Semi-supervised Learning Framework Research Articles

Deep collaborative learning with class-rebalancing for semi-supervised change detection in SAR images

Deep learning reveals excellent potential for accomplishing change detection in SAR imagery. Yet, it suffers from the problem of requiring large amounts of labeled samples, whilst labeling SAR imagery for change detection requires experts to label individual images at the pixel level, which is extremely tedious and time-consuming. Also, sample imbalance continues to present a serious challenge for the existing change detection techniques. To tackle these problems, in this study, a Deep Collaborative semi-supervised learning Framework with Class-Rebalancing (DCF-CRe) is proposed for SAR imagery change detection, by exploiting Convolutional Neural Network (CNN) and deep clustering. In particular, a Siamese Difference Fusion Network (SDFNet) is devised to implement change detection while effectively reducing the information loss due to the generation of difference images and highlighting features of the changed regions.In so doing, only a tiny batch of labeled samples is utilized to train SDFNet in order to obtain predicted change map and deep features. In addition, the Approximate Rank-Order Clustering (AROC) algorithm is employed to cluster the deep features, generating pseudo-labels for abundant unlabeled samples. DCF-CRe is then applied to select appropriate pseudo-labels and to add labeled samples to train SDFNet. Experimental results evaluated on six challenging datasets show that this proposed approach can achieve performance superior to state-of-the-art change detection methods for SAR imagery.

DML-PL: Deep metric learning based pseudo-labeling framework for class imbalanced semi-supervised learning

Traditional class imbalanced learning algorithms require training data to be labeled, whereas semi-supervised learning algorithms assume that the class distribution is balanced. However, class imbalance and insufficient labeled data problems often coexist in practical real-world applications. Currently, most existing class-imbalanced semi-supervised learning methods tackle these two problems separately, resulting in the trained model biased towards majority classes that have more data samples. In this study, we propose a deep metric learning based pseudo-labeling (DML-PL) framework that tackles both problems simultaneously for class-imbalanced semi-supervised learning. The proposed DML-PL framework comprises three modules: Deep Metric Learning, Pseudo-Labeling and Network Fine-tuning. An iterative self-training strategy is used to train the model multiple times. For each time of training, Deep Metric Learning trains a deep metric network to learn compact feature representations of labeled and unlabeled data. Pseudo-Labeling then generates reliable pseudo-labels for unlabeled data through labeled data clustering with nearest neighbors selection. Finally, Network Fine-tuning fine-tunes the deep metric network to generate better pseudo-labels in the subsequent training. The training ends when all the unlabeled data are pseudo-labeled. The proposed framework achieved state-of-the-art performance on the long-tailed CIFAR-10, CIFAR-100, and ImageNet127 benchmark datasets compared with baseline models.

MTSS-AAE: Multi-task semi-supervised adversarial autoencoding for COVID-19 detection based on chest X-ray images

Efficient diagnosis of COVID-19 plays an important role in preventing the spread of the disease. There are three major modalities to diagnose COVID-19 which include polymerase chain reaction tests, computed tomography scans, and chest X-rays (CXRs). Among these, diagnosis using CXRs is the most economical approach; however, it requires extensive human expertise to diagnose COVID-19 in CXRs, which may deprive it of cost-effectiveness. The computer-aided diagnosis with deep learning has the potential to perform accurate detection of COVID-19 in CXRs without human intervention while preserving its cost-effectiveness. Many efforts have been made to develop a highly accurate and robust solution. However, due to the limited amount of labeled data, existing solutions are evaluated on a small set of test dataset. In this work, we proposed a solution to this problem by using a multi-task semi-supervised learning (MTSSL) framework that utilized auxiliary tasks for which adequate data is publicly available. Specifically, we utilized Pneumonia, Lung Opacity, and Pleural Effusion as additional tasks using the ChesXpert dataset. We illustrated that the primary task of COVID-19 detection, for which only limited labeled data is available, can be improved by using this additional data. We further employed an adversarial autoencoder (AAE), which has a strong capability to learn powerful and discriminative features, within our MTSSL framework to maximize the benefit of multi-task learning. In addition, the supervised classification networks in combination with the unsupervised AAE empower semi-supervised learning, which includes a discriminative part in the unsupervised AAE training pipeline. The generalization of our framework is improved due to this semi-supervised learning and thus it leads to enhancement in COVID-19 detection performance. The proposed model is rigorously evaluated on the largest publicly available COVID-19 dataset and experimental results show that the proposed model attained state-of-the-art performance.

A framework for lossy compression and recovery of battery cell voltage

High-quality historical data of battery is important for state estimation and management. However, limited by bandwidth of 4G network and storage capacity, the cloud only receives low-frequency (LF) data and a few high-frequency (HF) signals. This paper proposes a model-based method to transfer adequate battery data between vehicle and cloud at a low cost. Firstly, a training dataset is built from real-world vehicle data. Then, a multi-task learning (MTL) model under semi-supervised learning (SSL) framework is proposed to learn HF voltage representation of each battery cell. SSL framework uses a large number of unlabeled low-frequency voltages to improve the voltage frequency recovery accuracy. MTL framework enables the model to focus on battery aging and solve the problem of target domain drifting over time. Finally, both real-world vehicle and experimental data are used to compare the results of different methods on the voltage recovering task. The results show that the proposed method can reduce the average voltage recovery error to less than 7mV at various driving conditions.

A Novel Semi-Supervised Long-Tailed Learning Framework With Spatial Neighborhood Information for Hyperspectral Image Classification

Deep learning technologies have been successfully applied to hyperspectral (HS) image classification with remarkable performance. However, compared with traditional machine learning methods, neural networks usually need more data. In remote sensing (RS) research, obtaining a large number of labeled HS data is very difficult and expensive work. Simultaneously, the distribution of feature information is bound to be unbalanced, and tends to conform to the long tail. At present, the neighborhood information of unlabeled samples is usually ignored in HS image classification tasks based on semi-supervised learning. In this letter, we propose a new semi-supervised long-tail learning framework based on spatial neighborhood information (SLN-SNI), which can complete the HS image classification task under unbalanced small sample data. Specifically, a new semi-supervised learning strategy is proposed. On this basis, a new method to determine the label of unlabeled samples based on spatial neighborhood information (SNI) is proposed. The coarse classification results divided into three situations are judged again, and the accuracy of pseudo labels is improved. The performance of the proposed method is tested on three public HS image datasets. Compared with the current advanced methods have achieved a certain improvement.

A semi-supervised multi-task learning framework for cancer classification with weak annotation in whole-slide images.

Cancer region detection (CRD) and subtyping are two fundamental tasks in digital pathology image analysis. The development of data-driven models for CRD and subtyping on whole-slide images(WSIs) would mitigate the burden of pathologists and improve their accuracy in diagnosis. However, the existing models are facing two major limitations. Firstly, they typically require large-scale datasets with precise annotations, which contradicts with the original intention of reducing labor effort. Secondly, for the subtyping task, the non-cancerous regions are treated as the same as cancerous regions within a WSI, which confuses a subtyping model in its training process. To tackle the latter limitation, the previous research proposed to perform CRD first for ruling out the non-cancerous region, then train a subtyping model based on the remaining cancerous patches. However, separately training ignores the interaction of these two tasks, also leads to propagating the error of the CRD task to the subtyping task. To address these issues and concurrently improve the performance on both CRD and subtyping tasks, we propose a semi-supervised multi-task learning(MTL) framework for cancer classification. Our framework consists of a backbone feature extractor, two task-specific classifiers, and a weight control mechanism. The backbone feature extractor is shared by two task-specific classifiers, such that the interaction of CRD and subtyping tasks can be captured. The weight control mechanism preserves the sequential relationship of these two tasks and guarantees the error back-propagation from the subtyping task to the CRD task under the MTL framework. We train the overall framework in a semi-supervised setting, where datasets only involve small quantities of annotations produced by our minimal point-based(min-point) annotation strategy. Extensive experiments on four large datasets with different cancer types demonstrate the effectiveness of the proposed framework in both accuracy and generalization.

Semi-Supervised Pixel Contrastive Learning Framework for Tissue Segmentation in Histopathological Image.

Accurate tissue segmentation in histopathological images is essential for promoting the development of precision pathology. However, the size of the digital pathological image is great, which needs to be tiled into small patches containing limited semantic information. To imitate the pathologist's diagnosis process and model the semantic relation of the whole slide image, We propose a semi-supervised pixel contrastive learning framework (SSPCL) which mainly includes an uncertainty-guided mutual dual consistency learning module (UMDC) and a cross image pixel-contrastive learning module (CIPC). The UMDC module enables efficient learning from unlabeled data through mutual dual-consistency and consensus-based uncertainty. The CIPC module aims at capturing the cross-patch semantic relationship by optimizing a contrastive loss between pixel embeddings. We also propose several novel domain-related sampling methods by utilizing the continuous spatial structure of adjacent image patches, which can avoid the problem of false sampling and improve the training efficiency. In this way, SSPCL significantly reduces the labeling cost on histopathological images and realizes the accurate quantitation of tissues. Extensive experiments on three tissue segmentation datasets demonstrate the effectiveness of SSPCL, which outperforms state-of-the-art up to 5.0% in mDice.

Accelerating Semi-Supervised Text Classification by K-Way Projecting Networks

The state of the art semi-supervised learning framework has greatly shown its potential in making deep and complex language models such as BERT highly effective for text classification tasks when labeled data is limited. However, the large size and low inference speed of such models may hinder their application on resources-limited or real-time use cases. In this paper, we propose a new approach in semi-supervised learning framework to distill large complex teacher model into a fairly lightweight student model which has the ability of acquiring knowledge from different layers of teacher with the usage of <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$K$ </tex-math></inline-formula> -way projecting networks. Across four English datasets in text classification benchmarks and one dataset collected from an Chinese online course, our experiment shows that this student model achieves comparable results with the state of the art Transformer-based semi-supervised text classification methods, while using only 0.156MB parameters and having an inference speed 785 times faster than the teacher model.

IDN: inner-class dense neighbours for semi-supervised learning-based remote sensing scene classification

ABSTRACT In remote sensing scene classification, manual annotation is time-consuming and expensive to collect, which limits the effective deployment of supervised classification networks. Semi-supervised learning (SSL), which can benefit from unlabelled and labelled samples simultaneously, attracts much attention. In this letter, we introduce a cluster group construction method for semi-supervised high-resolution remote sensing (HRRS) scene classification. The low-density separation assumption, which assumes that the decision boundary of a classification network should lie in the low-density region, is a vital assumption to guide the construction of clusters. Following this, we propose to construct inner-class dense neighbours (IDN) for the semi-supervised classification framework. The proposed IDN can increase the density of clusters for a better determination of high-density regions to help the network training in the SSL framework. The experiments on the NWPU-RESIS45 and RSSCN7 datasets demonstrate the effectiveness of our proposed method, which show that the performance of the classification network with our proposed IDN has a prominent classification accuracy improvement.

Deep semi-supervised multiple instance learning with self-correction for DME classification from OCT images.

Supervised deep learning has achieved prominent success in various diabetic macular edema (DME) recognition tasks from optical coherence tomography (OCT) volumetric images. A common problematic issue that frequently occurs in this field is the shortage of labeled data due to the expensive fine-grained annotations, which increases substantial difficulty in accurate analysis by supervised learning. The morphological changes in the retina caused by DME might be distributed sparsely in B-scan images of the OCT volume, and OCT data is often coarsely labeled at the volume level. Hence, the DME identification task can be formulated as a multiple instance classification problem that could be addressed by multiple instance learning (MIL) techniques. Nevertheless, none of previous studies utilize unlabeled data simultaneously to promote the classification accuracy, which is particularly significant for a high quality of analysis at the minimum annotation cost. To this end, we present a novel deep semi-supervised multiple instance learning framework to explore the feasibility of leveraging a small amount of coarsely labeled data and a large amount of unlabeled data to tackle this problem. Specifically, we come up with several modules to further improve the performance according to the availability and granularity of their labels. To warm up the training, we propagate the bag labels to the corresponding instances as the supervision of training, and propose a self-correction strategy to handle the label noise in the positive bags. This strategy is based on confidence-based pseudo-labeling with consistency regularization. The model uses its prediction to generate the pseudo-label for each weakly augmented input only if it is highly confident about the prediction, which is subsequently used to supervise the same input in a strongly augmented version. This learning scheme is also applicable to unlabeled data. To enhance the discrimination capability of the model, we introduce the Student-Teacher architecture and impose consistency constraints between two models. For demonstration, the proposed approach was evaluated on two large-scale DME OCT image datasets. Extensive results indicate that the proposed method improves DME classification with the incorporation of unlabeled data and outperforms competing MIL methods significantly, which confirm the feasibility of deep semi-supervised multiple instance learning at a low annotation cost.

Semi-Supervised SAR ATR via Epoch- and Uncertainty-Aware Pseudo-Label Exploitation

Automatic target recognition (ATR) is a significant application scenario for Synthetic aperture radar (SAR) image interpretation. In recent years, deep-learning based SAR ATR approaches have made great progress, with the require of large amounts of labeled data for network training. However, the labeled SAR images are scarce, and the annotation of SAR images is expensive and time-consuming. In this paper, a novel semi-supervised learning (SSL) framework is proposed for SAR ATR, which effectively alleviates the need of labeled samples for network training, and allows for excavating the intrinsic semantic relationship information between samples belonging to different categories, greatly improving recognition performance. This method develops an epoch- and uncertainty-aware pseudo-label selection (EUAPS) mechanism, which takes advantage of the underutilized consistency between training epochs, and introduces the uncertainty estimates. EUAPS can select pseudo-label samples with high-confidence, allowing the network to perform well even when labels are extremely scarce. Furthermore, we propose a novel loss function, group relationship consistency loss (GRCL) that explicitly enforces the consistency of relations between different samples under perturbations. GRCL facilitates the network to learn rich sample relationship information and more discriminative features in sample group level, effectively reducing the recognition error rate. Extensive experiments were carried out on three public benchmark SAR datasets. And the experimental results illustrate that the proposed approach achieves the accuracy of 98.58% when 10 labeled samples per class and 88.65% when only 5 labeled samples per class respectively, outperforming the state-of-the-art SSL methods and demonstrating the superiority.

Source Free Semi-Supervised Transfer Learning for Diagnosis of Mental Disorders on fMRI Scans.

The high prevalence of mental disorders gradually poses a huge pressure on the public healthcare services. Deep learning-based computer-aided diagnosis (CAD) has emerged to relieve the tension in healthcare institutions by detecting abnormal neuroimaging-derived phenotypes. However, training deep learning models relies on sufficient annotated datasets, which can be costly and laborious. Semi-supervised learning (SSL) and transfer learning (TL) can mitigate this challenge by leveraging unlabeled data within the same institution and advantageous information from source domain, respectively. This work is the first attempt to propose an effective semi-supervised transfer learning (SSTL) framework dubbed S3TL for CAD of mental disorders on fMRI data. Within S3TL, a secure cross-domain feature alignment method is developed to generate target-related source model in SSL. Subsequently, we propose an enhanced dual-stage pseudo-labeling approach to assign pseudo-labels for unlabeled samples in target domain. Finally, an advantageous knowledge transfer method is conducted to improve the generalization capability of the target model. Comprehensive experimental results demonstrate that S3TL achieves competitive accuracies of 69.14%, 69.65%, and 72.62% on ABIDE-I, ABIDE-II, and ADHD-200 datasets, respectively. Furthermore, the simulation experiments also demonstrate the application potential of S3TL through model interpretation analysis and federated learning extension.

A Probabilistic Contrastive Framework for Semi-Supervised Learning

Semi-supervised learning is a common way that investigates how to improve performance of a visual learning model, while data annotation is far from sufficient. Recent works in semi-supervised deep learning have successfully applied consistency regularization, which encourages a model to maintain consistent predictions for different perturbed versions of an image. However, most of such methods ignore the category correlation of image features, especially when exploiting strong augmentation methods for unlabeled images. To address this problem, we propose PConMatch, a model that leverages a probabilistic contrastive learning framework to separate the features of strongly-augmented versions from different classes. A semi-supervised probabilistic contrastive loss is designed, which takes both labeled and unlabeled samples into account and develops an auxiliary module to generate a probability score to measure the model prediction confidence for each sample. Specifically, PConMatch first generates a pair of weakly-augmented versions for each labeled sample, and produces a weakly-augmented version and a corresponding pair of strongly-augmented versions for each unlabeled sample. Second, a probability score module is proposed to assign pseudo-labeling confidence scores to strongly-augmented unlabeled images. Finally, the probability score of each sample is further passed to the contrastive loss, combining with consistency regularization to enable the model to learn better feature representations. Extensive experiments on four publicly available image classification benchmarks demonstrate that the proposed approach achieves state-of-the-art performance in image classification. Several rigorous ablation studies are conducted to validate the effectiveness of the method.

Dynamic Super-Pixel Normalization for Robust Hyperspectral Image Classification

Deep neural networks (DNNs) have underpinned most of recent progress of hyperspectral image (HSI) classification. One premise of their success lies on the high image quality without noise corruption. However, due to the limitation of imaging sensor and imaging conditions, HSIs captured in practice inevitably suffer from random noise, which will degrade the generalization performance and robustness of most existing DNN based methods. In this study, we propose a dynamic super-pixel normalization based DNN for HSI classification, which can adaptively relieve the negative effect caused by various types of noise corruption and improve the generalization performance. To achieve this goal, we propose a dynamic super-pixel normalization module, for a given super-pixel which normalizes the inner pixel features using parameters dynamically generated based on themselves. By doing this, such a module enables adaptively restoring the similarity among pixels within the super-pixel corrupted by random noise through aligning their feature distribution, thus enhancing the generalization performance on noisy HSI. Moreover, it can be directly plugged into any other existing DNN architectures. To appropriately train the proposed DNN model, we further present a semi-supervised learning framework, which integrates the cross entropy loss and Kullback-Leibler divergence loss on labeled samples with the infomation entropy loss on the unlabeled samples for joint learning to well sidestep over-fitting. Experiments on three benchmark HSI classification datasets demonstrate the advantages of the proposed method over several state-of-the-art competitors in handling HSIs under different types of noise corruption.

Semi-Supervised Learning Framework for Crack Segmentation Based on Contrastive Learning and Cross Pseudo Supervision

Semi-Supervised Learning Framework for Crack Segmentation Based on Contrastive Learning and Cross Pseudo Supervision

SDCDNet: A Semi-Dual Change Detection Network Framework With Super-Weak Label for Remote Sensing Image

Most current change detection methods require a large amount of labeled data to train huge parameters. To break this limitation, this paper proposes a novel semi-supervised learning framework for remote sensing change detection, named a semi-dual change detection network (SDCDNet). The SDCDNet consists of a dual shared network and dual branching networks. The dual shared network is designed to exploit the full potential of the data, and the dual branching network is proposed to differentiate the kinds of annotated data and eliminate the disturbance between different types of data. In addition, the adaptive weighting module (AWM) enhances the features of weak branching, and the mask constraint module (MCM) is proposed to increase the ability of the network to extract foreground features. To solve the complex problem of data labeling, a patch-based weak label construction method is proposed to build super-weak labels. Experiments show that the proposed SDCDNet achieves excellent results on two remote sensing image change detection datasets.

Cost-Sensitive Laplacian Logistic Regression for Ship Detention Prediction

Port state control (PSC) is the last line of defense for substandard ships. During a PSC inspection, ship detention is the most severe result if the inspected ship is identified with critical deficiencies. Regarding the development of ship detention prediction models, this paper identifies two challenges: learning from imbalanced data and learning from unlabeled data. The first challenge, imbalanced data, arises from the fact that a minority of inspected ships were detained. The second challenge, unlabeled data, arises from the fact that in practice not all foreign visiting ships receive a formal PSC inspection, leading to a missing data problem. To address these two challenges, this paper adopts two machine learning paradigms: cost-sensitive learning and semi-supervised learning. Accordingly, we expand the traditional logistic regression (LR) model by introducing a cost parameter to consider the different misclassification costs of unbalanced classes and incorporating a graph regularization term to consider unlabeled data. Finally, we conduct extensive computational experiments to verify the superiority of the developed cost-sensitive semi-supervised learning framework in this paper. Computational results show that introducing a cost parameter into LR can improve the classification rate for substandard ships by almost 10%. In addition, the results show that considering unlabeled data in classification models can increase the classification rate for minority and majority classes by 1.33% and 5.93%, respectively.

A fault diagnosis framework for rotating machinery of marine equipment: A semi-supervised learning framework based on contractive stacked autoencoder

Rotating machinery is one of the key components of marine equipment. Due to the complex and harsh offshore environment, the health status of rotating machinery is more likely to be affected. Therefore, fault diagnosis is of great significance to normal operation and maintenance of rotating machinery in marine equipment. Traditional data-driven fault diagnosis tasks require massive label data for training, and it takes time and manpower to obtain enough label samples. At the same time, it is considered that the noise can interfere with the performance of the fault diagnosis framework. To overcome the above two defects, this paper proposes a fault diagnosis framework based on semi-supervised learning, where the contractive stacked autoencoder (CSA) and the classifier multilayer perceptron (MLP) extract features from unlabeled data and realize fault classification respectively. Compared with the Stacked Autoencoder (SAE)-MLP and Stacked Denoising Autoencoder (SDAE)-MLP frameworks, the proposed learning framework has better fault diagnosis accuracy and robustness.

Semi-supervised graph learning framework for apicomplexan parasite classification

Apicomplexan parasites cause diseases including malaria, toxoplasmosis, and babesiosis, affecting large parts of the world and hampering economic development to a considerable severity. Early recognition of pathogenic parasites can effectively constrain the spread of these infectious diseases. Recently, deep-learning-based computer-aided parasite recognition systems provide many practical tools for automated apicomplexan parasite recognition. However, existing research works are limited by the complicated and expensive manual annotating workload in supervised deep learning methods, which require many experts’ manually annotated samples. Moreover, they also ignore the significance of correlations among different parasite cells, which retain helpful information for identification. Therefore, this paper is the first to employ a semi-supervised learning strategy to reduce the manual labeling workload and introduce a hybrid graph learning algorithm into parasite recognition. Our approach has a superior parasite recognition capability with semi-supervised techniques and the graph-based model. Specifically, this research proposes a Semi-Supervised Graph Learning (SSGL) framework, composing CNN (Convolutional Neural Networks) feature embedding, learnable graph building, and graph feature learning components. To verify the performance of the SSGL model, the paper applies multiple evaluation indicators to appraise our model, and the approach can achieve high accuracy of 91.75%, AUC of 91.83%, the sensitivity of 91.75% and specificity of 97.25% with only a small amount of labeled data (20%). In conclusion, SSGL model provides high parasite recognition capability with minimal labels, greatly reduces the workload of relevant experts, and promotes the popularization of computer-aided parasite diagnosis tools.

Issues in Melanoma Detection: Semisupervised Deep Learning Algorithm Development via a Combination of Human and Artificial Intelligence.

Automatic skin lesion recognition has shown to be effective in increasing access to reliable dermatology evaluation; however, most existing algorithms rely solely on images. Many diagnostic rules, including the 3-point checklist, are not considered by artificial intelligence algorithms, which comprise human knowledge and reflect the diagnosis process of human experts. In this paper, we aimed to develop a semisupervised model that can not only integrate the dermoscopic features and scoring rule from the 3-point checklist but also automate the feature-annotation process. We first trained the semisupervised model on a small, annotated data set with disease and dermoscopic feature labels and tried to improve the classification accuracy by integrating the 3-point checklist using ranking loss function. We then used a large, unlabeled data set with only disease label to learn from the trained algorithm to automatically classify skin lesions and features. After adding the 3-point checklist to our model, its performance for melanoma classification improved from a mean of 0.8867 (SD 0.0191) to 0.8943 (SD 0.0115) under 5-fold cross-validation. The trained semisupervised model can automatically detect 3 dermoscopic features from the 3-point checklist, with best performances of 0.80 (area under the curve [AUC] 0.8380), 0.89 (AUC 0.9036), and 0.76 (AUC 0.8444), in some cases outperforming human annotators. Our proposed semisupervised learning framework can help with the automatic diagnosis of skin disease based on its ability to detect dermoscopic features and automate the label-annotation process. The framework can also help combine semantic knowledge with a computer algorithm to arrive at a more accurate and more interpretable diagnostic result, which can be applied to broader use cases.