In this work we present an end-to-end system for text spotting—localising and recognising text in natural scene images—and text based image retrieval. This system is based on a region proposal ...mechanism for detection and deep convolutional neural networks for recognition. Our pipeline uses a novel combination of complementary proposal generation techniques to ensure high recall, and a fast subsequent filtering stage for improving precision. For the recognition and ranking of proposals, we train very large convolutional neural networks to perform word recognition on the whole proposal region at the same time, departing from the character classifier based systems of the past. These networks are trained solely on data produced by a synthetic text generation engine, requiring no human labelled data. Analysing the stages of our pipeline, we show state-of-the-art performance throughout. We perform rigorous experiments across a number of standard end-to-end text spotting benchmarks and text-based image retrieval datasets, showing a large improvement over all previous methods. Finally, we demonstrate a real-world application of our text spotting system to allow thousands of hours of news footage to be instantly searchable via a text query.
Although highly valuable for a variety of applications, urban mobility data are rarely made openly available, as it contains sensitive personal information. Synthetic data aims to solve this issue by ...generating artificial data that resembles an original dataset in structural and statistical characteristics, but omits sensitive information. For mobility data, a large number of corresponding models have been proposed in the past decade. This systematic review provides a structured comparative overview of the current state of this heterogeneous, active field of research. A special focus is put on the applicability of the reviewed models in practice.
Lithium-ion batteries undergo capacity loss and power fade over time. Despite indicating degradation, these changes lack internal insights. Degradation modes group various mechanisms but are ...challenging to quantify due to aging complexity. In this paper, a noninvasive and comprehensive diagnostic framework is proposed for the accurate estimation of the state of health (SOH) and degradation modes. A large amount of synthetic data is generated from the mechanistic model to cover many typical aging paths without using redundant experimental data. A data-driven multistep diagnosis method is developed by using incremental capacity sequences. This method analyses unique sensitivities to voltage responses from different degradation modes and identifies them sequentially to simplify complex interactions. In addition, overpotential correction is incorporated to estimate battery degradation modes under normal usage. This framework is validated through experimental data under various operating conditions, affirming the 2 % accuracy of SOH estimation and the effective automatic quantification of degradation modes. Furthermore, the method is applicable to common state-of-charge (SOC) windows ranging from 25% to 85 % (3.6 V–4.1 V for the voltage window) and a high current rate up to C/4. The diagnostic framework does not rely on any regular calibration data during model training and thus has high potential for practical application.
•A novel framework without regular calibration is proposed for battery diagnosis.•Synthetic training dataset with typical aging paths is generated via mechanistic model.•Data-driven multistep diagnosis is employed to estimate SOH and degradation modes.•Common charging SOC window and high current rate enable practical aging diagnosis.
Biclustering is increasingly used in biomedical data analysis, recommendation tasks, and text mining domains, with hundreds of biclustering algorithms proposed. When assessing the performance of ...these algorithms, more than real datasets are required as they do not offer a solid ground truth. Synthetic data surpass this limitation by producing reference solutions to be compared with the found patterns. However, generating synthetic datasets is challenging since the generated data must ensure reproducibility, pattern representativity, and real data resemblance.
We propose G-Bic, a dataset generator conceived to produce synthetic benchmarks for the normative assessment of biclustering algorithms. Beyond expanding on aspects of pattern coherence, data quality, and positioning properties, it further handles specificities related to mixed-type datasets and time-series data.G-Bic has the flexibility to replicate real data regularities from diverse domains. We provide the default configurations to generate reproducible benchmarks to evaluate and compare diverse aspects of biclustering algorithms. Additionally, we discuss empirical strategies to simulate the properties of real data.
G-Bic is a parametrizable generator for biclustering analysis, offering a solid means to assess biclustering solutions according to internal and external metrics robustly.
Celotno besedilo
Dostopno za:
DOBA, IZUM, KILJ, NUK, PILJ, PNG, SAZU, SIK, UILJ, UKNU, UL, UM, UPUK
Quantile mapping bias correction algorithms are commonly used to correct systematic distributional biases in precipitation outputs from climate models. Although they are effective at removing ...historical biases relative to observations, it has been found that quantile mapping can artificially corrupt future model-projected trends. Previous studies on the modification of precipitation trends by quantile mapping have focused on mean quantities, with less attention paid to extremes. This article investigates the extent to which quantile mapping algorithms modify global climate model (GCM) trends in mean precipitation and precipitation extremes indices. First, a bias correction algorithm, quantile delta mapping (QDM), that explicitly preserves relative changes in precipitation quantiles is presented. QDM is compared on synthetic data with detrended quantile mapping (DQM), which is designed to preserve trends in the mean, and with standard quantile mapping (QM). Next, methods are applied to phase 5 of the Coupled Model Intercomparison Project (CMIP5) daily precipitation projections over Canada. Performance is assessed based on precipitation extremes indices and results from a generalized extreme value analysis applied to annual precipitation maxima. QM can inflate the magnitude of relative trends in precipitation extremes with respect to the raw GCM, often substantially, as compared to DQM and especially QDM. The degree of corruption in the GCM trends by QM is particularly large for changes in long period return values. By the 2080s, relative changes in excess of +500% with respect to historical conditions are noted at some locations for 20-yr return values, with maximum changes by DQM and QDM nearing +240% and +140%, respectively, whereas raw GCM changes are never projected to exceed +120%.
Celotno besedilo
Dostopno za:
BFBNIB, DOBA, IZUM, KILJ, NUK, PILJ, PNG, SAZU, SIK, UILJ, UKNU, UL, UM, UPUK
Infrastructure scene understanding from image data aids diverse applications in construction and maintenance. Recently, deep learning models have been employed to extract information regarding ...infrastructure from visual data. The performance of these models depends significantly on the volume of training data. However, preparing the training data is time-consuming and laborious, as it entails labeling numerous images. To address this issue, this paper proposes a method for generating high-quality synthetic data that includes the automatic annotation of infrastructure scenes. The method consists of three steps: 1) translating building information model (BIM) images into real-world images, 2) automatically labeling them using the spatial information contained in the BIM to generate various synthetic datasets, and 3) splicing the selected synthetic datasets together to form the final synthetic dataset. The Mask R-CNN models trained with building and bridge synthetic data achieved average precisions of 71.6% and 84.9%, respectively.
•The proposed method generates high-quality synthetic data on infrastructure scenes.•The annotation on the synthetic image is conducted automatically.•Synthetic images are generated by transforming BIM images using CycleGAN.•Two-stage synthetic data generation is proposed for better segmentation performance.
We develop metrics for measuring the quality of synthetic health data for both education and research. We use novel and existing metrics to capture a synthetic dataset’s resemblance, privacy, utility ...and footprint. Using these metrics, we develop an end-to-end workflow based on our generative adversarial network (GAN) method, HealthGAN, that creates privacy preserving synthetic health data. Our workflow meets privacy specifications of our data partner: (1) the HealthGAN is trained inside a secure environment; (2) the HealthGAN model is used outside of the secure environment by external users to generate synthetic data. This second step facilitates data handling for external users by avoiding de-identification, which may require special user training, be costly, or cause loss of data fidelity. This workflow is compared against five other baseline methods. While maintaining resemblance and utility comparable to other methods, HealthGAN provides the best privacy and footprint. We present two case studies in which our methodology was put to work in the classroom and research settings. We evaluate utility in the classroom through a data analysis challenge given to students and in research by replicating three different medical papers with synthetic data. Data, code, and the challenge that we organized for educational purposes are available.
Bringing artificial intelligence on board Earth observation satellites unlocks unprecedented possibilities to extract actionable items from various image modalities at the global scale in real time. ...This is of paramount importance nowadays, as downlinking large amounts of imagery is not only prohibitively expensive but also time-consuming. However, building deep learning solutions that could be deployed on board an edge device is challenging due to the limited manually-annotated satellite datasets and hardware constraints of an edge device. This paper addresses these challenges through harnessing a blend of data-centric and model-centric approaches to build a well-generalizing yet efficient and resource-frugal deep learning model for multi-class satellite image classification in the few-shot learning settings. This integrated strategy is formulated to enhance classification performance while accommodating the unique demands of an image analysis chain on board OPS-SAT, a nanosatellite operated by the European Space Agency. The experiments performed over a real-world dataset of OPS-SAT images delves into the interactions between data- and model-centric techniques, underscores the significance of synthesizing artificial training data and emphasizes the value of ensemble learning. However, they also caution against negative transfer in domain adaptation. This study sheds light on effective model training strategies and highlights the multifaceted challenges inherent in deep learning for practical Earth observation, contributing insights to the field of satellite image classification within the constraints of nanosatellite operations. To ensure reproducibility of our study, the implementation is available at https://github.com/ShendoxParadox/Few-shot-satellite-image-classification-OPS-SAT.
•We propose data- and model-centric approach to few-shot image classification.•We establish a framework for building deep learning models from limited samples.•We investigate our approaches over the benchmark images captured by OPS- SAT.•We ensure full reproducibility of our study.•The proposed techniques offer high classification accuracy and resource frugality.
•We propose a deep learning‐based method for tomato plant disease detection.•We generate synthetic images using C‐GAN for data augmentation purposes.•A DenseNet121 model is trained on the original ...tomato leaf and synthetic images.•The proposed data augmentation technique improves network generalizability.•Proposed method achieves the best accuracy of 99.51% for 5‐class classification.
Plant diseases and pernicious insects are a considerable threat in the agriculture sector. Therefore, early detection and diagnosis of these diseases are essential. The ongoing development of profound deep learning methods has greatly helped in the detection of plant diseases, granting a vigorous tool with exceptionally precise outcomes but the accuracy of deep learning models depends on the volume and the quality of labeled data for training. In this paper, we have proposed a deep learning-based method for tomato disease detection that utilizes the Conditional Generative Adversarial Network (C-GAN) to generate synthetic images of tomato plant leaves. Thereafter, a DenseNet121 model is trained on synthetic and real images using transfer learning to classify the tomato leaves images into ten categories of diseases. The proposed model has been trained and tested extensively on publicly available PlantVillage dataset. The proposed method achieved an accuracy of 99.51%, 98.65%, and 97.11% for tomato leaf image classification into 5 classes, 7 classes, and 10 classes, respectively. The proposed approach shows its superiority over the existing methodologies.