Improvement In Accuracy Research Articles

Unravelling uncertainty in trajectory prediction using a non-parametric approach

Predicting the trajectories of road agents is fundamental for self-driving cars. Trajectory prediction contains many sources of uncertainty in data and modelling. A thorough understanding of this uncertainty is crucial in a safety-critical task like auto-piloting a vehicle. In practice, it is necessary to distinguish between the uncertainty caused by partial observability of all factors that may affect a driver’s near-future decisions, the so-called aleatoric uncertainty, and the uncertainty of deploying a model in new scenarios that are possibly not present in the training set, the so-called epistemic uncertainty. They reflect the trade-off between data collection and model improvement In this paper, we propose a new framework to systematically quantify both sources of uncertainty. Specifically, to approximate the spatial distribution of an agent’s future position, we propose a 2D histogram-based deep learning model combined with deep ensemble techniques for measuring aleatoric and epistemic uncertainty by entropy-based quantities. The proposed Uncertainty Quantification Network (UQnet) employs a causal part to enhance its generalizability so rare driving behaviours can be effectively identified. Experiments on the INTERACTION dataset show that UQnet is able to give more robust predictions in generalizability tests compared to the correlation-based models. Further analysis presents that high aleatoric uncertainty cases are mainly caused by heterogeneous driving behaviours and unknown intended directions. Based on this aleatoric uncertainty component, we estimate the lower bounds of mean-square-error and final-displacement-error as indicators for the predictability of trajectories. Furthermore, the analysis of epistemic uncertainty illustrates that domain knowledge of speed-dependent driving behaviour is essential for adapting a model from low-speed to high-speed situations. Our paper contributes to motion forecasting with a new framework, that recasts the problem of accuracy improvement in a way that focuses on differentiating between unpredictable components and rare cases for which more and different data should be collected.

Science in the service of politics. The cartographic representation of a local territorial context of the Kingdom of Naples in the early nineteenth century

The erudite priest Marciano Di Leo (1751–1819), a prominent personality in the historical and geographical panorama of his time, not only in his home territory, authored a vast literary and poetic production, but also tried his hand at producing some maps, referring to a province of the Kingdom of Naples. At a time when the principles of geodetic cartography had become increasingly known, even locally, hand in hand with improvements in technology and accuracy of measurements, the author reflects the historical narratives of progress of the European (and Neapolitan) Enlightenment and translates them into an unpublished manuscript of statistical, historical, and geographical nature, accompanied by numerous maps of various scales. The rediscovery of a largely unknown—and therefore not very thorough—minor cartographic production underscores the spread even in more marginal contexts of the most innovative ideas and increasingly precise scientific foundations in the cartographic-mathematical representation of the territory. It also illustrates the role of a number of intellectuals in the service of the political choices of their time, in an attempt-often unrealized-to bring about a decisive change of course in public administration, in accordance with Enlightenment ideals and in the spirit of reform that spread throughout Europe thanks to the French Revolution.

Lightweight Deep Learning for Resource-Constrained Environments: A Survey

Over the past decade, the dominance of deep learning has prevailed across various domains of artificial intelligence, including natural language processing, computer vision, and biomedical signal processing. While there have been remarkable improvements in model accuracy, deploying these models on lightweight devices, such as mobile phones and microcontrollers, is constrained by limited resources. In this survey, we provide comprehensive design guidance tailored for these devices, detailing the meticulous design of lightweight models, compression methods, and hardware acceleration strategies. The principal goal of this work is to explore methods and concepts for getting around hardware constraints without compromising the model’s accuracy. Additionally, we explore two notable paths for lightweight deep learning in the future: deployment techniques for TinyML and Large Language Models. Although these paths undoubtedly have potential, they also present significant challenges, encouraging research into unexplored areas.

Estimating Forest Canopy Height based on GEDI Lidar Data and Multi-source Remote Sensing Images

Abstract. Estimating forest canopy height is crucial for assessing aboveground biomass and carbon sequestration. Light detection and ranging (lidar) is an important technology for its ability in capturing vertical structural information. However, due to instrument limitations and cost constraints, acquiring large-scale and continuous forest data solely through lidar is challenging. To compensate this, remote sensing images can be used to cover wide regions. Therefore, leveraging multi-source data for constructing canopy height models (CHMs) holds great promise in this field. The objective of this study is to evaluate and compare the contributions of multi-source remote sensing data and methods in estimating forest canopy height. In constructing the CHM, the commonly used random forest (RF) and fully convolutional network (FCN) are assessed. The canopy height obtained from GEDI was used as the reference data, and Landsat 8 and Sentinel-2 data were used for prediction. Multiple CHMs were constructed for the Dabie Mountains, Central China, in 2019 based on different data sources and methods, respectively, which are then comparatively analysed. The results showed that (1) the accuracy of the CHM using Sentinel-2 as input is marginally better than that using Landsat 8 based on RF, where the difference is insignificant; and (2) FCN is less accurate than RF despite domain-specific fine-tuning, although further improvement in accuracy is expected by weighing in more FCN models.

Integrating Data and Model Optimization for Improved Multi-Class Land Cover Classification Using Multispectral Remote Sensing Data

Abstract. With the rapid development of artificial intelligence, significant progress has been made in land cover classification using deep learning methods. However, in existing research, most studies focus more on improving classification accuracy by optimizing the model structure and less on mining the value of the data itself. In this paper, experiments on remote sensing multi-class land cover classification were conducted based on Worldview3 data, and strategies to improve classification accuracy were proposed in terms of sampling methods, band combination, loss function, and model optimization. Experiment results show that the proposed improvement strategies are effective for multi-class land cover classification, with recall, F1, and IoU improved by 29%, 17%, and 19%, respectively. The significant improvement in classification accuracy for less-represented targets confirms that enhancing data richness and balance leads to greater improvement than just optimizing the model.

Pengembangan Sistem Oracle Forms untuk Optimisasi Proses Penjualan di PT.XYZ

This research focuses on the development of an efficient sales system for PT. XYZ a retail company in Indonesia. The company faces challenges in its sales processes due to the lack of efficiency in the manual system, which is susceptible to errors. The research aims to optimize the sales process by utilizing Oracle Forms 10g and implementing the Rapid Application Development (RAD) methodology to ensure timely solutions. The developed system demonstrates a significant improvement in efficiency, accuracy, and speed. The use of Oracle Forms and SQL Developer simplifies data entry and management processes, while the implementation of List of Values (LOV) reduces errors and enhances data consistency. Furthermore, the system provides real-time insights into sales performance, enabling better decision-making and strategic planning for the company.

Error logs in the second language classroom: Exploring the relationship between learner engagement with written corrective feedback and improvements in writing accuracy

This classroom-based study investigates how engagement with written corrective feedback (WCF) through the completion of error logs is related to improvements in linguistic accuracy in new pieces of writing. Conducted over eight weeks in a Hong Kong secondary school English as a foreign language (EFL) classroom with 20 student participants, the study involved learners completing error logs after receiving metalinguistic WCF on their written texts. The error logs required learners to record each error and type, their corrections, and explanations for each correction. Using a pre-test, post-test design, the study compared learners' linguistic accuracy before and after the error log intervention, and explored the relationship between accuracy gains and successful error log completion. Learners' attitudes towards using error logs were also assessed through questionnaires and interviews. The findings indicated that linguistic accuracy significantly improved from pre-test to post-test, there was a significant positive correlation between accuracy gains and level of success in explaining corrections, and that learners generally held positive attitudes towards using error logs. The positive perceptions stemmed mainly from learners’ recognition that the error logs raised their awareness of common errors and fostered their linguistic development.

MISATO: machine learning dataset of protein-ligand complexes for structure-based drug discovery.

Large language models have greatly enhanced our ability to understand biology and chemistry, yet robust methods for structure-based drug discovery, quantum chemistry and structural biology are still sparse. Precise biomolecule-ligand interaction datasets are urgently needed for large language models. To address this, we present MISATO, a dataset that combines quantum mechanical properties of small molecules and associated molecular dynamics simulations of ~20,000 experimental protein-ligand complexes with extensive validation of experimental data. Starting from the existing experimental structures, semi-empirical quantum mechanics was used to systematically refine these structures. A large collection of molecular dynamics traces of protein-ligand complexes in explicit water is included, accumulating over 170 μs. We give examples of machine learning (ML) baseline models proving an improvement of accuracy by employing our data. An easy entry point for ML experts is provided to enable the next generation of drug discovery artificial intelligence models.

An improvement method for pancreas CT segmentation using superpixel-based active contour

Objective. Pancreas is one of the most challenging organs for Computed Tomograph (CT) image automatic segmentation due to its complex shapes and fuzzy edges. It is simple and universal to use the traditional segmentation method as a post-processor of deep learning method for segmentation accuracy improvement. As the most suitable traditional segmentation method for pancreatic segmentation, the active contour model (ACM), still suffers from the problems of weak boundary leakage and slow contour evolution speed. Therefore, a convenient post-processor for any deep learning methods using superpixel-based active contour model (SbACM) is proposed to improve the segmentation accuracy. Approach. Firstly, the superpixels with strong adhesion to edges are used to guide the design of narrowband and energy function. A multi-scale evolution strategy is also proposed to reduce the weak boundary leakage and comprehensively improve the evolution speed. Secondly, using the original image and the coarse segmentation results obtained from deep learning methods as inputs, the proposed SbACM method is used as a post-processor for fine segmentation. Finally, the pancreatic segmentation public dataset TCIA from the National Institutes of Health(NIH, USA) is used for evaluation, and the Wilcoxon Test confirmed that the improvement of proposed method is statistically significant. Main results. (1) the superpixel-based narrowband shape and dynamic edge energy of the proposed SbACM work for boundary leakage reduction, as well as the multi-scale evolution strategy and dynamic narrowband width for the evolution speed improvement; (2) as a post-processor, SbACM can increase the Dice similarity coefficients (DSC) of five typical UNet-based models, including UNet, SS-UNet, PBR UNet, ResDSN, and nnUNet, 2.35% in average and 9.04% in maximum. (3) Based on the best backbone nnUNet, the proposed post-processor performs better than either adding edge awareness or adding edge loss in segmentation enhancement without increasing the complexity and training time of deep learning models. Significance. The proposed SbACM can improve segmentation accuracy with the lowest cost, especially in cases of squeezed fuzzy edges with similar neighborhood , and complex edges.

A Novel Improved Variational Mode Decomposition-Temporal Convolutional Network-Gated Recurrent Unit with Multi-Head Attention Mechanism for Enhanced Photovoltaic Power Forecasting

Photovoltaic (PV) power forecasting plays a crucial role in optimizing renewable energy integration into the grid, necessitating accurate predictions to mitigate the inherent variability of solar energy generation. We propose a novel forecasting model that combines improved variational mode decomposition (IVMD) with the temporal convolutional network-gated recurrent unit (TCN-GRU) architecture, enriched with a multi-head attention mechanism. By focusing on four key environmental factors influencing PV output, the proposed IVMD-TCN-GRU framework targets a significant research gap in renewable energy forecasting methodologies. Initially, leveraging the sparrow search algorithm (SSA), we optimize the parameters of VMD, including the mode component K-value and penalty factor, based on the minimum envelope entropy principle. The optimized VMD then decomposes PV power, while the TCN-GRU model harnesses TCN’s proficiency in learning local temporal features and GRU’s capability in rapidly modeling sequence data, while leveraging multi-head attention to better utilize the global correlation information within sequence data. Through this design, the model adeptly captures the correlations within time series data, demonstrating superior performance in prediction tasks. Subsequently, the SSA is employed to optimize GRU parameters, and the decomposed PV power mode components and environmental feature attributes are inputted into the TCN-GRU neural network. This facilitates dynamic temporal modeling of multivariate feature sequences. Finally, the predicted values of each component are summed to realize PV power forecasting. Validation using real data from a PV station corroborates that the novel model demonstrates a substantial reduction in RMSE and MAE of up to 55.1% and 54.5%, respectively, particularly evident in instances of pronounced photovoltaic power fluctuations during inclement weather conditions. The proposed method exhibits marked improvements in accuracy compared to traditional PV power prediction methods, underscoring its significance in enhancing forecasting precision and ensuring the secure scheduling and stable operation of power systems.

Improving Urban Mapping Accuracy: Investigating the Role of Data Acquisition Methods and SfM Processing Modes in UAS-Based Survey Through Explainable AI Metrics

In this study, we investigated the accuracy of surface models and orthophoto mosaics generated from images acquired using different data acquisition methods at different processing levels in two urban study areas with different characteristics. Experimental investigations employed single- and double-grid flight directions with nadir and tilted (60°) camera angles, alongside the Perimeter 3D method. Three processing levels (low, medium, and high) were applied using SfM software, resulting in 42 models. Ground truth data from RTK GNSS points and aerial LiDAR surveys were used to assess horizontal and vertical accuracies. For the horizontal accuracy test, neither the oblique camera angle nor the double grid resulted in an improvement in accuracy. In contrast, when examining the vertical accuracy, it was concluded that for several processing levels, the tilted camera angle yielded better results, and in these cases, the double grid also improved accuracy. Feature importance analysis revealed that, among the four variables, the data acquisition method was the most important factor affecting accuracy in two out of three cases.

A Dual-FSM GI LiDAR Imaging Control Method Based on Two-Dimensional Flexible Turntable Composite Axis Tracking

In this study, a tracking and pointing control system with a dual-FSM (fast steering mirror) two-dimensional flexible turntable composite axis is proposed. It is applied to the target-tracking accuracy control in a GI LiDAR (ghost imaging LiDAR) system. Ghost imaging is a multi-measurement imaging method; the dual-FSM GI LiDAR tracking and pointing imaging control system proposed in this study mainly solves the problems of the high-resolution remote sensing imaging of high-speed moving targets and various nonlinear disturbances when this technology is transformed into practical applications. Addressing the detrimental effects of nonlinear disturbances originating from internal flexible mechanisms and assorted external environmental factors on motion control’s velocity, stability, and tracking accuracy, a nonlinear active disturbance rejection control (NLADRC) method based on artificial neural networks is advanced. Additionally, to overcome the limitations imposed by receiving aperture constraints in GI LiDAR systems, a novel optical path design for the dual-FSM GI LiDAR tracking and imaging system is put forth. The implementation of the described methodologies culminated in the development of a dual-FSM GI LiDAR tracking and imaging system, which, upon thorough experimental validation, demonstrated significant improvements. Notably, it achieved an improvement in the coarse tracking accuracy from 193.29 μrad (3σ) to 87.21 μrad (3σ) and enhanced the tracking accuracy from 10.1 μrad (σ) to 1.5 μrad (σ) under specified operational parameters. Furthermore, the method notably diminished the overshoot during the target capture process from 28.85% to 12.8%, concurrently facilitating clear recognition of the target contour. This research contributes significantly to the advancement of GI LiDAR technology for practical application, showcasing the potential of the proposed control and design strategies in enhancing system performance in the face of complex disturbances.

Large-scale DSM registration via motion averaging

Abstract. Generating wide-area digital surface models (DSMs) requires registering a large number of individual, and partially overlapped DSMs. This presents a challenging problem for a typical registration algorithm, since when a large number of observations from these multiple DSMs are considered, it may easily cause memory overflow. Sequential registration algorithms, although can significantly reduce the computation, are especially vulnerable for small overlapped pairs, leading to a large error accumulation. In this work, we propose a novel solution that builds the DSM registration task as a motion averaging problem: pair-wise DSMs are registered to build a scene graph, with edges representing relative poses between DSMs. Specifically, based on the grid structure of the large DSM, the pair-wise registration is performed using a novel nearest neighbor search method. We show that the scene graph can be optimized via an extremely fast motion average algorithm with O(N) complexity (N refers to the number of images). Evaluation of high-resolution satellite-derived DSM demonstrates significant improvement in computation and accuracy.

Multivariate regional deep learning prediction of soil properties from near-infrared, mid-infrared and their combined spectra

Artificial neural network (ANN) models have been successfully used in infrared spectroscopy research for the prediction of soil properties. They often show better performance than conventional methods such as partial least squares regression (PLSR). In this paper we develop and evaluate a multivariate extension of ANN for predicting correlated soil properties: total carbon (C), total nitrogen (N), clay, silt, and sand contents, using visible near-infrared (vis-NIR), mid-infrared (MIR) or combined spectra (vis-NIR + MIR). We hypothesize that accounting for the correlation through joint modelling of soil properties with a single model can eliminate “pedological chimera”: unrealistic values that may arise when properties are predicted independently such as when calculating ratio or soil texture values. We tested two types of ANN models, a univariate (ANN-UV) and a multivariate model (ANN-MV), using a dataset of 228 soil samples collected from Murehwa district in Zimbabwe at two soil depth intervals (0–20 and 20–40 cm). The models were compared with results from a univariate PLSR (PLSR-UV) model. We found that the multivariate ANN model was better at conserving the observed correlations between properties and consequently gave realistic soil C:N and C:Clay ratios, but that there was no improvement in prediction accuracy over using a univariate model (ANN or PLSR). The use of combined spectra (vis-NIR + MIR) did not make any significant improvements in prediction accuracy of the multivariate ANN model compared to using the vis-NIR or MIR only. We conclude that the multivariate ANN model is better suited for the prediction of multiple correlated soil properties and that it is flexible and can account for compositional constrains. The multivariate ANN model helps to keep realistic ratio values – with strong implications for assessment studies that make use of such predicted soil values.

On-Orbit Geometric Calibration of the HJ-2 A/B Satellites' Infrared Sensors

Abstract. With the advancement of China's satellite technology, the HuanJingJianZai-2 A/B (HJ-2 A/B) satellites, equipped with whisk-broom infrared sensors, represent a significant leap forward in environmental monitoring and Earth observation capabilities. This technological leap, however, introduces new challenges in calibration. The unique structure of the HJ-2 A/B infrared spectroradiometer (IRS) necessitates innovative calibration techniques, as traditional methods primarily focused on exterior orientation parameters (EOPs) and often overlooked the importance of interior orientation accuracy, which is essential for accurate multispectral band registration and color rendering. Addressing this gap, we introduce an innovative multi-focal-plane-array joint calibration method specifically designed for whisk-broom cameras. Our method involves selecting a master band from each focal plane array for accurate focal length calibration and deriving ground control points from image matching and altitude interpolation for comprehensive bundle adjustment. This adjustment refines EOPs and interior orientation parameters (IOPs), ensuring globally optimal EOPs and enhanced IOPs calibration stability. The application of our method to the HJ-2 A/B IRS yielded substantial improvements in georeferencing and band registration accuracies, surpassing traditional methods. This paper details the multi-focal-plane-array joint calibration method, describes the IRS and experimental setup, presents the experimental results, and concludes with the implications and potential applications of our findings.

Transmission Line Fault Classification Using Conformer Convolution-Augmented Transformer Model

Ensuring a consistently reliable power supply is paramount in power systems. Researchers are engaged in the pursuit of categorizing transmission line failures to design countermeasures for mitigating the associated financial losses. Our study employs a machine learning-based methodology, specifically the Conformer Convolution-Augmented Transformer model, to classify transmission line fault types. This model processes time series input data directly, eliminating the need for expert feature extraction. The training and validation datasets are generated through simulations conducted on a two-terminal transmission line, while testing is conducted on historical data consisting of 108 events that occurred in the Taiwan power system. Due to the limited availability of historical data, they are utilized solely for inference purposes. Our simulations are meticulously designed to encompass potential faults based on an analysis of historical data. A significant aspect of our investigation focuses on the impact of the sampling rate on input data, establishing that a rate of four samples per cycle is sufficient. This suggests that, for our specific classification tasks, relying on lower frequency data might be adequate, thereby challenging the conventional emphasis on high-frequency analysis. Eventually, our methodology achieves a validation accuracy of 100%, although the testing accuracy is lower at 88.88%. The discrepancy in testing accuracy can be attributed to the limited information and the small number of historical events, which pose challenges in bridging the gap between simulated data and real-world measurements. Furthermore, we benchmarked our method against the ELM model proposed in 2023, demonstrating significant improvements in testing accuracy.

Sandwich boosting for accurate estimation in partially linear models for grouped data

Abstract We study partially linear models in settings where observations are arranged in independent groups but may exhibit within-group dependence. Existing approaches estimate linear model parameters through weighted least squares, with optimal weights (given by the inverse covariance of the response, conditional on the covariates) typically estimated by maximizing a (restricted) likelihood from random effects modelling or by using generalized estimating equations. We introduce a new ‘sandwich loss’ whose population minimizer coincides with the weights of these approaches when the parametric forms for the conditional covariance are well-specified, but can yield arbitrarily large improvements in linear parameter estimation accuracy when they are not. Under relatively mild conditions, our estimated coefficients are asymptotically Gaussian and enjoy minimal variance among estimators with weights restricted to a given class of functions, when user-chosen regression methods are used to estimate nuisance functions. We further expand the class of functional forms for the weights that may be fitted beyond parametric models by leveraging the flexibility of modern machine learning methods within a new gradient boosting scheme for minimizing the sandwich loss. We demonstrate the effectiveness of both the sandwich loss and what we call ‘sandwich boosting’ in a variety of settings with simulated and real-world data.

Development and Deployment of a High-Precision Power Factor Measurement Meter

This research article presents the development and deployment of a high-accuracy power factor measurement meter. Power factor measurement is crucial for assessing the efficiency of electrical systems and ensuring optimal power usage. The proposed meter employs advanced signal processing techniques to accurately measure both leading and lagging power factors with high precision. The design incorporates robust hardware components such as ATMega8 micro- controlling chip and sophisticated algorithms to minimize errors and provide reliable measurements even in challenging operating conditions. Experimental results demonstrate the effectiveness of the proposed meter in accurately measuring power factor across a wide range of electrical loads. The developed meter offers significant improvements in accuracy and reliability compared to existing solutions, making it suitable for various industrial and commercial applications where precise power factor measurement is essential for efficient energy management.

Closed-Loop Transcranial Electrical Neurostimulation for Sustained Attention Enhancement: A Pilot Study towards Personalized Intervention Strategies

Sustained attention is pivotal for tasks like studying and working for which focus and low distractions are necessary for peak productivity. This study explores the effectiveness of adaptive transcranial direct current stimulation (tDCS) in either the frontal or parietal region to enhance sustained attention. The research involved ten healthy university students performing the Continuous Performance Task-AX (AX-CPT) while receiving either frontal or parietal tDCS. The study comprised three phases. First, we acquired the electroencephalography (EEG) signal to identify the most suitable metrics related to attention states. Among different spectral and complexity metrics computed on 3 s epochs of EEG, the Fuzzy Entropy and Multiscale Sample Entropy Index of frontal channels were selected. Secondly, we assessed how tDCS at a fixed 1.0 mA current affects attentional performance. Finally, a real-time experiment involving continuous metric monitoring allowed personalized dynamic optimization of the current amplitude and stimulation site (frontal or parietal). The findings reveal statistically significant improvements in mean accuracy (94.04 vs. 90.82%) and reaction times (262.93 vs. 302.03 ms) with the adaptive tDCS compared to a non-stimulation condition. Average reaction times were statistically shorter during adaptive stimulation compared to a fixed current amplitude condition (262.93 vs. 283.56 ms), while mean accuracy stayed similar (94.04 vs. 93.36%, improvement not statistically significant). Despite the limited number of subjects, this work points out the promising potential of adaptive tDCS as a tailored treatment for enhancing sustained attention.

Accuracy improvement of two-dimensional shape reconstruction based on OFDR using first-order differential local filtering

The accuracy of two-dimensional (2D) shape reconstruction is highly susceptible to fake peaks in the strain distribution measured by optical frequency domain reflectometry (OFDR). In this paper, a post-processing method using first-order differential local filtering is proposed to suppress fake peaks and further improve the accuracy of shape reconstruction. By analyzing the principles of 2D shape reconstruction, an explanation of how fake peaks lead to shape reconstruction errors is provided, along with the introduction of an error evaluation standard. The principle of first-order differential local filtering is presented, and its feasibility is verified by simulation. An OFDR 2D shape reconstruction system is built, with three groups of 2D shape reconstruction experiments carried out, including up bending, down bending and arch bending. The experimental results show that the end errors of the three groups of shape reconstruction are respectively reduced from 2.33%, 2.97%, and 1.07% to 0.25%, 0.78%, and 0.20%, at the shape reconstruction length of 0.5 m. The research demonstrates that the accuracy of OFDR 2D shape reconstruction can be improved by using first-order differential local filtering.