Deep Learning Based Image Classification Research Articles

Ensemble Deep Learning-Based Image Classification for Breast Cancer Subtype and Invasiveness Diagnosis from Whole Slide Image Histopathology.

Cancer diagnosis and classification are pivotal for effective patient management and treatment planning. In this study, a comprehensive approach is presented utilizing ensemble deep learning techniques to analyze breast cancer histopathology images. Our datasets were based on two widely employed datasets from different centers for two different tasks: BACH and BreakHis. Within the BACH dataset, a proposed ensemble strategy was employed, incorporating VGG16 and ResNet50 architectures to achieve precise classification of breast cancer histopathology images. Introducing a novel image patching technique to preprocess a high-resolution image facilitated a focused analysis of localized regions of interest. The annotated BACH dataset encompassed 400 WSIs across four distinct classes: Normal, Benign, In Situ Carcinoma, and Invasive Carcinoma. In addition, the proposed ensemble was used on the BreakHis dataset, utilizing VGG16, ResNet34, and ResNet50 models to classify microscopic images into eight distinct categories (four benign and four malignant). For both datasets, a five-fold cross-validation approach was employed for rigorous training and testing. Preliminary experimental results indicated a patch classification accuracy of 95.31% (for the BACH dataset) and WSI image classification accuracy of 98.43% (BreakHis). This research significantly contributes to ongoing endeavors in harnessing artificial intelligence to advance breast cancer diagnosis, potentially fostering improved patient outcomes and alleviating healthcare burdens.

Artificial intelligence-driven microalgae autotrophic batch cultivation: A comparative study of machine and deep learning-based image classification models

The goal of this study is to classify microalgae of different species, using machine learning (ML) and deep learning (DL) methods. At present, we applied gray-scaling, bilateral filtering, adaptive thresholding, Sobel edge detection, and Canny edge detection, for the segmentation of microalgae. Morphological and texture descriptors, which are part of the important geometrical features, were used for feature extraction. Results indicates that the final combined features, with optimised image pre-processing techniques, produced high accuracy of 96.93 % and 97.63 % for k-nearest neighbours (k−NN) and support vector machine (SVM) classifiers, respectively. Overall, the Azure custom vision model performed the best with the highest accuracy of 97.67 % and 97.86 % at probability threshold of 50 % and 80 %, respectively. Our study aimed to bridge artificial intelligence technologies to microalgae based on understanding of shape, texture, and convolution features, which could accelerate the development of real-time monitoring, as well as rapid and precise microalgae classification.

Finding the input features that reduce the entropy of a neural network’s prediction

In deep learning-based image classification, the entropy of a neural network’s output is often taken as a measure of its uncertainty. We introduce an explainability method that identifies those features in the input that impact most this uncertainty. Learning the corresponding features by straightforward backpropagation typically leads to results that are hard to interpret. We propose an extension of the recently proposed oriented, modified integrated gradients (OMIG) technique as an alternative to produce perturbations of the input that have a visual quality comparable to explainability methods from the literature but marks features that have a substantially higher impact on the entropy. The potential benefits of the modified OMIG method are demonstrated by comparison with current state-of-the-art explainability methods on several popular databases. In addition to a qualitative analysis of explainability results, we propose a metric for their quantitative comparison, which evaluates the impact of identified features on the entropy of a prediction.

Optimal Deep Learning-Based Image Classification for IoT-Enabled UAVs in Remote Sensing Applications

Optimal Deep Learning-Based Image Classification for IoT-Enabled UAVs in Remote Sensing Applications

Intelligent analysis of corrosion characteristics of steel pipe piles of offshore construction wharfs based on computer vision

The lower part of offshore construction wharfs is mostly a steel structure system composed of steel pipe piles, whose corrosion level directly affects the structural safety performance of steel wharfs in service. The currently common corrosion detection methods can only sample and inspect steel pile after it has been dismantled, making it impractical for in-service monitoring during the operational period of the steel pile. In this paper, a deep learning-based image classification model is first established to recognize the type of corroded area on steel pipe piles. The model achieves a recognition accuracy of 99.14 % in automatically identifying different types of corroded areas, including full immersion zone, tidal range zone, and splash zone. Subsequently, digital image processing technology is utilized to automatically calculate the corroded area of steel pipe piles. The method proposed in this paper can obtain the key information, such as type of corrosion area and area of the steel pipe pile corrosion area, without damaging their structural performance during the service. With this data, the mechanical performance of steel pipe piles can be analyzed, and the structural safety of the in-service steel pipe piles can be determined, thereby ensuring the safety of the construction wharf.

Deep learning-based image classification of MRI brain image

This article reviews the latest research on MRI brain image classification techniques based on deep learning. Firstly, MRI brain image classification and traditional machine learning methods applied to MRI image classification were briefly introduced. Then, a review was conducted on existing MRI brain image classification methods based on deep learning. The article reviews past and recent related research and provides a detailed introduction to the application of deep learning methods in MRI brain image classification. This includes some traditional machine learning methods and research achievements in deep neural networks (DNN), convolutional neural networks (CNN), and transfer learning. The most commonly used deep learning architecture for image classification is CNN. Research has shown that deep learning methods have high accuracy and performance in MRI brain image classification and can automatically extract image features for effective classification. Therefore, deep learning methods provide doctors with more comprehensive information, help them make more accurate diagnoses and formulate treatment plans, and have broad prospects for application in MRI brain image classification. In the future, the further development of deep learning technology will achieve better auxiliary effects on diagnosis, treatment, and scientific research related to the brain.

Using green background for dermatological images to improve deep learning-based image classification.

Using green background for dermatological images to improve deep learning-based image classification.

Deep learning-based image classification of turtles imported into Korea

Although turtles play a key role in maintaining healthy and balanced environments, these are endangered due to global trade to meet the high demand for food, medicine, and pets in Asia. In addition, imported non-native turtles have been controlled as alien invasive species in various countries, including Korea. Therefore, a rapid and accurate classification of imported turtles is needed to conserve and detect those in native ecosystems. In this study, eight Single Shot MultiBox Detector (SSD) models using different backbone networks were used to classify 36 imported turtles in Korea. The images of these species were collected from Google and were identified using morphological features. Then, these were divided into 70% for training, 15% for validation, and 15% for test sets. In addition, data augmentation was applied to the training set to prevent overfitting. Among the eight models, the Resnet18 model showed the highest mean Average Precision (mAP) at 88.1% and the fastest inference time at 0.024 s. The average correct classification rate of 36 turtles in this model was 82.8%. The results of this study could help in management of the turtle trade, specifically in improving detection of alien invasive species in the wild.

Indoor/Outdoor Deep Learning Based Image Classification for Object Recognition Applications

With the rapid development of smart devices, people's lives have become easier, especially for visually disabled or special-needs people. The new achievements in the fields of machine learning and deep learning let people identify and recognise the surrounding environment. In this study, the efficiency and high performance of deep learning architecture are used to build an image classification system in both indoor and outdoor environments. The proposed methodology starts with collecting two datasets (indoor and outdoor) from different separate datasets. In the second step, the collected dataset is split into training, validation, and test sets. The pre-trained GoogleNet and MobileNet-V2 models are trained using the indoor and outdoor sets, resulting in four trained models. The test sets are used to evaluate the trained models using many evaluation metrics (accuracy, TPR, FNR, PPR, FDR). Results of Google Net model indicate the high performance of the designed models with 99.34% and 99.76% accuracies for indoor and outdoor datasets, respectively. For Mobile Net models, the result accuracies are 99.27% and 99.68% for indoor and outdoor sets, respectively. The proposed methodology is compared with similar ones in the field of object recognition and image classification, and the comparative study proves the transcendence of the propsed system.

Benign and Malignant Oral Lesion Image Classification Using Fine-Tuned Transfer Learning Techniques.

Oral lesions are a prevalent manifestation of oral disease, and the timely identification of oral lesions is imperative for effective intervention. Fortunately, deep learning algorithms have shown great potential for automated lesion detection. The primary aim of this study was to employ deep learning-based image classification algorithms to identify oral lesions. We used three deep learning models, namely VGG19, DeIT, and MobileNet, to assess the efficacy of various categorization methods. To evaluate the accuracy and reliability of the models, we employed a dataset consisting of oral pictures encompassing two distinct categories: benign and malignant lesions. The experimental findings indicate that VGG19 and MobileNet attained an almost perfect accuracy rate of 100%, while DeIT achieved a slightly lower accuracy rate of 98.73%. The results of this study indicate that deep learning algorithms for picture classification demonstrate a high level of effectiveness in detecting oral lesions by achieving 100% for VGG19 and MobileNet and 98.73% for DeIT. Specifically, the VGG19 and MobileNet models exhibit notable suitability for this particular task.

MRI Kidney Tumor Image Classification with SMOTE Preprocessing and SIFT-tSNE Features using CNN

Kidney tumor detection is a challenging task due to the complexity of tumor characteristics and variability in imaging modalities. In this paper, we propose a deep learning-based approach for detecting kidney tumors with 98.5% accuracy. Our method addresses the issue of an imbalanced dataset by applying the Synthetic Minority Over-sampling Technique (SMOTE) to balance the distribution of images. SMOTE generates synthetic samples of the minority class to increase the number of samples, thus providing a balanced dataset. We utilize a convolutional neural network (CNN) architecture that is trained on this balanced dataset of kidney tumor images. The CNN can learn and extract relevant features from the images, resulting in precise tumor classification. We evaluated our approach on a separate dataset and compared it with state-of-the-art methods. The results demonstrate that our method not only outperforms other methods but also shows robustness in detecting kidney tumors with a high degree of accuracy. By enabling early detection and diagnosis of kidney tumors, our proposed method can potentially improve patient outcomes. Additionally, addressing the imbalance in the dataset using SMOTE demonstrates the usefulness of this technique in improving the performance of deep learning-based image classification systems.

Application of residual neural networks to detect and quantify milk adulterations

In order to develop a milk adulteration detection system, a method based on optimized convolutional neural networks has been developed. First, milks of different animal origins and lactose content have been artificially adulterated with water within the 1 to 25 ppm range to prepare samples that will lead to a broad image database. The milks chosen for this work were from cow, goat, sheep, as well as lactose-free cow milk. The samples were photographed in two light conditions (1/30 s for bright and 1/500 s for dark), resulting in a total of 10,400 images. As for the mathematical algorithm, two residual neural networks based on transfer learning were trained to perform the classification of the milk samples, leading to accuracies above 93%. These results open the door to the design of prototypes for fast, easy-to-use, and reliable adulterant detection in the dairy industry by means of deep learning-based image classification.

Deep Learning-Based Image Classification for Major Mosquito Species Inhabiting Korea.

Mosquitoes are one of the deadliest insects, causing harm to humans worldwide. Preemptive prevention and forecasting are important to prevent mosquito-borne diseases. However, current mosquito identification is mostly conducted manually, which consumes time, wastes labor, and causes human error. In this study, we developed an automatic image analysis method to identify mosquito species using a deep learning-based object detection technique. Color and fluorescence images of live mosquitoes were acquired using a mosquito capture device and were used to develop a deep learning-based object detection model. Among the deep learning-based object identification models, the combination of a swine transformer and a faster region-convolutional neural network model demonstrated the best performance, with a 91.7% F1-score. This indicates that the proposed automatic identification method can be rapidly applied for efficient analysis of species and populations of vector-borne mosquitoes with reduced labor in the field.

Deep Learning-Based Business Recommendation System in Intelligent Vehicles

The advancements in intelligent vehicle technologies are facilitating the growth of information technology (IT) platforms, unlike conventional automobiles. In-vehicle-infotainment (IVI) is becoming an appealing element in intelligent vehicles as it offers various experiences to users; however, it requires personalized services to provide even more sophisticated user experiences. It is supposed that passengers search for businesses that provide products or services they found interesting in videos played via IVIs while the vehicle is driving autonomously. In that case, it could be more effective to use images that can express the user’s preference as a query for the search than to utilize texts such as product names. Accordingly, this study proposes a recommendation system that informs users of businesses near an intelligent vehicle when a passenger inputs an image of a product or service into an IVI system. The proposed recommendation system involves training deep learning-based image classification models with the user’s interest images to classify the category, measure the similarity with the business category using Word2vec, and finally provide the locations of the businesses with a high degree of similarity via IVI, using a smartphone. The experimental results indicated that the user’s interest image exhibited 85% accuracy for category classification via the EfficientNet B0 model, while the similarity between the image and business categories using Word2vec was particularly high in the business category similar to the actual image category.

Improved Classification of Blurred Images with Deep-Learning Networks Using Lucy-Richardson-Rosen Algorithm

Pattern recognition techniques form the heart of most, if not all, incoherent linear shift-invariant systems. When an object is recorded using a camera, the object information is sampled by the point spread function (PSF) of the system, replacing every object point with the PSF in the sensor. The PSF is a sharp Kronecker Delta-like function when the numerical aperture (NA) is large with no aberrations. When the NA is small, and the system has aberrations, the PSF appears blurred. In the case of aberrations, if the PSF is known, then the blurred object image can be deblurred by scanning the PSF over the recorded object intensity pattern and looking for pattern matching conditions through a mathematical process called correlation. Deep learning-based image classification for computer vision applications gained attention in recent years. The classification probability is highly dependent on the quality of images as even a minor blur can significantly alter the image classification results. In this study, a recently developed deblurring method, the Lucy-Richardson-Rosen algorithm (LR2A), was implemented to computationally refocus images recorded in the presence of spatio-spectral aberrations. The performance of LR2A was compared against the parent techniques: Lucy-Richardson algorithm and non-linear reconstruction. LR2A exhibited a superior deblurring capability even in extreme cases of spatio-spectral aberrations. Experimental results of deblurring a picture recorded using high-resolution smartphone cameras are presented. LR2A was implemented to significantly improve the performances of the widely used deep convolutional neural networks for image classification.

102 AI-Based Molecular Classification of Diffuse Gliomas using Rapid, Label-Free Optical Imaging

Molecular classification has transformed the management of brain tumors by enabling more accurate prognostication and personalized treatment. Access to timely molecular diagnostic testing for brain tumor patients is limited, complicating surgical and adjuvant treatment and obstructing clinical trial enrollment. By combining stimulated Raman histology (SRH), a rapid, label-free, non-consumptive, optical imaging method, and deep learning-based image classification, we are able to predict the molecular genetic features used by the World Health Organization (WHO) to define the adult-type diffuse glioma taxonomy, including IDH-1/2, 1p19q-codeletion, and ATRX loss. We developed a multimodal deep neural network training strategy that uses both SRH images and large-scale, public diffuse glioma genomic data (i.e. TCGA, CGGA, etc.) in order to achieve optimal molecular classification performance. One institution was used for model training (University of Michigan) and four institutions (NYU, UCSF, Medical University of Vienna, and University Hospital Cologne) were included for patient enrollment in the prospective testing cohort. Using our system, called DeepGlioma, we achieved an average molecular genetic classification accuracy of 93.2% and identified the correct diffuse glioma molecular subgroup with 91.5% accuracy within 2 minutes in the operating room. DeepGlioma outperformed conventional IDH1-R132H immunohistochemistry (94.2% versus 91.4% accuracy) as a first-line molecular diagnostic screening method for diffuse gliomas and can detect canonical and non-canonical IDH mutations. Our results demonstrate how artificial intelligence and optical histology can be used to provide a rapid and scalable alternative to wet lab methods for the molecular diagnosis of brain tumor patients during surgery.

An abbreviated review of deep learning-based image classification models

Image classification is an extensively researched sub-fields of computer vision implemented in face recognition, self-driving, medical image segmentation, biological identification, and others. Traditional models of image classification require manual construction of feature extraction techniques and classification accuracy which are closely associated with these utilized techniques. During the rapid progress of multimedia technologies, the number of images that require classification got bigger, and this led to making image classification more complicated, hence, the manual construction of feature extraction techniques consumes more time and provides lower accuracy. In the recent decade, deep learning-based models have appeared in various applications. These models hold the merits of an effective extraction of image features, low-weight features filtering, a large capacity for processing, and higher classification speed and accuracy. Thus, lots of researchers have attempted to utilize deep learning algorithms, especially convolutional neural networks (CNNs) for image classification. Therefore, this paper concentrates on providing an abbreviated review of deep learning-based image classification models, by covering the recently utilized deep learning algorithms, comparing various related works and benchmark datasets mentioned in this paper, and summarizing the fundamental analysis and discussion.

Transfer Learning based Image Classification of Diseased Tomato Leaves with Optimal Fine-Tuning combined with Heat Map Visualization

India is one of the leading countries in Tomato cultivation and production. Coimbatore, Dharmapuri, Salem,Krishnagiri are some of the foremost districts in tomato production. According to a survey, Tomato Leaf Curl virus is one of the most ravage class of disease and the maximum incidence of the disease is reported as 95.3% in Dharmapuri district of Tamil Nadu. The objective of the work is to identify the most recurrent tomato plant diseases using deep learning-based image classification techniques and to develop a plant disease classification web application. In this paper, Tomato based leaf disease images alone is used from Plant Village Dataset downloaded from Kaggle platform. Inception V3, a pre-trained transfer learning model is used to classify this multi-class tomato plant disease by using the leaf symptoms such as dark brown lesions, concentric rings etc. Data augmentation is performed while training the model on the images using Keras Image Data Generator on various layers and feature extraction is performed. The training accuracy and validation accuracy of the model before and after fine-tuning is 80.21%,76.29% and 81.72%,77.63% respectively. The highest precision, recall and F1 score is 95, 93 and 94 respectively is recorded for Tomato Yellow Leaf Curl Virus Disease for which the yield losses is severe. Further, activations are used to generate an attention map in the form of Heat Maps which is included as a post processing step before the classification of the output. Plant Leaf Disease Classification- A web application is deployed using Streamlit Python library and Ngrok services

A survey of automated data augmentation algorithms for deep learning-based image classification tasks

In recent years, one of the most popular techniques in the computer vision community has been the deep learning technique. As a data-driven technique, deep model requires enormous amounts of accurately labelled training data, which is often inaccessible in many real-world applications. A data-space solution is Data Augmentation (DA), that can artificially generate new images out of original samples. Image augmentation strategies can vary by dataset, as different data types might require different augmentations to facilitate model training. However, the design of DA policies has been largely decided by the human experts with domain knowledge, which is considered to be highly subjective and error-prone. To mitigate such problem, a novel direction is to automatically learn the image augmentation policies from the given dataset using Automated Data Augmentation (AutoDA) techniques. The goal of AutoDA models is to find the optimal DA policies that can maximize the model performance gains. This survey discusses the underlying reasons of the emergence of AutoDA technology from the perspective of image classification. We identify three key components of a standard AutoDA model: a search space, a search algorithm and an evaluation function. Based on their architecture, we provide a systematic taxonomy of existing image AutoDA approaches. This paper presents the major works in AutoDA field, discussing their pros and cons, and proposing several potential directions for future improvements.

CAD-based data augmentation and transfer learning empowers part classification in manufacturing

Especially in manufacturing systems with small batches or customized products, as well as in remanufacturing and recycling facilities, there is a wide variety of part types that may be previously unseen. It is crucial to accurately identify these parts based on their type for traceability or sorting purposes. One approach that has shown promising results for this task is deep learning–based image classification, which can classify a part based on its visual appearance in camera images. However, this approach relies on large labeled datasets of real-world images, which can be challenging to obtain, especially for parts manufactured for the first time or whose appearance is unknown. To overcome this challenge, we propose generating highly realistic synthetic images based on photo-realistically rendered computer-aided design (CAD) data. Using this commonly available source, we aim to reduce the manual effort required for data generation and preparation and improve the classification performance of deep learning models using transfer learning. In this approach, we demonstrate the creation of a parametric rendering pipeline and show how it can be used to train models for a 30-class classification problem with typical engineering parts in an industrial use case. We also demonstrate how our method’s entropy gain improves the classification performance in various deep image classification models.