current paradigm shift towards medical endometriosis treatment. The inclusion of information about the type of adenomyosis in the mr#Enzian classification could enhance diagnostic accuracy and inform treatment planning.

Pictorial Task Assistance System using Electroencephalography Signals

Neuromuscular disorders are a significant health problem globally. Patients may experience paralysis, muscle weakness, and communication problems because of these disorders. We propose a Pictorial Task Assistance System to help patients with communication issues using Electroencephalography (EEG). We developed an interface for patients containing an image of food and water. We collected EEG data from 25 healthy students using the Muse headset and Muse monitor app for our study, while they selected one of the images. The EEG data was used to train three supervised machine-learning algorithms for classification. The labels were acquired manually from participants. Using 10-fold cross-validation the results demonstrated that the Random Forest (RF) classifier achieved 88% accuracy, K-Nearest Neighbors (KNN) 80%, and 76% accuracy in Logistic Regression (LR) in the classification of food and water images. These results suggest that the proposed system has the potential to be a useful tool for patients suffering from neuromuscular disorders to perform communication for their necessary tasks.

Clinical and Biomedical Applications of Lensless Holographic Microscopy

AbstractMany clinical procedures and biomedical research workflows rely on microscopy, including diagnosis of cancer, genetic disorders, autoimmune diseases, infections, and quantification of cell culture. Despite its widespread use, traditional image acquisition and review by trained microscopists is often lengthy and expensive, limited to large hospitals or laboratories, precluding use in point‐of‐care settings. In contrast, lensless or lensfree holographic microscopy (LHM) is inexpensive and widely deployable because it can achieve performance comparable to expensive and bulky objective‐based benchtop microscopes while relying on components that cost only a few hundred dollars or less. Lab‐on‐a‐chip integration is practical and enables LHM to be combined with single‐cell isolation, sample mixing, and in‐incubator imaging. Additionally, many manual tasks in conventional microscopy are instead computational in LHM, including image focusing, stitching, and classification. Furthermore, LHM offers a field of view hundreds of times greater than that of conventional microscopy without sacrificing resolution. Here, the basic LHM principles are summarized, as well as recent advances in artificial intelligence integration and enhanced resolution. How LHM is applied to the above clinical and biomedical applications is discussed in detail. Finally, emerging clinical applications, high‐impact areas for future research, and some current challenges facing widespread adoption are identified.

Dynamic and Integrated Security Model in Inter Cloud for Image Classification

Cloud computing has transformed software and database accessibility, utilizing the Internet and server hosting. However, security risks arise, including malware attacks and website hacking. To address these challenges, deep learning models like ResNet50 have been developed. Trained on encrypted images, ResNet50 enhances the speed and accuracy of image recognition, enabling the identification of hidden data without decryption. Despite inter-cloud communication issues, cloud servers prioritize data security, user privacy, and integrity maintenance. The ResNet50 model exhibits impressive performance, achieving 99.5%accuracy and precision-recall scores of 99.5% and 99.5% using the ImageNet Dataset. Cloud computing offers significant advantages, but data security remains a critical concern. Encrypted image recognition powered by deep learning models offers efficient and private solutions. Cloud providers continually strive to improve inter-cloud communication, ensuring comprehensive protection for data and system integrity. The remarkable capabilities of ResNet50 highlight its potential in encrypted image analysis tasks.

On-device Online Learning and Semantic Management of TinyML Systems

Recent advances in Tiny Machine Learning (TinyML) empower low-footprint embedded devices for real-time on-device Machine Learning (ML). While many acknowledge the potential benefits of TinyML, its practical implementation presents unique challenges. This study aims to bridge the gap between prototyping single TinyML models and developing reliable TinyML systems in production: (1) Embedded devices operate in dynamically changing conditions. Existing TinyML solutions primarily focus on inference, with models trained offline on powerful machines and deployed as static objects. However, static models may underperform in the real world due to evolving input data distributions. We propose online learning to enable training on constrained devices, adapting local models towards the latest field conditions. (2) Nevertheless, current on-device learning methods struggle with heterogeneous deployment conditions and the scarcity of labeled data when applied across numerous devices. We introduce federated meta-learning incorporating online learning to enhance model generalization, facilitating rapid learning. This approach ensures optimal performance among distributed devices by knowledge sharing. (3) Moreover, TinyML’s pivotal advantage is widespread adoption. Embedded devices and TinyML models prioritize extreme efficiency, leading to diverse characteristics ranging from memory and sensors to model architectures. Given their diversity and non-standardized representations, managing these resources becomes challenging as TinyML systems scale up. We present semantic management for the joint management of models and devices at scale. We demonstrate our methods through a basic regression example and then assess them in three real-world TinyML applications: handwritten character image classification, keyword audio classification, and smart building presence detection. The results confirm the effectiveness of our approaches from various perspectives, such as accuracy improvement, resource savings, and engineering effort reduction.

Information bottleneck-based Hebbian learning rule naturally ties working memory and synaptic updates

Deep neural feedforward networks are effective models for a wide array of problems, but training and deploying such networks presents a significant energy cost. Spiking neural networks (SNNs), which are modeled after biologically realistic neurons, offer a potential solution when deployed correctly on neuromorphic computing hardware. Still, many applications train SNNs offline, and running network training directly on neuromorphic hardware is an ongoing research problem. The primary hurdle is that back-propagation, which makes training such artificial deep networks possible, is biologically implausible. Neuroscientists are uncertain about how the brain would propagate a precise error signal backward through a network of neurons. Recent progress addresses part of this question, e.g., the weight transport problem, but a complete solution remains intangible. In contrast, novel learning rules based on the information bottleneck (IB) train each layer of a network independently, circumventing the need to propagate errors across layers. Instead, propagation is implicit due the layers' feedforward connectivity. These rules take the form of a three-factor Hebbian update a global error signal modulates local synaptic updates within each layer. Unfortunately, the global signal for a given layer requires processing multiple samples concurrently, and the brain only sees a single sample at a time. We propose a new three-factor update rule where the global signal correctly captures information across samples via an auxiliary memory network. The auxiliary network can be trained a priori independently of the dataset being used with the primary network. We demonstrate comparable performance to baselines on image classification tasks. Interestingly, unlike back-propagation-like schemes where there is no link between learning and memory, our rule presents a direct connection between working memory and synaptic updates. To the best of our knowledge, this is the first rule to make this link explicit. We explore these implications in initial experiments examining the effect of memory capacity on learning performance. Moving forward, this work suggests an alternate view of learning where each layer balances memory-informed compression against task performance. This view naturally encompasses several key aspects of neural computation, including memory, efficiency, and locality.

All-optical Fourier Convolutional Neural Network with High Performance in Image Classification

All-optical Fourier Convolutional Neural Network with High Performance in Image Classification

AI-based image classification algorithms for Leukaemia diagnostics and hematologic cytomorphology: From single cells to molecular features

Due to the progress of image analysis and classification systems in recent years, algorithms have been developed that support morphologic examination of both single cells and tissue samples. These algorithms are typically developed using data-driven strategies, which require comprehensive, large-scale datasets. In the diagnostic workup of hematopoietic malignancies, cytomorphologic examination and differentiation represents a key first step. In recent years, the availability of large-scale, high-quality datasets of single leukocytes from peripheral blood and bone marrow has led to the development of diagnostic support algorithms for this modality. These methods not only allow a faster and more consistent classification of diagnostically relevant cell types, but also pave the way for integrated analysis of cytomorphologic and molecular findings.

Integration and performance analysis of artificial intelligence and computer vision based on deep learning algorithms

This paper focuses on the analysis of the application effectiveness of the integration of deep learning and computer vision technologies. Deep learning achieves a historic breakthrough by constructing hierarchical neural networks, enabling end-to-end feature learning and semantic understanding of images. The successful experiences in the field of computer vision provide strong support for training deep learning algorithms. The tight integration of these two fields has given rise to a new generation of advanced computer vision systems, significantly surpassing traditional methods in tasks such as machine vision image classification and object detection. In this paper, typical image classification cases are combined to analyze the superior performance of deep neural network models while also pointing out their limitations in generalization and interpretability, proposing directions for future improvements. Overall, the efficient integration and development trend of deep learning with massive visual data will continue to drive technological breakthroughs and application expansion in the field of computer vision, making it possible to build truly intelligent machine vision systems. This deepening fusion paradigm will powerfully promote unprecedented tasks and functions in computer vision, providing stronger development momentum for related disciplines and industries.

Enhancing Brain Tumour Identification through Convolutional Neural Network

A tumor is carried on by rapid and uncontrolled cell proliferation in the brain. If it is not treated in the early stages, it could prove fatal. Despite multiple significant efforts and encouraging outcomes, accurate segmentation and classification still pose a difficulty. Detection of brain tumors is extremely complicated by the distinctions in tumor position, structure, and proportions. Using computational intelligence and statistical image processing techniques, proposed in our project provide multiple approaches to detect brain cancer and tumors. Also shows an evaluation matrix for a specific system using particular systems and type of dataset. Also explains the morphology of brain tumors, accessible data sets, augmentation techniques, component extraction, and classification of Deep Learning (DL), Transfer Learning (TL), and Machine Learning (ML) models. Key Words: brain tumour, image classification, image sementation, convolutional neural networks.

Vision transformer-convolution for breast cancer classification using mammography images: A comparative study

Breast cancer is a significant global health concern, highlighting the critical importance of early detection for effective treatment of women’s health. While convolutional networks (CNNs) have been the best for analysing medical images, recent interest has emerged in leveraging vision transformers (ViTs) for medical data analysis. This study aimed to conduct a comprehensive comparison of three systems a self-attention transformer (VIT), a compact convolution transformer (CCT), and a tokenlearner (TVIT) for binary classification of mammography images into benign and cancerous tissue. Thorough experiments were performed using the DDSM dataset, which consists of 5970 benign and 7158 malignant images. The performance accuracy of the proposed models was evaluated, yielding results of 99.81% for VIT, 99.92% for CCT, and 99.05% for TVIT. Additionally, the study compared these results with the current state-of-the-art performance metrics. The findings demonstrate how convolution-attention mechanisms can effectively contribute to the development of robust computer-aided systems for diagnosing breast cancer. Notably, the proposed approach achieves high-performance results while also minimizing the computational resources required and reducing decision time.

Texture-Based Classification to Overcome Uncertainty between COVID-19 and Viral Pneumonia Using Machine Learning and Deep Learning Techniques

The SARS-CoV-2 virus, responsible for COVID-19, often manifests symptoms akin to viral pneumonia, complicating early detection and potentially leading to severe COVID pneumonia and long-term effects. Particularly affecting young individuals, the elderly, and those with weakened immune systems, the accurate classification of COVID-19 poses challenges, especially with highly dimensional image data. Past studies have faced limitations due to simplistic algorithms and small, biased datasets, yielding inaccurate results. In response, our study introduces a novel classification model that integrates advanced texture feature extraction methods, including GLCM, GLDM, and wavelet transform, within a deep learning framework. This innovative approach enables the effective classification of chest X-ray images into normal, COVID-19, and viral pneumonia categories, overcoming the limitations encountered in previous studies. Leveraging the unique textures inherent to each dataset class, our model achieves superior classification performance, even amidst the complexity and diversity of the data. Moreover, we present comprehensive numerical findings demonstrating the superiority of our approach over traditional methods. The numerical results highlight the accuracy (random forest (RF): 0.85; SVM (support vector machine): 0.70; deep learning neural network (DLNN): 0.92), recall (RF: 0.85, SVM: 0.74, DLNN: 0.93), precision (RF: 0.86, SVM: 0.71, DLNN: 0.87), and F1-Score (RF: 0.86, SVM: 0.72, DLNN: 0.89) of our proposed model. Our study represents a significant advancement in AI-based diagnostic systems for COVID-19 and pneumonia, promising improved patient outcomes and healthcare management strategies.

Automatic ceramic identification using machine learning. Lusitanian amphorae and Faience. Two Portuguese case studies

ABSTRACT This article presents a novel approach to classifying archaeological artefacts using machine learning, specifically deep learning, rather than relying on traditional, time-consuming human-based methods. By employing Convolutional Neural Networks (CNNs), this approach aims to expedite and enhance the identification process, making it more accessible to a wider audience. The study focuses on two types of artefacts- Roman Lusitanian amphorae (2nd-5th centuries) and Portuguese faience (16th-18th centuries)- chosen for their diversity. While Lusitanian amphorae lack decoration, Portuguese faience poses challenges with subtle colour variations. The study demonstrates the potential of this approach to overcome these hurdles. The paper outlines the methodology, dataset creation, and model training, emphasizing the importance of extensive data and computational resources. The ultimate objective of this research is to develop a mobile application that utilizes image classification techniques to accurately classify ceramic sherds and bring about a significant transformation in archaeological classification.

Human attention guided explainable artificial intelligence for computer vision models

Explainable artificial intelligence (XAI) has been increasingly investigated to enhance the transparency of black-box artificial intelligence models, promoting better user understanding and trust. Developing an XAI that is faithful to models and plausible to users is both a necessity and a challenge. This work examines whether embedding human attention knowledge into saliency-based XAI methods for computer vision models could enhance their plausibility and faithfulness. Two novel XAI methods for object detection models, namely FullGrad-CAM and FullGrad-CAM++, were first developed to generate object-specific explanations by extending the current gradient-based XAI methods for image classification models. Using human attention as the objective plausibility measure, these methods achieve higher explanation plausibility. Interestingly, all current XAI methods when applied to object detection models generally produce saliency maps that are less faithful to the model than human attention maps from the same object detection task. Accordingly, human attention-guided XAI (HAG-XAI) was proposed to learn from human attention how to best combine explanatory information from the models to enhance explanation plausibility by using trainable activation functions and smoothing kernels to maximize the similarity between XAI saliency map and human attention map. The proposed XAI methods were evaluated on widely used BDD-100K, MS-COCO, and ImageNet datasets and compared with typical gradient-based and perturbation-based XAI methods. Results suggest that HAG-XAI enhanced explanation plausibility and user trust at the expense of faithfulness for image classification models, and it enhanced plausibility, faithfulness, and user trust simultaneously and outperformed existing state-of-the-art XAI methods for object detection models.

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.

Semi-Supervised Medical Image Classification with Pseudo Labels Using Coalition Similarity Training

The development of medical image classification models necessitates a substantial number of labeled images for model training. In real-world scenarios, sample sizes are typically limited and labeled samples often constitute only a small portion of the dataset. This paper aims to investigate a collaborative similarity learning strategy that optimizes pseudo-labels to enhance model accuracy and expedite its convergence, known as the joint similarity learning framework. By integrating semantic similarity and instance similarity, the pseudo-labels are mutually refined to ensure their quality during initial training. Furthermore, the similarity score is utilized as a weight to guide samples away from misclassification predictions during the classification process. To enhance the model’s generalization ability, an adaptive consistency constraint is introduced into the loss function to improve performance on untrained datasets. The model achieved a satisfactory accuracy of 93.65% at 80% labeling ratio, comparable to supervised learning methods’ performance. Even with very low labeling ratio (e.g., 5%), the model still attained an accuracy of 74.28%. Comparison with other techniques such as Mean Teacher and FixMatch revealed that our approach significantly outperforms them in medical image classification tasks through improving accuracy by approximately 2%, demonstrating this framework’s leadership in medical image classification.