Labeling Cost Research Articles

Use of Multi-Feature Extraction and Transfer Learning to Identify Urban Villages in China

Urban villages (UVs) are the most typical urban informal settlements in China, and the study of an effective identification method for UVs can help to provide a reference for the development of locally adapted UV transformation policies. In order to reduce the cost of labeling and enhance transferability, this study integrates remote sensing and social sensing data and applies sample migration from a labeled area to a less labeled area based on the theory of transfer learning. There are two main results of this study: (1) This study constructed a feature system for UV identification based on multi-feature extraction using a block as a unit, and experiments based on Tianhe District achieved an overall accuracy of 90% and a kappa coefficient of 0.76. (2) Using Tianhe District as the source domain and Jiangan District as the target domain, samples from the source domain were reused based on the KMM, TCA, and CORAL algorithms. The CORAL+RF algorithm showed the best performance, where its overall accuracy reached 97.06% and its kappa coefficient reached 0.89, and its overall accuracy reached 91.17% and its kappa coefficient reached 0.67 in the case of no target domain labeling. To sum up, the identification method for UVs proposed in the present study provides theoretical references for identification methods for UVs in different geographical areas.

MaTableGPT: GPT-Based Table Data Extractor from Materials Science Literature.

Efficiently extracting data from tables in the scientific literature is pivotal for building large-scale databases. However, the tables reported in materials science papers exist in highly diverse forms; thus, rule-based extractions are an ineffective approach. To overcome this challenge, the study presents MaTableGPT, which is a GPT-based table data extractor from the materials science literature. MaTableGPT features key strategies of table data representation and table splitting for better GPT comprehension and filtering hallucinated information through follow-up questions. When applied to a vast volume of water splitting catalysis literature, MaTableGPT achieves an extraction accuracy (total F1 score) of up to 96.8%. Through comprehensive evaluations of the GPT usage cost, labeling cost, and extraction accuracy for the learning methods of zero-shot, few-shot, and fine-tuning, the study presents a Pareto-front mapping where the few-shot learning method is found to be the most balanced solution owing to both its high extraction accuracy (total F1 score>95%) and low cost (GPT usage cost of 5.97 US dollars and labeling cost of 10 I/O paired examples). The statistical analyses conducted on the database generated by MaTableGPT revealed valuable insights into the distribution of the overpotential and elemental utilization across the reported catalysts in the water splitting literature.

Unsupervised global-local domain adaptation with self-training for remote sensing image semantic segmentation

ABSTRACT Unsupervised domain adaptation (UDA) techniques have the potential to enhance the transferability of neural network models in unknown scenarios and reduce the labelling costs associated with unlabelled datasets. Popular solutions to this challenging UDA task are adversarial training and self-training. However, current adversarial-based UDA methods emphasize only global or local feature alignment, which is insufficient for tackling the domain shift. In addition, self-training-based methods inevitably produce many wrong pseudo labels on the target domain due to bias towards the source domain. To tackle the above problems, this paper proposes a hybrid training framework that integrates global-local adversarial training and self-training strategies to effectively tackle global-local domain shift. First, the adversarial approach measures the discrepancies between domains from domain and category-level perspectives. The adversarial network incorporates discriminators at the local-category and global-domain levels, thereby facilitating global-local feature alignment through multi-level adversarial training. Second, the self-training strategy is integrated to acquire domain-specific knowledge, effectively mitigating negative migration. By combining these two domain adaptation strategies, we present a more efficient approach for mitigating the domain gap. Finally, a self-labelling mechanism is introduced to directly explore the inherent distribution of pixels, allowing for the rectification of pseudo labels generated during the self-training stage. Compared to state-of-the-art UDA methods, the proposed method gains 3.2 % , 1.21 % , 5.86 % , 6.16 % mIoU improvements on Rural → Urban, Urban → Rural, Potsdam → Vaihingen, Vaihingen → Potsdam, respectively.

Leveraging Unlabeled Data for Superior ROC Curve Estimation via a Semiparametric Approach

The receiver operating characteristic (ROC) curve is a widely utilized tool in various fields, including economics, medicine, and machine learning, for evaluating classification performance and comparing treatment effect. The absence of clear and readily labels is a frequent phenomenon in estimating ROC owing to various reasons like labeling cost, time constraints, data privacy and information asymmetry. Traditional supervised estimators commonly rely solely on labeled data, where each sample is associated with a fully-observed response variable. We propose a new set of semi-supervised (SS) estimators to exploit available unlabeled data (samples lack of observations for responses) to enhance the estimation precision under the semi-parametric setting assuming that the distribution of the response variable for one group is known up to unknown parameters. The newly proposed SS estimators have attractive properties such as adaptability and efficiency by leveraging the flexibility of kernel smoothing method. We establish the large sample properties of the SS estimators, which demonstrate that the SS estimators outperform the supervised estimator consistently under mild assumptions. Numeric experiments provide empirical evidence to support our theoretical findings. Finally, we showcase the practical applicability of our proposed methodology by applying it to two real datasets.

Towards Integrating ChatGPT into Satellite Image Annotation Workflows. A Comparison of Label Quality and Costs of Human and Automated Annotators

Towards Integrating ChatGPT into Satellite Image Annotation Workflows. A Comparison of Label Quality and Costs of Human and Automated Annotators

Category knowledge-guided few-shot bearing fault diagnosis

Real-time bearing fault diagnosis plays a vital role in maintaining the safety and reliability of sophisticated industrial systems. However, the scarcity of labeled data in fault diagnosis, due to the difficulty of collecting fault samples and the high cost of labeling, poses a significant challenge in learning discriminative fault features from limited and complex monitoring signals. Few-shot learning (FSL) emerges as a potent method for extracting and accurately classifying features from severe fault signals. Nonetheless, challenges such as data scarcity and environmental noise significantly impede the efficacy of existing FSL methods in diagnosing incipient faults effectively. These limitations are primarily due to the inadequate consideration of inter-class correlations within noisy contexts by current FSL strategies, which restricts their ability to extrapolate familiar features to new classes. Consequently, there is a pressing demand for an FSL approach that can exploit inter-class correlations to address the hurdles of data insufficiency and environmental complexities, thereby facilitating the diagnosis of incipient faults in few-shot settings. This paper proposes a novel category-knowledge-guided model tailored for few-shot multi-task scenarios. By leveraging attribute data from base categories and the similarities across new class samples, our model efficiently establishes mapping relations for unencountered tasks, significantly enhancing its generalization capabilities for early-stage fault diagnosis and multi-task applications. This model ensures swift and precise FSL fault diagnosis under uncharted operational conditions. Comparative analyses utilizing the Case Western Reserve University bearing dataset and the Early Mild Fault Traction Motor bearing dataset demonstrate our model’s superior performance against leading FSL and transfer learning approaches.

Augmenting a spine CT scans dataset using VAEs, GANs, and transfer learning for improved detection of vertebral compression fractures

In recent years, deep learning has become a popular tool to analyze and classify medical images. However, challenges such as limited data availability, high labeling costs, and privacy concerns remain significant obstacles. As such, generative models have been extensively explored as a solution to generate new images and overcome the stated challenges. In this paper, we augment a dataset of chest CT scans for Vertebral Compression Fractures (VCFs) collected from the American University of Beirut Medical Center (AUBMC), specifically targeting the detection of incidental fractures that are often overlooked in routine chest CTs, as these scans are not typically focused on spinal analysis. Our goal is to enhance AI systems to enable automated early detection of such incidental fractures, addressing a critical healthcare gap and leading to improved patient outcomes by catching fractures that might otherwise go undiagnosed. We first generate a synthetic dataset based on the segmented CTSpine1K dataset to simulate real grayscale data that aligns with our specific scenario. Then, we use this generated data to evaluate the generative capabilities of Deep Convolutional Generative Adverserial Networks (DCGANs), variational autoencoders (VAEs), and VAE-GAN models. The VAE-GAN model demonstrated the highest performance, achieving a Fréchet Inception Distance (FID) five times lower than the other architectures. To adapt this model to real-image scenarios, we perform transfer learning on the GAN, training it with the real dataset collected from AUBMC and generating additional samples. Finally, we train a CNN using augmented datasets that include both real and generated synthetic data and compare its performance to training on real data alone. We then evaluate the model exclusively on a test set composed of real images to assess the effect of the generated data on real-world performance. We find that training on augmented datasets significantly improves the classification accuracy on a test set composed of real images by 16 %, increasing it from 73 % to 89 %. This improvement demonstrates that the generated data is of high quality and enhances the model's ability to perform well against unseen, real data.

Utilizing active learning and attention-CNN to classify vegetation based on UAV multispectral data

This paper presents a deep learning model based on an active learning strategy. The model achieves accurate identification of vegetation types in the study area by utilizing multispectral data obtained from preprocessing of unmanned aerial vehicle (UAV) remote sensing equipment. This approach offers advantages such as high data accuracy, mobility, and easy data collection. In active learning, the minimum confidence scoring method and a sampling technique based on a data pool are employed to reduce labeling costs. The deep learning model incorporates a semantic segmentation gated full fusion module that integrates a dual attention mechanism. This module enhances the capture of detailed texture information, optimally allocates spectral weights, and improves the model’s ability to distinguish between similar categories. At a labeling cost of 20%, the average accuracy of the model is 93.2%. Compared with other models, the proposed model achieved the highest classification accuracy in the case of limited training samples. At full annotation cost, the average accuracy is 95.32%, with only a difference of about 2%, but saving 80% of annotation cost. Therefore, active learning strategies can filter out high-value samples that are beneficial for model training, greatly reducing the annotation cost of samples Finally, the recognition results of surface vegetation cover types in the study area are presented, and the model’s accuracy is verified through field investigation.

Weakly supervised bounding‐box generation for camera‐trap image based animal detection

AbstractIn ecology, deep learning is improving the performance of camera‐trap image based wild animal analysis. However, high labelling cost becomes a big challenge, as it requires involvement of huge human annotation. For example, the Snapshot Serengeti (SS) dataset contains over 900,000 images, while only 322,653 contains valid animals, 68,000 volunteers were recruited to provide image level labels such as species, the no. of animals and five behaviour attributes such as standing, resting and moving etc. In contrast, the Gold Standard SS Bounding‐Box Coordinates (GSBBC for short) contains only 4011 images for training of object detection algorithms, as the annotation of bounding‐box for animals in the image, is much more costive. Such a no. of training images, is obviously insufficient. To address this, the authors propose a method to generate bounding‐boxes for a larger dataset using limited manually labelled images. To achieve this, the authors first train a wild animal detector using a small dataset (e.g. GSBBC) that is manually labelled to locate animals in images; then apply this detector to a bigger dataset (e.g. SS) for bounding‐box generation; finally, we remove false detections according to the existing label information of the images. Experiments show that detector trained with images whose bounding‐boxes are generated using the proposal, outperformed the existing camera‐trap image based animal detection, in terms of mean average precision (mAP). Compared with the traditional data augmentation method, our method improved the mAP by 21.3% and 44.9% for rare species, also alleviating the long‐tail issue in data distribution. In addition, detectors trained with the proposed method also achieve promising results when applied to classification and counting tasks, which are commonly required in wildlife research.

Online Active Continual Learning for Robotic Lifelong Object Recognition.

In real-world applications, robotic systems collect vast amounts of new data from ever-changing environments over time. They need to continually interact and learn new knowledge from the external world to adapt to the environment. Particularly, lifelong object recognition in an online and interactive manner is a crucial and fundamental capability for robotic systems. To meet this realistic demand, in this article, we propose an online active continual learning (OACL) framework for robotic lifelong object recognition, in the scenario of both classes and domains changing with dynamic environments. First, to reduce the labeling cost as much as possible while maximizing the performance, a new online active learning (OAL) strategy is designed by taking both the uncertainty and diversity of samples into account to protect the information volume and distribution of data. In addition, to prevent catastrophic forgetting and reduce memory costs, a novel online continual learning (OCL) algorithm is proposed based on the deep feature semantic augmentation and a new loss-based deep model and replay buffer update, which can mitigate the class imbalance between the old and new classes and alleviate confusion between two similar classes. Moreover, the mistake bound of the proposed method is analyzed in theory. OACL allows robots to select the most representative new samples to query labels and continually learn new objects and new variants of previously learned objects from a nonindependent and identically distributed (i.i.d.) data stream without catastrophic forgetting. Extensive experiments conducted on real lifelong robotic vision datasets demonstrate that our algorithm, even trained with fewer labeled samples and replay exemplars, can achieve state-of-the-art performance on OCL tasks.

Localizing From Classification: Self-Directed Weakly Supervised Object Localization for Remote Sensing Images.

In recent years, object localization and detection methods in remote sensing images (RSIs) have received increasing attention due to their broad applications. However, most previous fully supervised methods require a large number of time-consuming and labor-intensive instance-level annotations. Compared with those fully supervised methods, weakly supervised object localization (WSOL) aims to recognize object instances using only image-level labels, which greatly saves the labeling costs of RSIs. In this article, we propose a self-directed weakly supervised strategy (SD-WSS) to perform WSOL in RSIs. To specify, we fully exploit and enhance the spatial feature extraction capability of the RSIs' classification model to accurately localize the objects of interest. To alleviate the serious discriminative region problem exhibited by previous WSOL methods, the spatial location information implicit in the classification model is carefully extracted by GradCAM ++ to guide the learning procedure. Furthermore, to eliminate the interference from complex backgrounds of RSIs, we design a novel self-directed loss to make the model optimize itself and explicitly tell it where to look. Finally, we review and annotate the existing remote sensing scene classification dataset and create two new WSOL benchmarks in RSIs, named C45V2 and PN2. We conduct extensive experiments to evaluate the proposed method and six mainstream WSOL methods with three backbones on C45V2 and PN2. The results demonstrate that our proposed method achieves better performance when compared with state-of-the-arts.

Dandelion center detection in perennial ryegrass with Heat maps using Convolutional Neural Networks

Industrial agricultural practices largely involve automated weed control processes and techniques. This phenomenon can also be seen in horticulture and landscaping. Generally, dandelion weeds (Taraxacum officinale) are a common pest and their detection is necessary for any type of removal. To address this detection issue, a cost-effective and intuitive labeling method using Heat maps is proposed for marking dandelion plant centers within perennial rye-grass. This method relies on approximate localization as opposed to pinpoint accuracy. An expandable lightweight Convolutional Neural Network (CNN) is built on a base network to generate detection output maps at two resolutions. Multiple loss functions are expanded to multi-instance predictions and their combinations are examined through ablation to assess and rank their performance. Different methods of computing standard performance metrics are also explored. Also, different backbone networks are also shown to reveal varying performance advantages. Through these methods, dandelion weed centers can consistently be located with robustness to noise and erroneous labels and with good precision. Furthermore, our method is almost entirely end-to-end. The experimental results demonstrate that our methods outperform Semantic Segmentation models in the precision of output maps while avoiding the need of intensive labeling costs. In addition, when applying Hierarchical Clustering to the segmentation maps for a complete comparison in center detection, our methods double the accuracy and do not require the manual tuning of cluster parameters. Our proposed application of soft computing can be used in the landscaping industry and adapted to other fields with relative ease. The binary classification and object detection tasks of locating dandelion plants can be extended to multi-class problems with other plants.

A Survey on Open-Vocabulary Detection and Segmentation: Past, Present, and Future.

As the most fundamental scene understanding tasks, object detection and segmentation have made tremendous progress in deep learning era. Due to the expensive manual labeling cost, the annotated categories in existing datasets are often small-scale and pre-defined, i.e., state-of-the-art fully-supervised detectors and segmentors fail to generalize beyond the closed vocabulary. To resolve this limitation, in the last few years, the community has witnessed an increasing attention toward Open-Vocabulary Detection (OVD) and Segmentation (OVS). By "open-vocabulary", we mean that the models can classify objects beyond pre-defined categories. In this survey, we provide a comprehensive review on recent developments of OVD and OVS. A taxonomy is first developed to organize different tasks and methodologies. We find that the permission and usage of weak supervision signals can well discriminate different methodologies, including: visual-semantic space mapping, novel visual feature synthesis, region-aware training, pseudo-labeling, knowledge distillation, and transfer learning. The proposed taxonomy is universal across different tasks, covering object detection, semantic/instance/panoptic segmentation, 3D and video understanding. The main design principles, key challenges, development routes, methodology strengths, and weaknesses are thoroughly analyzed.

Few-Shot Learning Method Based on Data Enhancement and Transfer Learning

With the significant progress of deep learning technology in many fields, the dependence of model training on a large amount of labeled data is increasingly prominent. However, in many practical application scenarios, especially in tasks with high labeling costs, data scarcity often occurs. This realistic challenge has promoted the rise of Few-Shot Learning (FSL) technology, which seeks to achieve effective learning of models with extremely limited samples. This article provides a comprehensive overview of the theoretical background, key technologies of FSL, to explore its potential and effectiveness in solving the problem of small-sample learning. In the method overview section, this article pays special attention to FSL strategies based on data augmentation and transfer learning. By reviewing and analyzing these methods, this article aims to provide some theoretical support and technical references for further exploration in this field and hopes to contribute to solving the problem of data scarcity and promote the sustainable development of this field.

Cross-domain fine grained strip steel defect detection method based on semi-supervised learning and Multi-head Self Attention coordination

The identification of steel strip defects plays a pivotal role in assessing steel quality and advancing production technology. However, the majority of intelligent defect recognition algorithms for steel strips, based on deep learning, primarily focus on supervised learning. These methods depend on a multitude of training samples, incurring additional manual labelling costs, and exhibit low recognition efficiency. In contrast to supervised learning, we integrate the fine-grained characteristics of strip defects. We propose a cross-domain, fine-grained strip defect detection method based on semi-supervised learning and Multi-head self-attention coordination, along with an improvement strategy, resulting in a novel network structure: Multi-head Self Attention and Semi-supervised collaborative detection network (MSD Net). This method initiates the cross-domain migration of defect samples through Cycle Generative Adversarial Networks (Cycle GAN), creating new semi-supervised training samples from source domain and target domain data to enhance data distribution diversity. The detection model is then constructed leveraging the advantages of Multi-head Self Attention (MSA) in augmenting the global receptive field of feature extraction. The proposed semi-supervised learning method employs a pseudo-label allocation strategy to guide the model in fully utilizing the distribution fitting of unlabeled samples. This allows the deep neural network to learn a more comprehensive multivariate data distribution within the training set, thereby enhancing the generalization ability of the semi-supervised model. Experimental results on the benchmark dataset for steel strip defect detection demonstrate that the cross-domain semi-supervised method achieves a test accuracy of 96.1 % on mAP@0.5, surpassing the supervised baseline model by 4.8 %. Our method also outperforms the baseline supervised model in the accuracy of small target recognition on PASCAL VOC 2007 datasets. Additionally, we have implemented a strip defect detection system based on edge computing for real-time deployment of the proposed algorithm. Testing in an actual industrial setting further validates the efficacy of our proposed method in practical applications. Our work encourages further exploration, the task of public datasets can be obtained at https://github.com/songzhiweiknight/NEU-DET-Datasets.git.

Semi-supervised convolutional generative adversarial networks for dynamic fault classification with manifold regularization

Data-driven fault classification identifies and categorizes faults or anomalies in industrial systems, to ensure efficient and safe operations. This paper addresses the challenges of semi-supervised learning for dynamic fault classification in industrial processes. The key problems involve utilizing three unique characteristics of typical industrial process dataset: 1) the presence of unlabeled samples due to high labeling costs and required expert knowledge, 2) the local manifold structure arising from strong variable coupling, and 3) time correlation inherent in time-series sensing data of dynamic processes. To tackle these challenges, a semi-supervised fault classification model, SSDCGAN, is proposed, based on deep convolutional Generative Adversarial Networks. The model captures dynamic temporal information through a moving window technique and leverages manifold regularization to maintain classifier consistency along the manifold’s tangent direction. Evaluations on the Tennessee Eastman (TE) benchmark demonstrate that SSDCGAN enhances fault classification accuracy, outperforming current methods.

Unsupervised attention-guided domain adaptation model for Acute Lymphocytic Leukemia (ALL) diagnosis

Background and Objective:Acute lymphocytic leukemia (ALL) is a dangerous disease characterized by an increased number of abnormal blood cells in the blood. Its early diagnosis and treatment are crucial, as it can lead to severe consequences if left untreated. Manual examination of blood samples by pathologists and/or hematologists is time-consuming and requires expert skill, so automated and fast solutions need to be developed. However, the marginal data distribution of samples taken from subjects under certain conditions is a major obstacle to building a model that works on datasets obtained under different conditions. Labeling the new dataset also means extra costs. Methods:Considering these reasons, this study proposes an attention-enhanced generative adversarial network (GAN) model to move two datasets with different structures into the same feature space. The proposed model is used to transfer the blast and normal cells to the target domain regardless of the background to eliminate the domain difference between the datasets. Results:By learning the complex structure and class features of the cells in an unsupervised manner, the labeling cost is eliminated and it is shown that the trained classifier achieves better results than other domain adaptation methods in the literature. Conclusion:At the end of the study, it was seen that attention mechanisms are highly skilled in extracting the useful parts from the data. In this way, domain mismatch between datasets could be eliminated.

Active Learning with Particle Swarm Optimization for Enhanced Skin Cancer Classification Utilizing Deep CNN Models.

Skin cancer is a critical global health issue, with millions of non-melanoma and melanoma cases diagnosed annually. Early detection is essential to improving patient outcomes, yet traditional deep learning models for skin cancer classification are often limited by the need for large, annotated datasets and extensive computational resources. The aim of this study is to address these limitations by proposing an efficient skin cancer classification framework that integrates active learning (AL) with particle swarm optimization (PSO). The AL framework selectively identifies the most informative unlabeled instances for expert annotation, minimizing labeling costs while optimizing classifier performance. PSO, a nature-inspired metaheuristic algorithm, enhances the selection process within AL, ensuring the most relevant data points are chosen. This method was applied to train multiple Convolutional Neural Network (CNN) models on the HAM10000 skin lesion dataset. Experimental results demonstrate that the proposed AL-PSO approach significantly improves classification accuracy, with the Least Confidence strategy achieving approximately 89.4% accuracy while using only 40% of the labeled training data. This represents a substantial improvement over traditional approaches in terms of both accuracy and efficiency. The findings indicate that the integration of AL and PSO can accelerate the adoption of AI in clinical settings for skin cancer detection. The code for this study is publicly available at ( https://github.com/Sayantani-31/AL-PSO ).

AI-Guided Feature Segmentation Techniques to Model Features from Single Crystal Diamond Growth

Improving diamond growth to achieve high-quality materials for applications like optics and quantum detectors is a key objective. Leveraging machine learning, especially with its challenges such as complex, time-dependent data features and the sheer data volume per growth cycle, presents a promising solution. The extraction of accurate spatial features from images for real-time diamond growth monitoring is challenging due to data's complex and sparse nature. This paper evaluates various feature extraction methods in diamond growth, introducing a novel deep learning approach for precise geometric feature segmentation. It utilizes a human-in-the-loop system for efficient data annotation, significantly cutting down labeling time and costs. Our approach, using the DeeplabV3plus model, showcases high efficiency in feature classification, with accuracies up to 96.31% for pocket holders, 98.60% for diamond tops, and 91.64% for sides, demonstrating deep learning's potential in handling complex dataset features effectively.

Exploring better sparsely annotated shadow detection

Sparsely annotated image segmentation has attracted increasing attention due to its low labeling cost. However, existing weakly-supervised shadow detection methods require complex training procedures, and there is still a significant performance gap compared to fully-supervised methods. This paper summarizes two current challenges in sparsely annotated shadow detection, i.e., weak supervision diffusion and poor structure recovery, and attempts to alleviate them. To this end, we propose a one-stage weakly-supervised learning framework to facilitate sparsely annotated shadow detection. Specifically, we first design a simple yet effective semantic affinity module (SAM) that adaptively propagates scribble supervision to unlabeled regions using a gradient diffusion scheme. Then, to better recover shadow structures, we introduce a feature-guided edge-aware loss, which leverages higher-level semantic relations to perceive shadow boundaries, while avoiding interference from ambiguous regions. Finally, we present an intensity-guided structure consistency loss to ensure that the same images with different brightness are predicted to be consistent shadow masks, which can be regarded as a self-consistent mechanism to improve the model’s generalization ability. Experimental results on three benchmark datasets demonstrate that our approach significantly outperforms previous weakly-supervised methods and achieves competitive performance in comparison to recent state-of-the-art fully-supervised methods.