Large Image Datasets Research Articles

An Optimized Machine Learning and Deep Learning Framework for Facial and Masked Facial Recognition

In this study, we aimed to find an optimized approach to improving facial and masked facial recognition using machine learning and deep learning techniques. Prior studies only used a single machine learning model for classification and did not report optimal parameter values. In contrast, we utilized a grid search with hyperparameter tuning and nested cross-validation to achieve better results during the verification phase. We performed experiments on a large dataset of facial images with and without masks. Our findings showed that the SVM model with hyperparameter tuning had the highest accuracy compared to other models, achieving a recognition accuracy of 0.99912. The precision values for recognition without masks and with masks were 0.99925 and 0.98417, respectively. We tested our approach in real-life scenarios and found that it accurately identified masked individuals through facial recognition. Furthermore, our study stands out from others as it incorporates hyperparameter tuning and nested cross-validation during the verification phase to enhance the model's performance, generalization, and robustness while optimizing data utilization. Our optimized approach has potential implications for improving security systems in various domains, including public safety and healthcare.

A study on phenotypic micro-variation of stored melon based on weight loss rate

This study proposes a novel deep learning algorithm that uses weight loss rate as a basis for classification and incorporates phenotypic micro-variation features for the dynamic prediction and analysis of melon storage processes. The Resnet50 backbone network was used as a basis to construct deformable residual blocks for nonrigid objects with high similarity over time, allowing the convolutional kernel shape to adapt to the melon surface shape. A hybrid spatial and channel attention mechanism was used to focus more attentional resources on the target regions requiring precise attention. The experimental results indicate that the proposed method outperforms traditional computer vision and machine learning methods in terms of effectiveness, performance, and generalization ability. We evaluate our method on a large dataset of melon fruit images and show that it outperforms existing methods in accurately predicting storage status and optimizing control strategies in time. The proposed method will serve as a powerful tool for other nondestructive detection and accurate prediction of the shelf life of large fruit.

RGB-D Face Recognition With Identity-Style Disentanglement and Depth Augmentation

Deep learning approaches achieve highly accurate face recognition by training the models with huge face image datasets. Unlike 2D face image datasets, there is a lack of large 3D face datasets available to the public. Existing public 3D face datasets were usually collected with few subjects, leading to the over-fitting problem. This paper proposes two CNN models to improve the RGB-D face recognition task. The first is a segmentation-aware depth estimation network, called DepthNet, which estimates depth maps from RGB face images by exploiting semantic segmentation for more accurate face region localization. The other is a novel segmentation-guided RGB-D face recognition model that contains an RGB recognition branch, a depth map recognition branch, and an auxiliary segmentation mask branch. In our multi-modality face recognition model, a feature disentanglement scheme is employed to factorize the feature representation into identity-related and style-related components. DepthNet is applied to augment a large 2D face image dataset to a large RGB-D face dataset, which is used for training our RGB-D face recognition model. Our experimental results show that DepthNet can produce more reliable depth maps from face images with the segmentation mask. Our multi-modality face recognition model fully exploits the depth map and outperforms state-of-the-art methods on several public 3D face datasets with challenging variations.

Skin Cancer Detection

Abstract: Skin Cancer Detection using VGG-16 in CNN. Skin cancer is one of the most common and potentially life-threatening forms of cancer, and early detection plays a crucial role in improving patient outcomes. In recent years, Convolutional Neural Networks (CNNs) have shown remarkable success in various computer vision tasks, including medical image analysis. This abstract presents a study on the application of the VGG-16 architecture, a widely used CNN, for the detection of skin cancer. The proposed method involves training a VGG-16 model using a large dataset of dermoscopic images of skin lesions. The dataset comprises both malignant (cancerous) and benign (non-cancerous) cases, providing a diverse range of examples for the network to learn from. Through a process of supervised learning, the VGG-16 model is trained to recognize patterns and features indicative of skin cancer in these images

Toward Automated Detection of Silent Cerebral Infarcts in Children and Young Adults With Sickle Cell Anemia.

Silent cerebral infarcts (SCI) in sickle cell anemia (SCA) are associated with future strokes and cognitive impairment, warranting early diagnosis and treatment. Detection of SCI, however, is limited by their small size, especially when neuroradiologists are unavailable. We hypothesized that deep learning may permit automated SCI detection in children and young adults with SCA as a tool to identify the presence and extent of SCI in clinical and research settings. We utilized UNet-a deep learning model-for fully automated SCI segmentation. We trained and optimized UNet using brain magnetic resonance imaging from the SIT trial (Silent Infarct Transfusion). Neuroradiologists provided the ground truth for SCI diagnosis, while a vascular neurologist manually delineated SCI on fluid-attenuated inversion recovery and provided the ground truth for SCI segmentation. UNet was optimized for the highest spatial overlap between automatic and manual delineation (dice similarity coefficient). The optimized UNet was externally validated using an independent single-center prospective cohort of SCA participants. Model performance was evaluated through sensitivity and accuracy (%correct cases) for SCI diagnosis, dice similarity coefficient, intraclass correlation coefficient (metric of volumetric agreement), and Spearman correlation. The SIT trial (n=926; 31% with SCI; median age, 8.9 years) and external validation (n=80; 50% with SCI; age, 11.5 years) cohorts had small median lesion volumes of 0.40 and 0.25 mL, respectively. Compared with the neuroradiology diagnosis, UNet predicted SCI presence with 100% sensitivity and 74% accuracy. In magnetic resonance imaging with SCI, UNet reached a moderate spatial agreement (dice similarity coefficient, 0.48) and high volumetric agreement (intraclass correlation coefficient, 0.76; ρ=0.72; P<0.001) between automatic and manual segmentations. UNet, trained using a large pediatric SCA magnetic resonance imaging data set, sensitively detected small SCI in children and young adults with SCA. While additional training is needed, UNet may be integrated into the clinical workflow as a screening tool, aiding in SCI diagnosis.

Drone Detection and Classification Using Music (Multiple Signal Classification) Algorithm

Abstract: The rapid augmentation of drones in different sectors has raised concerns regarding privacy, security, and safety. This project proposes a novel drone detection and classification approach by integrating the YOLOv5 framework with a MUSICbased algorithm to address the above issues. The aim is to postulate an efficient and accurate model capable of identifying and classifying different types of drones in real time. The project leverages the YOLOv5 architecture incorporated with the MUSIC algorithm for superior speed and accuracy. By training YOLOv5 on a large dataset of drone images, the model can learn to identify visual features and specific characteristics unique to drones, permitting effective detection and classification

Classification of Brain Tumor as Benign or Malignant Using Image Processing and Deep Learning Techniques: A Comparative Study

bstract: In India, the dominance of brain tumors is 5-10 per 100,000 people. According to the International Association of Cancer Registries (IARC), over 28,000 cases of brain tumors are recorded each year in India, with over 24,000 individuals dying from them each year. The proposed approach includes pre-processing of the Magnetic Resonance Imaging (MRI) scans followed by the extraction of the Region of Interest (ROI) using image processing techniques. The extracted parts are then used as input to train a deep neural network model for classification. The Convolutional Neural(CNN) is trained on a large dataset of MRI images of brain tumors, and evaluated using various performance metrics. Deep Learning models like VGG16, ResNet50 are used for the classification of images. Deep learning model with VGG16 gives 98.5 % accuracy which is better than CNN and ResNet50.

Generative Data Augmentation and ARMD Classification

Abstract: Age Related Macular Degeneration (ARMD) is a type of eye disease which normally have an effect on the central vision of a person. This Disease might sometimes lead to permanent vision loss for some people. It affects the people over the age of 50. So, basically there are 2 different types of ARMD i.e., Dry and Wet. Dry ARMD will generate a tiny amount of protein deposits called drusen, whereas Wet ARMD occurs whenever any abnormal blood vessel is developed under the retina, so sometimes this blood vessels might leak blood fluid, this type of ARMD is very severe and can even lead to permanent central vision loss. Therefore, it is necessary for early detection of the disease. Generative Data Augmentation for ARMD Classification is deep learning based which uses Convolutional Neural Network (CNN) model for generating images to accurately identify the disease. Deep Learning Diagnostic models require expertly graded images from extensive data sets obtained in large scale clinical trials which may not exist. Therefore, (Generative Adversarial Networks) GAN-based generative data augmentation method called Style GAN is used for generating the images. Generative deep learning techniques is used to synthesize new large datasets of artificial retinal images from different stages of ARMD using the images from the already existing datasets. The performance of ARMD diagnostic DCNNs will be trained on the combination of both real and synthetic datasets. Images obtained by using GAN appear to be realistic, and increase the accuracy of the model. It then continues with classifying the retinal images into one of the three classes i.e., dry, wet or normal using CNN model. It also compares the accuracy against the model with traditional augmentation techniques, towards improving the performance of real-world ARMD classification tasks.

CROWD-SOURCED SURVEYING FOR BUILDING ARCHAEOLOGY: THE POTENTIAL OF STRUCTURE FROM MOTION (SFM) AND NEURAL RADIANCE FIELDS (NERF)

Abstract. This contribution presents a simple workflow for surveying historical buildings and sites using crowd-sourced images. The proposed approach involves collecting large datasets of images from the internet using free plugins, followed by automatic image analysis and filtering using AI-based tools. 3D reconstructions are then created with Structure from Motion (SfM) and neural radiance fields (NeRF). To assess the reliability of crowd-sourced surveys, the 3D reconstructions are compared to high-precision laser scans of large medieval churches. In addition, the paper demonstrates the potential of this workflow in the field of building archaeology through detailed geometrical analyses of several iconic domes such as Hagia Sofia. By enabling remote and 4D surveys, crowd-sourced reconstruction methods open up novel opportunities for rapid, affordable and borderless research on cultural heritage.

Al-Biruni Earth Radius Optimization with Transfer Learning Based Histopathological Image Analysis for Lung and Colon Cancer Detection.

An early diagnosis of lung and colon cancer (LCC) is critical for improved patient outcomes and effective treatment. Histopathological image (HSI) analysis has emerged as a robust tool for cancer diagnosis. HSI analysis for a LCC diagnosis includes the analysis and examination of tissue samples attained from the LCC to recognize lesions or cancerous cells. It has a significant role in the staging and diagnosis of this tumor, which aids in the prognosis and treatment planning, but a manual analysis of the image is subject to human error and is also time-consuming. Therefore, a computer-aided approach is needed for the detection of LCC using HSI. Transfer learning (TL) leverages pretrained deep learning (DL) algorithms that have been trained on a larger dataset for extracting related features from the HIS, which are then used for training a classifier for a tumor diagnosis. This manuscript offers the design of the Al-Biruni Earth Radius Optimization with Transfer Learning-based Histopathological Image Analysis for Lung and Colon Cancer Detection (BERTL-HIALCCD) technique. The purpose of the study is to detect LCC effectually in histopathological images. To execute this, the BERTL-HIALCCD method follows the concepts of computer vision (CV) and transfer learning for accurate LCC detection. When using the BERTL-HIALCCD technique, an improved ShuffleNet model is applied for the feature extraction process, and its hyperparameters are chosen by the BER system. For the effectual recognition of LCC, a deep convolutional recurrent neural network (DCRNN) model is applied. Finally, the coati optimization algorithm (COA) is exploited for the parameter choice of the DCRNN approach. For examining the efficacy of the BERTL-HIALCCD technique, a comprehensive group of experiments was conducted on a large dataset of histopathological images. The experimental outcomes demonstrate that the combination of AER and COA algorithms attain an improved performance in cancer detection over the compared models.

P-306 Generalizable AI model for microscopic and timelapse multifocal embryo images

Abstract Study question Can an AI model be applied to datasets with different characteristics? Summary answer The AI model predicting pregnancy developed on microscope images can be generally applied to multifocal images from time-lapse system. What is known already AI models require large datasets to be able to generalize and handle various scenarios, but it can be challenging to gather enough data for each case. Embryologists evaluate embryos using multiple focal planes and add color filters as needed, and timelapse images have varying features. AI models in past studies were trained using both microscopic and timelapse images, and embryologists question if images must be captured at a specific focal plane for AI to work effectively. In this study, the AI model trained on over 2,000 microscopic embryonic images was validated using timelapse images taken at multiple focal planes. Study design, size, duration We collected 2,555 microscopic images from 7 IVF clinics and 1299 timelapse images from 433 embryos from a single IVF clinic between July 2016 and December 2020. The timelapse images were divided into 3 groups, with Group 1 being the best visualized ICM images, Group 2 being 20 µm higher or lower than Group 1, and Group 3 being 20 µm higher or lower than Group 2. Participants/materials, setting, methods We built 2 CNN models. The “Microscopic model” and “Timelapse model” were trained and validated using 3-fold cross validation with 2,555 microscopic images, 433 timelapse Group 1 images, respectively. To examine whether the Timelapse model was able to infer well for images taken at different focal points, Group 1, 2, and 3 images were used to test the Microscopic model. Main results and the role of chance The AUROCs and accuracies in mean (SD) for the Microscopic model were 0.738 (0.003) and 0.705 (0.011) after 3-fold cross-validation. The AUROC and accuracy for the Timelapse model were 0.627 and 0.583. The AUROCs for Group 1,2, and 3 were 0.699 (0.014), 0.705 (0.001) and 0.701 (0.064), respectively. The accuracies for Group 1,2, and 3 were 0.655. (0.013), 0.647 (0.007), and 0.651 (0.007). The predictive power of the Microscopic model applied to the time-lapse images was better than that of the Timelapse model, although it was not as accurate as the Microscopic model applied to the microscope images. Limitations, reasons for caution The limitations of the study include its retrospective nature and a small dataset. Transfer learning of timelapse images to a microscopic model may be beneficial for analyzing timelapse images. Wider implications of the findings The study found that timelapse images with multifocal planes can be applied to an AI model built based on a large dataset of microscopic images collected from multiple centers. It may be prudent for a small IVF clinic to apply a generalizable model rather than building its own. Trial registration number not applicable

A large expert-curated cryo-EM image dataset for machine learning protein particle picking

Cryo-electron microscopy (cryo-EM) is a powerful technique for determining the structures of biological macromolecular complexes. Picking single-protein particles from cryo-EM micrographs is a crucial step in reconstructing protein structures. However, the widely used template-based particle picking process is labor-intensive and time-consuming. Though machine learning and artificial intelligence (AI) based particle picking can potentially automate the process, its development is hindered by lack of large, high-quality labelled training data. To address this bottleneck, we present CryoPPP, a large, diverse, expert-curated cryo-EM image dataset for protein particle picking and analysis. It consists of labelled cryo-EM micrographs (images) of 34 representative protein datasets selected from the Electron Microscopy Public Image Archive (EMPIAR). The dataset is 2.6 terabytes and includes 9,893 high-resolution micrographs with labelled protein particle coordinates. The labelling process was rigorously validated through 2D particle class validation and 3D density map validation with the gold standard. The dataset is expected to greatly facilitate the development of both AI and classical methods for automated cryo-EM protein particle picking.

Deep learning morphological distribution analysis of metal alloy catalysts in proton exchange membrane fuel cells

Measuring the variation of the morphological distribution of Pt-based catalyst particles on supports using transmission electron microscopy provides crucial information for understanding the performance degradation behaviors of proton exchange membrane fuel cells and for designing more durable electrocatalysts. However, interpretation based on a few micrographs is statistically insignificant, whereas manual analyses of large image datasets are time-consuming and often require subjective human decisions. To address this issue, an efficient method assisted by deep learning is proposed for the automated interpretation of massive image datasets of metal catalyst nanoparticles (NPs). Based on an attention-aided deep convolutional neural network for object detection (that is, Attention U-Net), the proposed approach rapidly measures the changes in the structural parameters of Pt/Co NPs on a porous support and quantitatively evaluates morphological changes of the NPs after cycling, with statistical significance for their sizes and separating distances.

Compression ensembles quantify aesthetic complexity and the evolution of visual art

To the human eye, different images appear more or less complex, but capturing this intuition in a single aesthetic measure is considered hard. Here, we propose a computationally simple, transparent method for modeling aesthetic complexity as a multidimensional algorithmic phenomenon, which enables the systematic analysis of large image datasets. The approach captures visual family resemblance via a multitude of image transformations and subsequent compressions, yielding explainable embeddings. It aligns well with human judgments of visual complexity, and performs well in authorship and style recognition tasks. Showcasing the functionality, we apply the method to 125,000 artworks, recovering trends and revealing new insights regarding historical art, artistic careers over centuries, and emerging aesthetics in a contemporary NFT art market. Our approach, here applied to images but applicable more broadly, provides a new perspective to quantitative aesthetics, connoisseurship, multidimensional meaning spaces, and the study of cultural complexity.

Deep learning‐based image processing for financial audit risk quantification in healthcare

AbstractRecently, deep learning technology has gradually penetrated various fields. Today, the field of healthcare is also closely linked to deep learning technology. Image processing technology based on deep learning can accurately segment medical images, which is convenient for medical research and pathological analysis. Accurate distribution of images can effectively save medical resources. Therefore, image processing techniques can contribute to the quantification and assessment of economic audit risks. In recent years, medical image segmentation has achieved many research results. However, with the improvement of accuracy, the segmentation standards of medical images are also becoming more and more stringent. For medical images, they tend to have rough and fuzzy boundaries and noise disturbances of different shapes. The above problems pose challenges for accurate localization and segmentation of lesion regions. On the other hand, in the field of medical images, there are also problems such as unbalanced number of samples and scarcity of large medical image datasets. In response to these problems, this paper conducts research work and proposes an Attention Mechanism and Multi‐Scale spatial Pooling‐based conditional Adversarial Network (AM‐MSP‐cGAN) model to achieve automatic segmentation of medical images. AM‐MSP‐cGAN can learn more detailed features from fuzzy boundaries, and effectively solve the problem of data lack, thereby promoting economic audit risk quantification and assessment in the healthcare field.

Automated Brain Tumor Detection from MRI Scans using Deep Convolutional Neural Networks

The brain, as the central nervous system's most critical part, can develop abnormal growths of cells known as tumors. Cancer is the term used to describe malignant tumors. Medical imaging modalities, such as computed tomography (CT) or magnetic resonance imaging (MRI), are commonly used to detect cancerous regions in the brain. Other techniques, such as positron emission tomography (PET), cerebral arteriography, lumbar puncture, and molecular testing, are also utilized for brain tumor detection. MRI scans provide detailed information concerning delicate tissue, which aids in diagnosing brain tumors. MRI scan images are analyzed to assess the disease condition objectively. The proposed system aims to identify abnormal brain images from MRI scans accurately. The segmented mask can estimate the tumor's density, which is helpful in therapy. Deep learning techniques are employed to automatically extract features and detect abnormalities from MRI images. The proposed system utilizes a convolutional neural network (CNN), a popular deep learning technique, to analyze MRI images and identify abnormal brain scans with high accuracy. The system's training process involves feeding the CNN with large datasets of normal and abnormal MRI images to learn how to differentiate between the two. During testing, the system classifies MRI images as either normal or abnormal based on the learned features. The system's ability to accurately identify abnormal brain scans can aid medical practitioners in making informed decisions and providing better patient care. Additionally, the system's ability to estimate tumor density from the segmented mask provides additional information to guide therapy. The proposed system offers a promising solution for improving the accuracy and efficiency of brain tumor detection from MRI images, which is critical for early detection and treatment.

Early Warning System to Prevent Animal-Train Collision

The project aims to prevent animal-train collisions by using an image processing system that can identify obstacles on the train tracks, particularly animals. The project utilizes machine learning algorithms to identify the obstacles and send a notification to the train if an obstacle is detected. The system uses an IoT application and Bluetooth transmitter to send a message to the train and trigger a buzzer notification. By utilizing technology, the project hopes to reduce the number of animal-train collisions and improve safety for both humans and animals. The system utilizes machine learning algorithms to identify the obstacles in real-time and send a notification to the train via an IoT application and Bluetooth transmitter. If an obstacle is detected, a buzzer notification will be triggered on the train, alerting the driver to slow down or stop. The image processing system is designed to be highly accurate and efficient in identifying animals on the tracks, even in challenging lighting or weather conditions. The system is trained using a large dataset of images of animals and non- animal objects on train tracks, enabling it to recognize animals of various sizes, shapes, and colors. Overall, the goal of this project is to reduce the number of animal-train collisions and improve the safety of train travel. By leveraging the power of machine learning and IoT technology, this system can provide an effective solution to a pressing safety issue." To accomplish this, the project utilizes an image processing system that is highly accurate and efficient in identifying animals on the tracks, even in challenging lighting or weather conditions. The system is trained using a large dataset of images of animals and non-animal objects on train tracks, enabling it to recognize animals of various sizes, shapes, and colors. The goal of this project is to reduce the number of animal- train collisions and improve the safety of train travel. By leveraging the power of machine learning and IoT technology, this system can provide an effective solution to a pressing safety issue.

Speckle noise removal via learned variational models

In this paper, we address the image denoising problem in presence of speckle degradation typically arising in ultra-sound images. Variational methods and Convolutional Neural Networks (CNNs) are considered well-established methods for specific noise types, such as Gaussian and Poisson noise. Nonetheless, the advances achieved by these two classes of strategies are limited when tackling the de-speckle problem. In fact, variational methods for speckle removal typically amounts to solve a non-convex functional with the related issues from the convergence viewpoint; on the other hand, the lack of large datasets of noise-free ultra-sound images has not allowed the extension of the state-of-the-art CNN denoiser methods to the case of speckle degradation. Here, we aim at combining the classical variational methods with the predictive properties of CNNs by considering a weighted total variation regularized model; the local weights are obtained as the output of a statistically inspired neural network that is trained on a small and composite dataset of natural and synthetic images. The resulting non-convex variational model, which is minimized by means of the Alternating Direction Method of Multipliers (ADMM) is proven to converge to a stationary point. Numerical tests show the effectiveness of our approach for the denoising of natural and satellite images.

Region-of-Interest Optimization for Deep-Learning-Based Breast Cancer Detection in Mammograms

The early detection and diagnosis of breast cancer may increase survival rates and reduce overall treatment costs. The cancer of the breast is a severe and potentially fatal disease that impacts individuals worldwide. Mammography is a widely utilized imaging technique for breast cancer surveillance and diagnosis. However, images produced with mammography frequently contain noise, poor contrast, and other anomalies that hinder radiologists from interpreting the images. This study develops a novel deep-learning technique for breast cancer detection using mammography images. The proposed procedure consists of two primary steps: region-of-interest (ROI) (1) extraction and (2) classification. At the beginning of the procedure, a YOLOX model is utilized to distinguish breast tissue from the background and to identify ROIs that may contain lesions. In the second phase, the EfficientNet or ConvNeXt model is applied to the data to identify benign or malignant ROIs. The proposed technique is validated using a large dataset of mammography images from various institutions and compared to several baseline methods. The pF1 index is used to measure the effectiveness of the technique, which aims to establish a balance between the number of false positives and false negatives, and is a harmonic mean of accuracy and recall. The proposed method outperformed existing methods by an average of 8.0%, obtaining superior levels of precision and sensitivity, and area under the receiver operating characteristics curve (ROC AUC) and the precision–recall curve (PR AUC). In addition, ablation research was conducted to investigate the effects of the procedure’s numerous components. According to the findings, the proposed technique is another choice that could enhance the detection and diagnosis of breast cancer using mammography images.

Evaluating synthetic pre-Training for handwriting processing tasks

In this work, we explore massive pre-training on synthetic word images for enhancing the performance on four benchmark downstream handwriting analysis tasks. To this end, we build a large synthetic dataset of word images rendered in several handwriting fonts, which offers a complete supervision signal. We use it to train a simple convolutional neural network (ConvNet) with a fully supervised objective. The vector representations of the images obtained from the pre-trained ConvNet can then be considered as encodings of the handwriting style. We exploit such representations for Writer Retrieval, Writer Identification, Writer Verification, and Writer Classification and demonstrate that our pre-training strategy allows extracting rich representations of the writers’ style that enable the aforementioned tasks with competitive results with respect to task-specific State-of-the-Art approaches.