Learning-based Classification Methods Research Articles

Raman spectroscopy for esophageal tumor diagnosis and delineation using machine learning and the portable Raman spectrometer

Esophageal cancer is one of the leading causes of cancer-related deaths worldwide. The identification of residual tumor tissues in the surgical margin of esophageal cancer is essential for the treatment and prognosis of cancer patients. But the current diagnostic methods, either pathological frozen section or paraffin section examination, are laborious, time-consuming, and inconvenient. Raman spectroscopy is a label-free and non-invasive analytical technique that provides molecular information with high specificity. Here, we report the use of a portable Raman system and machine learning algorithms to achieve accurate diagnosis of esophageal tumor tissue in surgically resected specimens. We tested five machine learning-based classification methods, including k-Nearest Neighbors, Adaptive Boosting, Random Forest, Principal Component Analysis-Linear Discriminant Analysis, and Support Vector Machine (SVM). Among them, SVM shows the highest accuracy (88.61 %) in classifying the esophageal tumor and normal tissues. The portable Raman system demonstrates robust measurements with an acceptable focal plane shift of up to 3 mm, which enables large-area Raman mapping on resected tissues. Based on this, we finally achieve successful Raman visualization of tumor boundaries on surgical margin specimens, and the Raman measurement time is less than 5 min. This work provides a robust, convenient, accurate, and cost-effective tool for the diagnosis of esophageal cancer tumors, advancing toward Raman-based clinical intraoperative applications.

Identification of cloud-to-ground lightning and intra-cloud lightning based on their radiated electric field signatures using different types of neural networks and machine learning classifiers

The scope of this paper is to test the efficiency of three different neural networks in the area of cloud-to-ground/intra-cloud lightning classification using a not very large data set of electric field data. Effective neural network learning verified for a small lightning event dataset but with whole 2-seconds particular record length is an important feature of fast and efficient procedures implemented into lightning location systems among which discrimination between cloud-to-ground and intra-cloud lightning is one of the most important task. Measurement data obtained at the Lightning Observatory in Rzeszow were used for the analysis. About 100 extremely low/medium frequency bandwidth electric field of cloud-to ground lightning registrations and the same number of intra-cloud lightning events were taken into account. During the study 81% of those registrations were dedicated for learning of neural networks while the remaining 19% were put into the testing set. All data were preselected manually with consideration of the electric field variation and information from the lightning location system database. Lightning events were selected uniformly with the well-pronounced presence of their basic components and reported distance to the registration station which ensured a wide variation of expected lightning electric field signatures. The multilayer perceptron neural network, the radial basis function neural network and the convolutional neural network were tested and compared in different configurations while distinguishing lightning. The results were compared with other conventional classification approaches, such as traditional machine learning methods as well as one a little more contemporary architecture, the long short-term memory neural network. The multilayer perceptron neural network achieved the best detection accuracy overall. Presented research is an extension of lightning identification studies available so far mainly based on the convolutional neural network by the results achieved using the multilayer perceptron neural network, the radial basis function neural network, the long short-term memory neural network and several machine learning-based classification methods.

Event recognition method based on feature synthesizing for a zero-shot intelligent distributed optical fiber sensor

Phase-sensitive optical time domain reflectometer (Φ-OTDR) is an emergent distributed optical sensing system with the advantages of high localization accuracy and high sensitivity. It has been widely used for intrusion identification, pipeline monitoring, under-ground tunnel monitoring, etc. Deep learning-based classification methods work well for Φ-OTDR event recognition tasks with sufficient samples. However, the lack of training data samples is sometimes a serious problem for these data-driven algorithms. This paper proposes a novel feature synthesizing approach to solve this problem. A mixed class approach and a reinforcement learning-based guided training method are proposed to realize high-quality feature synthesis. Experiment results in the task of eight event classifications, including one unknown class, show that the proposed method can achieve an average classification accuracy of 42% for the unknown class and obtain its event type, meanwhile achieving a 74% average overall classification accuracy. This is 29% and 7% higher, respectively, than those of the ordinary instance synthesizing method. Moreover, this is the first time that the Φ-OTDR system can recognize a specific event and tell its event type without collecting its data sample in advance.

Dilated Spectral–Spatial Gaussian Transformer Net for Hyperspectral Image Classification

In recent years, deep learning-based classification methods for hyperspectral images (HSIs) have gained widespread popularity in fields such as agriculture, environmental monitoring, and geological exploration. This is owing to their ability to automatically extract features and deliver outstanding performance. This study provides a new Dilated Spectral–Spatial Gaussian Transformer Net (DSSGT) model. The DSSGT model incorporates dilated convolutions as shallow feature extraction units, which allows for an expanded receptive field while maintaining computational efficiency. We integrated transformer architecture to effectively capture feature relationships and generate deep fusion features, thereby enhancing classification accuracy. We used consecutive dilated convolutional layers to extract joint low-level spectral–spatial features. We then introduced Gaussian Weighted Pixel Embedding blocks, which leverage Gaussian weight matrices to transform the joint features into pixel-level vectors. By combining the features of each pixel with its neighbouring pixels, we obtained pixel-level representations that are more expressive and context-aware. The transformed vector matrix was fed into the transformer encoder module, enabling the capture of global dependencies within the input data and generating higher-level fusion features with improved expressiveness and discriminability. We evaluated the proposed DSSGT model using five hyperspectral image datasets through comparative experiments. Our results demonstrate the superior performance of our approach compared to those of current state-of-the-art methods, providing compelling evidence of the DSSGT model’s effectiveness.

BoostXML: Gradient Boosting for Extreme Multilabel Text Classification With Tail Labels.

Multilabel learning involving hundreds of thousands or even millions of labels is referred to as extreme multilabel learning (XML), in which the labels often follow a power-law distribution with the majority occurring in very few data points as tail labels. Recent years have witnessed the intensive use of deep-learning methods for high-performance XML, but they are typically optimized for the head labels with abundant training instances and less consider the performance on tail labels, which, however, like the needles in haystacks, are often the focus of attention in real-life applications. In light of this, we present BoostXML, a deep learning-based XML method for extreme multilabel text classification, enhanced greatly by gradient boosting. In BoostXML, we pay more attention to tail labels in each Boosting Step by optimizing the residual mostly from unfitted training instances with tail labels. A Corrective Step is further proposed to avoid the mismatching between the text encoder and weak learners during optimization, which reduces the risk of falling into local optima and improves model performance. A Pretraining Step is also introduced in the initial stage of BoostXML to avoid exorbitant bias to tail labels. Extensive experiments on five benchmark datasets with state-of-the-art baselines demonstrate the advantage of BoostXML in tail-label prediction.

Multiscale Feature Extraction by Using Convolutional Neural Network: Extraction of Objects from Multiresolution Images of Urban Areas

The emergence of deep learning-based classification methods has led to considerable advancements and remarkable performance in image recognition. This study introduces the Multiscale Feature Convolutional Neural Network (MSFCNN) for the extraction of complex urban land cover data, with a specific emphasis on buildings and roads. MSFCNN is employed to extract multiscale features from three distinct image types—Unmanned Aerial Vehicle (UAV) images, high-resolution satellite images (HR), and low-resolution satellite images (LR)—all collected within the Fengshan District of Kaohsiung, Taiwan. The model in this study demonstrated remarkable accuracy in classifying two key land cover categories. Its success in extracting multiscale features from different image resolutions. In the case of UAV images, MSFCNN achieved an accuracy rate of 91.67%, with a Producer’s Accuracy (PA) of 93.33% and a User’s Accuracy (UA) of 90.0%. Similarly, the model exhibited strong performance with HR images, yielding accuracy, PA, and UA values of 92.5%, 93.33%, and 91.67%, respectively. These results closely align with those obtained for LR imagery, which achieved respective accuracy rates of 93.33%, 95.0%, and 91.67%. Overall, the MSFCNN excels in the classification of both UAV and satellite images, showcasing its versatility and robustness across various data sources. The model is well suited for the task of updating cartographic data related to urban buildings and roads.

Artificial intelligence-based solutions for coffee leaf disease classification

Coffee is one of the most widely consumed beverages and the quantity and quality of coffee beans depend significantly on the health and condition of coffee plants, particularly their leaves. The automation of coffee leaf disease classification using AI is an essential need, providing not only economic benefits but also contributing to environmental conservation and creating better conditions for sustainable coffee cultivation. Through the application of AI, early disease detection is facilitated, thereby reducing pest and disease control costs, minimizing crop losses, increasing coffee productivity and product quality, and promoting environmental preservation. Many studies have proposed AI algorithms for coffee disease classification. However, numerous algorithms employ classical algorithms, while some utilize deep learning, the current state-of-the-art in computer vision. The challenge lies in the fact that when using deep learning, a substantial amount of data is required for training. The design of deep learning architectures to enhance model accuracy while still working with a small training dataset remains an area of ongoing research. In this study, we propose deep learning-based method for coffee leaf disease classification. We propose the combination of different deep convolutional neural networks to further improve overall classification performance. Early and late fusion have been conducted to evaluate the effectiveness of the pre-trained model. Our experimental results demonstrate that the ensemble method outperforms single-model approaches, achieving high accuracy and precision in BRACOL coffee disease leaf.

Predicting the Distribution of Mangrove Forests through Random Forest Algorithm by Combining Spectral Features and DEM Data

Using satellite data and machine learning-based classification methods for mangrove forest identification has gained popularity due to their effectiveness in producing high accuracy. Therefore, this research developed a random forest classification model using various dataset combinations representing spectral and topographical aspects to identify mangrove forests in Segara Anakan, Cilacap. We used the Sentinel-2 imagery acquired in 2022. Two types of digital elevation model (DEM) data were utilized, namely the National Digital Elevation Model (DEMNAS) and Multi-Error-Removed Improved-Terrain DEM (MERIT DEM). Another critical dataset used is the Normalized Difference Moisture Index (NDMI), derived from the ratio between near-infrared (NIR) and shortwave-infrared (SWIR) bands. The classification method used is the random forest algorithm on six different dataset combinations, including IMAGE, IMAGE+NDMI, IMAGE+DEMNAS, IMAGE+MERIT DEM, IMAGE+NDMI+DEMNAS, and IMAGE+NDMI+ MERIT DEM. The results showed that the combination of Image+DEMNAS and Image+NDMI+DEMNAS datasets was able to identify the mangrove forests more optimally. Incorporating DEM data alongside IMAGE and NDMI datasets resulted in a remarkable level of accuracy in mangrove forest mapping, exceeding 90%. DEM data has a very important role in increasing the classification accuracy of mangrove forests using the random forest algorithm.

Optimizing Seagrass Detection in Lampung Selatan: Leveraging Machine Learning Algorithms and Sentinel-2A Imagery

Seagrass beds are one of the coastal ecosystems that play an important role in maintaining the stability of blue carbon. However, high community activities threaten the existence of seagrass beds themselves. South Lampung Regency is one of the areas with considerable seagrass potential but the availability of distribution and density data is still minimal. This research aims to identify and map seagrass density as a first step for seagrass management. The data used in this research is Sentinel 2A multispectral image. Machine learning-based classification methods used are random forest (RF) and support vector machine (SVM) because these algorithms have a good ability to distinguish objects based on their features. This study uses a 2-level classification scheme, where level 1 consists of land, shallow sea, and deep-sea classes. Level 2 is the shallow marine bottom benthic habitat class. The type of seagrass found in this area is Enhalus acoroides. Based on the results of the analysis, low, medium, and high-density classes were obtained with an area of low around 20.12 ha, medium around 34.67 ha and high around 320.12ha with a total area of 374.91ha. RF has a higher overall accuracy of 88.00% while SVM accuracy is 84.00% so it can be concluded that Sentinel 2A images can be used to detect seagrass meadows.

Photometric Classification of Evolved Massive Stars: Spectroscopic Verification and Validation

Dorn-Wallenstein et al. utilized a novel machine-learning technique to classify a large sample of evolved massive stars. This resulted in new classifications for ∼2550 objects. We wish to validate the efficiency of the Dorn-Wallenstein et al. machine classifier. To this end we obtained new observations of four stars identified by Dorn-Wallenstein et al., with a focus on verifying newly identified emission-line objects and evolved supergiants. We identified a previously unconfirmed Be star, TYC 3740-1791-1, using these data. We assigned spectral types to the two stars in our sample with sufficient signal-to-noise data. We then used Gaia DR3 BP/RP spectra to validate an additional 73 stars from Dorn-Wallenstein et al. Our classifications support the completeness and contamination reported by the authors and confirm the validity of using machine learning-based classification methods on massive stars in the era of big data.

Deep Learning-Based Method for Classification and Ripeness Assessment of Fruits and Vegetables

Food waste is a global concern and is the focus of this research. Currently, no method in the state of the art classifies multiple fruits and vegetables and their level of ripening. The objective of the study is to design and develop an intelligent system based on deep learning techniques to classify between types of fruits and vegetables, and also to evaluate the level of ripeness of some of them. The system consists of two models using the MobileNet V2 architecture. One algorithm is for the classification of 32 classes of fruits and vegetables, and another is for the determination of the ripeness of 6 classes of them. The overall intelligent system is the union of the two models, predicting first the class of fruit or vegetable and then its ripeness. The fruits and vegetables classification model achieved 97.86% accuracy, 98% precision, 98% recall, and 98% F1-score, while the ripeness assessment model achieved 100% accuracy, 98% precision, 99% recall, and 99% F1-score. According to the results, the proposed system is able to classify between types of fruits and vegetables and evaluate their ripeness. To achieve the best performance indicators, it is necessary to obtain the appropriate hyperparameters for the artificial intelligence models, in addition to having an extensive database with well-defined classes.

An Overview of Hyperspectral Image Classification by Data-driven Deep Learning

Hyperspectral imaging (HSI) in remote sensing is gaining significant attention due to its complexity, posing challenges for conventional machine learning in achieving accurate classification. The inherent nonlinear relationship between captured spectral information and materials further complicates hyperspectral imaging. Deep learning has emerged as an effective tool for feature extraction, finding widespread applications in image processing tasks. Motivated by its success, this survey integrates deep learning into hyperspectral imaging (HSI) classification, demonstrating commendable performance. The paper systematically reviews existing literature, providing a comparative analysis of strategies. Primary challenges in HSI classification for traditional methods are outlined, emphasizing the advantages of deep learning. Our framework categorizes works into three types: spectral-feature networks, spatial-feature networks, and spectral-spatial-feature networks, offering a comprehensive review of recent achievements and diverse approaches. Considering limited training samples in remote sensing and substantial data requirements for deep networks, strategies to enhance classification performance are presented, offering valuable insights for future studies. Experiments apply representative deep learning-based classification methods to real HSIs, providing practical validation. The survey contributes to understanding the current landscape in deep learning-based HSI classification and lays a foundation for future research in this evolving field.

A Multimodal Transfer Learning Method for Classifying Images of Celestial Point Sources

A large fraction of celestial objects exhibit point shapes in CCD images, such as stars and QSOs, which contain less information due to their few pixels. Point source classification based solely on image data may lead to low accuracy. To address this challenge, this paper proposes a Multi-modal Transfer Learning-based classification method for celestial objects with point shape images. Considering that spectral data possess rich features and that there is a correlation between spectral data and image data, the proposed approach fully utilizes the knowledge gained from celestial spectral data and transfers it to the original image-based classification, enhancing the accuracy of classifying stars and QSOs. Initially, a one-dimensional residual network is employed to extract a 128-dimensional spectral feature vector from the original 3700-dimensional spectral data. This spectral feature vector captures important features of the celestial object. The Generative Adversarial Network is then utilized to generate a simulated spectral vector of 128 dimensions, which corresponds to the celestial object image. By generating simulated spectral vectors, data from two modals (spectral and image) for the same celestial object are available, enriching the input features of the model. In the upcoming multimodal classification model, we only require the images of celestial objects along with their corresponding simulated spectral data, and we no longer need real spectral data. With the assistance of spectral data, the proposed method alleviates the above disadvantages of the original image-based classification method. Remarkably, our method has improved the F1-score from 0.93 to 0.9777, while reducing the error rate in classification by 40%. These enhancements significantly increase the classification accuracy of stars and QSOs, providing strong support for the classification of celestial point sources.

Plant Image Classification with Nonlinear Motion Deblurring Based on Deep Learning

Despite the significant number of classification studies conducted using plant images, studies on nonlinear motion blur are limited. In general, motion blur results from movements of the hands of a person holding a camera for capturing plant images, or when the plant moves owing to wind while the camera is stationary. When these two cases occur simultaneously, nonlinear motion blur is highly probable. Therefore, a novel deep learning-based classification method applied on plant images with various nonlinear motion blurs is proposed. In addition, this study proposes a generative adversarial network-based method to reduce nonlinear motion blur; accordingly, the method is explored for improving classification performance. Herein, experiments are conducted using a self-collected visible light images dataset. Evidently, nonlinear motion deblurring results in a structural similarity index measure (SSIM) of 73.1 and a peak signal-to-noise ratio (PSNR) of 21.55, whereas plant classification results in a top-1 accuracy of 90.09% and F1-score of 84.84%. In addition, the experiment conducted using two types of open datasets resulted in PSNRs of 20.84 and 21.02 and SSIMs of 72.96 and 72.86, respectively. The proposed method of plant classification results in top-1 accuracies of 89.79% and 82.21% and F1-scores of 84% and 76.52%, respectively. Thus, the proposed network produces higher accuracies than the existing state-of-the-art methods.

Prompt-based for Low-Resource Tibetan Text Classification

Text classification is a critical and foundational task in Tibetan natural language processing, it plays a crucial role in various applications, such as sentiment analysis and information extraction. However, the limited availability of annotated data poses a significant challenge to Tibetan natural language processing. This paper proposes a prompt learning-based method for low-resource Tibetan text classification to overcome this challenge. This method utilizes pre-trained language models to learn text representation and generation capabilities on a large-scale unsupervised Tibetan corpus, enabling few-shot Tibetan text classification. Experimental results demonstrate that the proposed method significantly improves the performance of Tibetan text classification in low-resource scenarios. This work provides a new research idea and method for low-resource language processing, such as Tibetan natural language processing. Hopefully, it will inspire subsequent work on low-resource language processing.

Application of Machine Learning to Performance Assessment for a Class of PID-Based Control Systems

In this paper, a novel machine learning derived control performance assessment (CPA) classification system is proposed. It is dedicated for a wide class of PID-based control industrial loops with processes exhibiting dynamical properties close to second order plus delay time (SOPDT). The proposed concept is very general and easy to configure to distinguish between acceptable and poor closed loop performance. This approach allows for determining the best (but also robust and practically achievable) closed loop performance based on very popular and intuitive closed loop quality factors. Training set can be automatically derived off-line using a number of different, diverse control performance indices (CPIs) used as discriminative features of the assessed control system. The proposed extended set of CPIs is discussed with comprehensive performance assessment of different machine learning based classification methods and practical application of the suggested solution. As a result, a general-purpose CPA system is derived that can be immediately applied in practice without any preliminary or additional learning stage during normal closed loop operation. It is verified by practical application to assess the control system for a laboratory heat exchange and distribution setup.

Hygeia: A Multilabel Deep Learning-Based Classification Method for Imbalanced Electrocardiogram Data.

Electrocardiogram (ECG) is a common diagnostic indicator of heart disease in hospitals. Because of the low price and noninvasiveness of ECG diagnosis, it is widely used for prescreening and physical examination of heart diseases. In several studies on ECG analysis, only rough diagnoses are made to determine whether ECGs are abnormal or on a few kinds of ECG. In actual scenarios, doctors must analyze ECG samples in detail, which is a multilabel classification problem. Herein, we propose Hygeia, a multilabel deep learning-based ECG classification method that can analyze and classify 55 types of ECG. First, a guidance model is constructed to transform the multilabel classification problem into multiple interrelated two-classification models. This method ensures the good performance of each ECG analysis model, and the relationship between various types of ECG can be used in the analysis. The imbalance of samples in ECG datasets makes it difficult to analyze abnormal ECGs with high sensitivity and accuracy. We used data generation and mixed sampling methods for 11 ECG types with imbalanced problems to improve the average accuracy, sensitivity, F1 value, and accuracy from 87.74%, 43.11%, 0.3929, and 0.3929, to 92.68%, 96.92, 0.9287, and 99.47%, respectively. The average accuracy, sensitivity, F1 value, and accuracy of 44 of the 55 tags of the abnormal ECG analysis model are 99.69%, 95.81%, 0.9758, and 99.72%, respectively.

A two-stream network with complementary feature fusion for pest image classification

Pests are diverse and the available datasets often contain an uneven number of examples for different pests (a.k.a., the long-tail distribution). This poses a great challenge to learning-based classification methods, especially deep networks, and often leads to degraded performance, especially for the minority (tail) classes. This paper presents a deep learning integration architecture based on decoupling training and fusion learning, which integrates different models with complementary performance on pest datasets with a long-tailed distribution to improve the overall classification performance of pests. A deep neural network is designed that fuses two complementary deep learning models at the feature level, which consists of a convolution neural network (ConvNeXt) and a Swin Transformer model for decoupling training. Experiments are conducted using three datasets (d0, insect, and IP102), and evaluation on accuracy, recall, and F1-Score is reported. For the large-scale pest dataset with long-tailed distribution IP102, the accuracy achieves 76.1%, which outperforms the state-of-the-art methods. In addition, the accuracy for d0 and insect datasets are 98.5% and 92.3%, respectively.

Online classification of soybean seeds based on deep learning

To quickly evaluate soybean quality, we proposed a deep learning-based method for online classification of soybean seeds. Firstly, images of soybean seeds with uneven illumination were segmented based on the multi-scale Retinex with color restoration (MSRCR). Then, a convolutional neural network (CNN) was constructed to achieve soybean seed four-classification with appropriate parameters. The F-score of the normal, damaged, abnormal, and non-classifiable soybeans reached about 95.97%, 97.41%, 97.25%, and 96.14%, respectively. Finally, the method was successfully applied in NVIDIA Jetson TX2 with an accuracy of 95.63% and an average classification time of 4.92 ms for a soybean seed, which can meet the requirement of online soybean quality assessment.

Automated Sagittal Skeletal Classification of Children Based on Deep Learning.

Malocclusions are a type of cranio-maxillofacial growth and developmental deformity that occur with high incidence in children. Therefore, a simple and rapid diagnosis of malocclusions would be of great benefit to our future generation. However, the application of deep learning algorithms to the automatic detection of malocclusions in children has not been reported. Therefore, the aim of this study was to develop a deep learning-based method for automatic classification of the sagittal skeletal pattern in children and to validate its performance. This would be the first step in establishing a decision support system for early orthodontic treatment. In this study, four different state-of-the-art (SOTA) models were trained and compared by using 1613 lateral cephalograms, and the best performance model, Densenet-121, was selected was further subsequent validation. Lateral cephalograms and profile photographs were used as the input for the Densenet-121 model, respectively. The models were optimized using transfer learning and data augmentation techniques, and label distribution learning was introduced during model training to address the inevitable label ambiguity between adjacent classes. Five-fold cross-validation was conducted for a comprehensive evaluation of our method. The sensitivity, specificity, and accuracy of the CNN model based on lateral cephalometric radiographs were 83.99, 92.44, and 90.33%, respectively. The accuracy of the model with profile photographs was 83.39%. The accuracy of both CNN models was improved to 91.28 and 83.98%, respectively, while the overfitting decreased after addition of label distribution learning. Previous studies have been based on adult lateral cephalograms. Therefore, our study is novel in using deep learning network architecture with lateral cephalograms and profile photographs obtained from children in order to obtain a high-precision automatic classification of the sagittal skeletal pattern in children.