Sample Labeling Research Articles

Transfer Learning with Attributes for Improving the Landslide Spatial Prediction Performance in Sample-Scarce Area Based on Variational Autoencoder Generative Adversarial Network

Owing to the complexity of obtaining the landslide inventory data, it is a challenge to establish a landslide spatial prediction model with limited labeled samples. This paper proposed a novel strategy, namely transfer learning with attributes (TLAs), to make good use of existing landslide inventory data, a strategy that is based on a variational autoencoder of a generative adversarial network (VAEGAN) for improving the landslide spatial prediction performance in sample-scarce areas. Different from transfer learning (TL), TLAs are pretraining the model with the data reconstructed by VAEGAN, so that the models learn in advance the landslide attributes of sample-scarce areas. Accordingly, a database containing a total of 986 landslides in three study areas with 14 landslide-influencing factors was established, and each of the three models, i.e., convolutional neural networks (CNNs), bidirectional long short-term memory (BiLSTM) and gated recurrent units (GRUs), was respectively selected as the feature extractor of the VAEGAN to reconstruct the data with attributes and the prediction model to generate the landslide susceptibility maps to investigate and validate the proposed TLA strategy. The experimental results showed that the TLA strategy increased the mean value of evaluators, such as the area under the receiver-operating characteristic (AUROC), F1-score, precision, recall and accuracy by about 2–7% compared with TL, results that indicated that the generated data have the attribute of specific study areas and the effectiveness of TLA strategy in sample-scare areas.

Graph-Based Domain Adaptation Few-Shot Learning for Hyperspectral Image Classification

Due to a lack of labeled samples, deep learning methods generally tend to have poor classification performance in practical applications. Few-shot learning (FSL), as an emerging learning paradigm, has been widely utilized in hyperspectral image (HSI) classification with limited labeled samples. However, the existing FSL methods generally ignore the domain shift problem in cross-domain scenes and rarely explore the associations between samples in the source and target domain. To tackle the above issues, a graph-based domain adaptation FSL (GDAFSL) method is proposed for HSI classification with limited training samples, which utilizes the graph method to guide the domain adaptation learning process in a uniformed framework. First, a novel deep residual hybrid attention network (DRHAN) is designed to extract discriminative embedded features efficiently for few-shot HSI classification. Then, a graph-based domain adaptation network (GDAN), which combines graph construction with domain adversarial strategy, is proposed to fully explore the domain correlation between source and target embedded features. By utilizing the fully explored domain correlations to guide the domain adaptation process, a domain invariant feature metric space is learned for few-shot HSI classification. Comprehensive experimental results conducted on three public HSI datasets demonstrate that GDAFSL is superior to the state-of-the-art with a small sample size.

Semi-Supervised Classification of PolSAR Images Based on Co-Training of CNN and SVM with Limited Labeled Samples.

Recently, convolutional neural networks (CNNs) have shown significant advantages in the tasks of image classification; however, these usually require a large number of labeled samples for training. In practice, it is difficult and costly to obtain sufficient labeled samples of polarimetric synthetic aperture radar (PolSAR) images. To address this problem, we propose a novel semi-supervised classification method for PolSAR images in this paper, using the co-training of CNN and a support vector machine (SVM). In our co-training method, an eight-layer CNN with residual network (ResNet) architecture is designed as the primary classifier, and an SVM is used as the auxiliary classifier. In particular, the SVM is used to enhance the performance of our algorithm in the case of limited labeled samples. In our method, more and more pseudo-labeled samples are iteratively yielded for training through a two-stage co-training of CNN and SVM, which gradually improves the performance of the two classifiers. The trained CNN is employed as the final classifier due to its strong classification capability with enough samples. We carried out experiments on two C-band airborne PolSAR images acquired by the AIRSAR systems and an L-band spaceborne PolSAR image acquired by the GaoFen-3 system. The experimental results demonstrate that the proposed method can effectively integrate the complementary advantages of SVM and CNN, providing overall classification accuracy of more than 97%, 96% and 93% with limited labeled samples (10 samples per class) for the above three images, respectively, which is superior to the state-of-the-art semi-supervised methods for PolSAR image classification.

Research on Shaft Current Damage Identification of Variable Condition Motor Bearings Based on Multiscale Feature Label Propagation and Manifold Metric Transfer

The current damage is the most stubborn and difficult fault of high-power motor bearings because its vibration characteristics are easily confused with those of ordinary bearing mechanical faults. If it is discriminated as an ordinary mechanical fault without electrical insulation protection, the current damage of bearing shafts will still repeatedly appear. Aiming at the problem that it is difficult to identify the bearing current damage fault under variable working conditions, a bearing shaft current damage identification method based on multiscale feature label propagation and manifold metric transfer (MFLP-MMT) is proposed. Firstly, the multiscale sub-band signal is obtained by wavelet packet decomposition, and the multiscale sub-band fuzzy entropy is obtained by calculating its fuzzy entropy. Then, according to the extracted features, a neighbor graph is constructed on the source domain of the known fault label to obtain the pseudo label of the target domain sample, and the source domain label information is gradually diffused by way of the graph label propagation. The multiscale sub-band fuzzy entropy of the sample is mapped to the low-dimensional manifold space by locality preserving projections (LPP), and the source domain samples close to the target domain are given higher weights by cross-domain density ratio estimation to solve the problem of domain offset. Combined with the label samples of the target domain in label propagation, the manifold distance metric is learned to minimize the intra-class distance and maximize the inter-class distance in the domain and eliminate the overlapping phenomenon in the domain. By increasing the range of label propagation after each iteration, the label propagation error of the leading graph is gradually reduced, and unsupervised metric transfer learning is realized. The experimental results show that the new method is superior to the semi-supervised transfer learning method in fault identification ability; the highest fault identification accuracy can reach 100% and it has a good robustness.

Robust fuzzy rough approximations with kNN granules for semi-supervised feature selection

Fuzzy rough set theory has attracted much attention because of its successful application in uncertainty measurement. To improve the efficiency and robustness of uncertainty measure using the theory, robust fuzzy rough approximations have become popular research. In this work, it first designs a novel uncertainty measure based on kNN granules. By combining the relative distance, the method is effectively applicable to the datasets with large density variation. Using the measure a robust FRS model is studied and simply denoted KNN-FRS model. The new model takes a compromise between the maximum lower approximation degree and the minimum classification loss in computing fuzzy rough approximations, which reduces the sensitivity of the classical FRS model to noise. Besides, some properties of KNN-FRS model are summarized. Next, KNN-FRS model is used to design a feature selection algorithm which is tested to validate the feasibility of the KNN-FRS model. And then the KNN-FRS model is applied to design a semi-supervised feature selection algorithm which selects features while labeling samples and the labeling method is designed based on the proposed uncertainty measure. Some experiments are conducted to validate the KNN-FRS model and the designed algorithms. The results illustrate the proposed KNN-FRS model is feasible, effective and robust. And the designed semi-supervised feature selection algorithm performs excellent performance.

Auto-segmentation of pancreatic tumor in multi-modal image using transferred DSMask R-CNN network

Pancreatic tumor segmentation is a difficult task due to the high variable shape, small size and hidden position of organs in patients for adaptive radiation therapy plan. To address the problems of limited labeled data, intra-class inconsistency and inter-class indistinction in pancreas tumor segmentation, a transferred DenseSE-Mask R-CNN (TDSMask R-CNN) Network segmentation model using Dense and SE block embedded is proposed in this paper. The multi-scale features strategy is selected to deal with high variability of pancreas and their tumor. The proposed network can learn complementary information from different modes (PET/MR) images respectively by the attention mechanism to get pancreatic tumor regions in different domain. As a result, the irrelevant information for segmenting the tumor area can be suppressed and get low false positives. Furthermore, accurate tumor location from PET image is transferred MRI training model for guide Dense-SE network learning to alleviate the small label samples and reduce network overfitting. Experimental results show that the proposed method achieves average Dice Similarity Coefficient (DSC) of 78.33%, sensitivity (SEN) of 78.56%, and specificity (SPE) of 99.72% on the collected PET/MR data set, which is superior to the existing method of some literatures. This algorithm can improve the accuracy of pancreatic tumor segmentation.

An innovative deep active learning approach for improving unlabeled audio classification by selectively querying informative instance

Audio classification tasks like speech recognition and acoustic scene analysis require substantial labeled data, which is expensive. This work explores active learning to reduce annotation costs for a sound classification problem with rare target classes where existing datasets are insufficient. A deep convolutional recurrent neural network extracts spectro-temporal features and makes predictions. An uncertainty sampling strategy queries the most uncertain samples for manual labeling by experts and non-experts. A new alternating confidence sampling strategy and two other certainty-based strategies are proposed and evaluated. Experiments show significantly higher accuracy than passive learning baselines with the same labeling budget. Active learning generalizes well in a qualitative analysis of 20,000 unlabeled recordings. Overall, active learning with a novel sampling strategy minimizes the need for expensive labeled data in audio classification, successfully leveraging unlabeled data to improve accuracy with minimal supervision.

Federated-Learning-Based Synchrotron X-Ray Microdiffraction Image Screening for Industry Materials

Synchrotron X-ray microdiffraction ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">μ</i> XRD) services are conducted for industrial minerals to identify their crystal impurities in terms of crystallinity and potential impurities. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">μ</i> XRD services generate huge loads of images that have to be screened before further processing and storage. However, there are insufficient effective labeled samples to train a screening model since service consumers are unwilling to share their original experimental images. In this article, we propose a physics law-informed federated learning (FL) based <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">μ</i> XRD image screening method to improve the screening while protecting data privacy. In our method, we handle the unbalanced data distribution challenge incurred by service consumers with different categories and amounts of samples with novel client sampling algorithms. We also propose hybrid training schemes to handle asynchronous data communications between FL clients and servers. The experiments show that our method can ensure effective screening for industrial users conducting industrial material testing while keeping commercially confidential information.

Active Learning-Driven Siamese Network for Hyperspectral Image Classification

Hyperspectral image (HSI) classification has recently been successfully explored by using deep learning (DL) methods. However, DL models rely heavily on a large number of labeled samples, which are laborious to obtain. Therefore, finding a way to efficiently embed DL models in limited labeled samples is a hot topic in the field of HSI classification. In this paper, an active learning-based siamese network (ALSN) is proposed to solve the limited labeled samples problem in HSI classification. First, we designed a dual learning-based siamese network (DLSN), which consists of a contrastive learning module and a classification module. Secondly, in view of the problem that active learning is difficult to effectively sample under the extremely limited labeling cost, we proposed an adversarial uncertainty-based active learning (AUAL) method to query valuable samples, and to promote DLSN to learn a more complete feature distribution by fine-tuning. Finally, an active learning architecture, based on inter-class uncertainty (ICUAL), is proposed to construct a lightweight sample pair training set, fully extracting the inter-class information of sample pairs and improving classification accuracy. Experiments on three generic HSI datasets strongly demonstrated the effectiveness of ALSN for HSI classification, with performance improvements over other related DL methods.

DC-FUDA: Improving deep clustering via fully unsupervised domain adaptation

By transferring knowledge from a source domain, the performance of deep clustering on an unlabeled target domain can be greatly improved. When achieving this, traditional approaches assume that an adequate amount of labeled data are available in the source domain. However, this assumption is not always satisfied in practice. First, it cannot be guaranteed that rich labeled samples are readily available in the selected source domain. Second, the noisy data in the source domain may lead to negative transferring. In this paper, we propose a novel transfer learning framework to improve deep clustering via fully unsupervised domain adaptation, called DC-FUDA. Specifically, to select reliable instances in the source domain for transferring, we propose a novel adaptive threshold algorithm to select low entropy instances. To transfer important features of the selected instances, we propose a feature-level domain adaptation network (FeatureDA) that cancels an unstable instance generation process. With extensive experiments, we validate that our method effectively improves deep clustering. Besides, without using any labeled data in the source domain, our method achieves competitive results, compared to the state-of-the-art methods using labeled data in the source domain.

Semi-Supervised Remote Sensing Image Semantic Segmentation Method Based on Deep Learning

In this paper, we study the semi-supervised semantic segmentation problem via limited labeled samples and a large number of unlabeled samples. We propose a self-learning semi-supervised approach for the semantic segmentation of high-resolution remote sensing images. Our approach uses two networks (UNet and DeepLabV3) to predict the labels of the same unlabeled sample, and the pseudo labels samples with high prediction consistency are added to the training samples to improve the accuracy of semantic segmentation under the condition of limited labeled samples. Our method expands training data samples by using unlabeled data samples with pseudo labels. In order to verify the effectiveness of the proposed method, some experiments were conducted on the improved ISPRS Vaihingen 2D Semantic Labeling dataset using the method that we proposed. We focus on the extraction of forest and vegetation information and focus on the impact of a large number of unlabeled samples on the precision of semantic segmentation, we combine water, surface, buildings, cars, and background into one category and named others, and we call the changed dataset the improved ISPRS Vaihingen dataset. The experimental results show that the proposed method can effectively improve the semantic segmentation accuracy of high-scoring remote sensing images with limited samples than common deep semi-supervised learning.

Attention-Based Active Learning Framework for Segmentation of Breast Cancer in Mammograms

Breast cancer is one of most serious malignant tumors that affect women’s health. To carry out the early screening of breast cancer, mammography provides breast cancer images for doctors’ efficient diagnosis. However, breast cancer lumps can vary in size and shape, bringing difficulties for the accurate recognition of both humans and machines. Moreover, the annotation of such images requires expert medical knowledge, which increases the cost of collecting datasets to boost the performance of deep learning methods. To alleviate these problems, we propose an attention-based active learning framework for breast cancer segmentation in mammograms; the framework consists of a basic breast cancer segmentation model, an attention-based sampling scheme and an active learning strategy for labelling. The basic segmentation model performs multi-scale feature fusion and enhancement on the basis of UNet, thus improving the distinguishing representation capability of the extracted features for further segmentation. Afterwards, the proposed attention-based sampling scheme assigns different weights for unlabeled breast cancer images by evaluating their uncertainty with the basic segmentation model. Finally, the active learning strategy selects unlabeled images with the highest weights for manual labeling, thus boosting the performance of the basic segmentation model via retraining with new labeled samples. Testing on four datasets, experimental results show that the proposed framework could greatly improve segmentation accuracy by about 15% compared with an existing method, while largely decreasing the cost of data annotation.

Semi-Supervised Gaussian Processes Active Learning Model for Imbalanced Small Data Based on Tri-Training With Data Enhancement

To solve the problem that some imbalanced small sample datasets only contain a few labeled samples, a semi-supervised gaussian processes active learning model based on improved tri-training with enhanced data is proposed. Firstly, the label samples are balanced and enhanced, and we present a quantitative enhanced data evaluation criteria based on the JS distance and the similarity of information entropy between enhanced data and original data to select the best enhanced data. Secondly, an improved semi-supervised learning method based on tri-training is proposed to find the unlabeled samples which have high confidence, so the certainty of the labeled samples group can be increased, in order to ensure that the three classifiers of tri-training have both difference and robustness, random forest is introduced to divide the features of the dataset into three groups with equal contribution, and each classifier trains different combinations of two feature groups. Thirdly, in order to query and classify the most informative unlabeled samples more precisely, active learning based on the Gaussian process and JS distribution range is structured, because of the high uncertainty of the unlabeled samples predicted by active learning, the similarity distribution range of JS distance is introduced to compare the similarity of unlabeled samples and labeled samples in active learning`s classifier, so the model can classify more diverse samples. The final experimental results show that compared with several traditional models, the proposed model performs better on artificial datasets and imbalanced small-size UCI datasets.

Semi-supervised Deep Learning via Transformation Consistency Regularization for Remote Sensing Image Semantic Segmentation

Deep convolutional neural networks (CNNs) have gotten a lot of press in the last several years, especially in domains like computer vision (CV) and remote sensing (RS). However, achieving superior performance with deep networks highly depends on a massive amount of accurately labeled training samples. In real-world applications, gathering a large number of labeled samples is time-consuming and labor-intensive, especially for pixel-level data annotation. This dearth of labels in land-cover classification is especially pressing in the RS domain because high-precision, high-quality labeled samples are extremely difficult to acquire, but unlabeled data is readily available. In this study, we offer a new semi-supervised deep semantic labeling framework for semantic segmentation of high-resolution RS images to take advantage of the limited amount of labeled examples and numerous unlabeled samples. Our model uses transformation consistency regularization (TCR) to encourage consistent network predictions under different random transformations or perturbations. We try three different transforms to compute the consistency loss and analyze their performance. Then, we present a deep semi-supervised semantic labeling technique by using a hybrid transformation consistency regularization (HTCR). A weighted sum of losses, which contains a supervised term computed on labeled samples and an unsupervised regularization term computed on unlabeled data, may be used to update the network parameters in our technique. Our comprehensive experiments on two RS datasets confirmed that the suggested approach utilized latent information from unlabeled samples to obtain more precise predictions and outperformed existing semi-supervised algorithms in terms of performance. Our experiments further demonstrated that our semi-supervised semantic labeling strategy has the potential to partially tackle the problem of limited labeled samples for high-resolution RS image land-cover segmentation.

A Novel Spectral Indices-Driven Spectral-Spatial-Context Attention Network for Automatic Cloud Detection

Cloud detection is a fundamental step for optical satellite image applications. Although existing deep learning method can provide more accurate cloud detection result. However, performance of these methods rely on a large number of label samples, whose collection is time-consuming and high cost. In addition, cloud detection is challenging in high brightness scenes due to due to cloud and high brightness object have a similar spectral feature. In this study, we propose a cloud index driven spectral-spatial-context attention network (SSCA-net) for cloud detection, which relies on no effort to manually collect label samples and can improve the accuracy cloud detection in high brightness scenes. The label samples are automatically generated from cloud index by using dual-threshold, which is then expanded to improve the completeness of cloud mask labels. We designed SSCA-net with the spectral-spatial-context aware module and spectral-spatial-context information aggregation module, aimed to improve the accuracy cloud detection in high brightness scenes. The results show that the proposed SSCA-net achieved good performance with average overall accuracy of 97.69% and average Kappa coefficient of 92.71% on Sentinel-2 and Landsat-8 dataset. This paper provides a fresh insight into how advanced deep attention network and cloud index can be integrated to obtain high accuracy of cloud detection on high brightness scenes.

Cross-Domain Self-Taught Network for Few-Shot Hyperspectral Image Classification

In recent years, deep learning models, which possess powerful feature extraction abilities, have achieved remarkable success in the classification of hyperspectral images (HSIs). Nevertheless, a common challenge faced by most deep learning models, including few-shot learning models, is the scarcity of valid labeled samples. To address this issue, we propose a cross-domain self-taught network (CDSTN) for few-shot hyperspectral image classification. The proposed CDSTN merges domain adaptation and semi-supervised self-taught strategy to implement the few-shot learning, which utilizes adequate labeled and unlabeled samples from source as well as target domain respectively. For the feature information extraction of HSI, we propose a deep spatial-spectral feature embedded extractor composed of four residual blocks and a channel attention module. Additionally, a set of domain classifiers are introduced behind each residual block for the purpose of domain alignment by extracting more domain information at different depths of the network. Finally, plenty of unlabeled samples are assigned with pseudo labels through the trained network, and a pseudo label refinement module is designed to select the most confident pseudo label sample for each class to further enrich the labeled database of target domain. Experiments conducted on four widely used benchmark HSI data sets demonstrate that CDSTN can obtain superior and stable performance with limited labeled samples compared with some state of the arts.

Machine Learning Approach to Improve Data Connectivity in Text-based Personality Prediction using Multiple Data Sources Mapping

This paper considers the task of personality prediction using social media text data. Personality datasets with conventional personality labels are few, and collecting them is challenging due to privacy concerns and the high expense of hiring expert psychologists to label them. Pertaining to a smaller number of labelled samples available, existing studies usually adds a sentiment, statistical NLP features to the text data to improve the accuracy of the personality detection model. To overcome these concerns, this research proposes a new methodology to generate a large amount of labelled data that can be used by deep learning algorithms. The model has three components: general data representation, data mapping and classification. The model applies Personality correlation descriptors to incorporate correlation information and further use this information in generating dataset mapping algorithm. Experimental results clearly demonstrate that the proposed method beats strong baselines across a variety of evaluation metrics. The results had the highest accuracy of 86.24% and 0.915 F1 measure score on the combined MBTI and Essays dataset. Moreover, the new dataset constructed contains 3,84,089 labelled samples on the combined dataset and can be further considered for personality prediction using the famous Five Factor Model thereby alleviating the problem of limited labelled samples for the purpose of personality detection.

Refining Noisy Labels With Label Reliability Perception for Person Re-Identification

Most person re-identification (Re-ID) approaches rely excessively on a great quantity of annotated training data. However, due to sampling errors or annotated errors, the label noise is unavoidable, which usually causes a dramatic decrease in the performance of existing Re-ID methods. To address this problem, we propose the label reliability perception (LRP) for person Re-ID by refining noisy labels. Specifically, a feature-fusion block (FFB) is proposed to enhance the discrimen- ability of pedestrians' features by expanding the network's attention span due to the fused feature, which is generated by overlapping the coarse-grained feature obtained by global average pooling and fine-grained features obtained by evenly dividing the feature map in the height dimension and performing global max pooling. In addition, the label dual perception (LDP) is proposed to refine noisy labels instead of filtering samples by evaluating the reliability of each training sample's label. Specifically, we meticulously design five evaluation modes for each sample to perceive the reliability of the labels of the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">k</i> -nearest neighbor images. Finally, we utilize the most reliable label to replace the noisy label and optimize the network. Extensive experiments prove the superiority of the proposed model over the competing methods; for instance, on Market1501, our method achieves 88.8% rank-1 accuracy and 70.5% mAP (4.7% and 4.3% improvements over the state-of-the-arts) under noise ratio 20%, and similarly on DukeMTMC-ReID, our method achieves 77.7% and 60.3%.

Weakly Supervised Few-Shot Segmentation via Meta-Learning

Semantic segmentation is a classic computer vision task with multiple applications, which includes medical and remote sensing image analysis. Despite recent advances with deep-based approaches, labeling samples (pixels) for training models is laborious and, in some cases, unfeasible. In this paper, we present two novel meta learning methods, named WeaSeL and ProtoSeg, for the few-shot semantic segmentation task with sparse annotations. We conducted extensive evaluation of the proposed methods in different applications (12 datasets) in medical imaging and agricultural remote sensing, which are very distinct fields of knowledge and usually subject to data scarcity. The results demonstrated the potential of our method, achieving suitable results for segmenting both coffee/orange crops and anatomical parts of the human body in comparison with full dense annotation.

Cross-Domain Few-Shot Hyperspectral Image Classification With Class-Wise Attention

Few-shot learning (FSL) is an effective method to solve the problem of hyperspectral image (HSI) classification with few labeled samples. It learns transferable knowledge from sufficient labeled auxiliary data to classify unseen classes with limited labeled samples for training. However, the distribution difference between auxiliary data and unseen classes results in the learned transferable knowledge not being well applied to the new task. Therefore, a class-wise attentive cross-domain FSL (CA-CFSL) framework is proposed in this article, in which a feature extractor is learned to extract data features with discriminability and domain invariance. The class-wise attention metric module (CAMM) introduces class-wise attention on the FSL framework to learn more discriminative features, which improves the interclass decision boundaries. Furthermore, an asymmetric domain adversarial module (ADAM) is designed to enhance the ability of extracting domain-invariant representations, which combines asymmetric adversarial training with embedded domain-specific information. Experimental results on four public HSI datasets demonstrate that the proposed method outperforms the existing methods.