Robust Identification Method Research Articles

Automatic detection of grains in partially recrystallized microstructures using deep learning

Precise identification of recrystallizing grains in partially recrystallized microstructures is essential to obtain quantitative information regarding the recrystallization process. Automatic, robust, user-friendly, and unbiased identification methods that do not rely on hard-coded, preselected values would be highly advantageous. In this study, we test convolutional neural network instance segmentation models to achieve automatic segmentation of individual recrystallizing grains in partially recrystallized microstructures. Our training dataset includes micrographs obtained using electron backscattered diffraction from five alloys with different thermal-mechanical histories and more than 100,000 recrystallizing grains. We adapt and train two state of the art deep learning models, namely Mask R-CNN and PointRend. Both models provide instance segmentation results of good quality, enabling quantitative determination of the microstructural parameters. The PointRend model demonstrates better performance for grains with irregular shapes than Mask R-CNN. Compared to conventional methods, the trained deep learning approach is easier to use, more flexible, and applicable to a wide range of materials.

An improved robust identification method for position independent geometric errors of the swing axis of the gear grinding machine

An improved robust identification method for position independent geometric errors of the swing axis of the gear grinding machine

Automated rice mapping using multitemporal Sentinel-1 SAR imagery using dynamic threshold and slope-based index methods

Rice cultivation plays a crucial role in food security and economic development, particularly in regions like India, due to its vast population and position as the top rice producer globally. This work introduces a novel framework, the Rice Mapping Method (RMM), which leverages Multitemporal Sentinel-1 Synthetic Aperture Radar (SAR) imagery for automated rice mapping. Contrary to the traditional approaches, RMM combines the Dynamic Threshold Method (DTM) for robust rice field identification and a slope-based index for classifying single and double cropping practices. By analyzing VH backscatter patterns and employing specific thresholds, DTM separates rice pixels from the other background pixels. The DTM, which relies on VH backscatter values during the growing season, has been tested across various rice cultivation landscapes, demonstrating high accuracy upto 0.95. DTM is also tested on different rice-growing areas such as the hilly Kodagu district, with an F1 Score of 0.96, and in the flooded delta region of Kuttanad, achieving an F1 Score of 0.93. The Slope-based Index I(r,c) is introduced to differentiate the single and double cropping pixels by calculating the index for the second season of cropping and gives F1 Score of 0.81. The DTM’s effectiveness in rice field identification is evaluated by comparing it to the classification of the Bi-directional Gated Recurrent Unit (Bi-GRU) network. Similarly, the Slope-based Index is compared with other established automated rice mapping methods to assess its accuracy in distinguishing cropping patterns. RMM was successfully applied in mapping rice-growing areas in the Udupi district for 2021, estimating Kharif and Rabi season areas, the estimated rice area is compared to official statistics by the Directorate of Economics and Statistics, Karnataka State. The proposed RMM approach offers a robust solution for mapping rice fields, particularly in regions with complex cropping landscapes, and enhances agricultural monitoring and decision-making processes contributing to sustainable rice production and food security initiatives.

Modal identification of wind turbine tower based on optimal fractional order statistical moments

AbstractIn vibration testing of civil engineering structures, the first two vibration modes are crucial in representing the global dynamic behavior of the structure measured. In the present study, a comprehensive method is proposed to identify the first two vibration modes of wind turbine towers, which is based on the analysis of fractional order statistical moments (FSM). This study offers novel contributions in two key aspects: (1) theoretical derivations of the relationship between FSM and vibration mode; and (2) successful use of 32/7‐order displacement statistical moment as the optimal FSM to identify wind turbine tower modes, by combining with noise resistance analysis, sensitivity analysis, and stability analysis, respectively. Using the proposed method, the FSM was first used to identify the modal vibration of wind turbine towers. By obtaining the response of the structure on the same vertical line, FSM was then calculated to estimate the corresponding structural modal vibration. Considering other influencing factors in the field test, the modal identification results of this index under different excitation forms and noise conditions were analyzed based on numerical simulation and verified with field wind tower test data. The results of the evaluation show that the proposed statistical moments of can accurately identify the first two vibration modes of wind turbine towers. This presents a new robust method for modal vibration identification, that is, simple and effective in its implementation.

A robust force identification method based on L1+2-norm approximation theory

Obtaining the real force of the structure is the basic condition for efficient vibration reduction design. However, the uneasiness and instability of the inverse problem make the force identification still challenging. On the basis of L2-norm regularization, we propose a new method for force identification based on L1+2-norm approximation theory. By introducing the L1-norm penalty operator, a joint norm approximation optimization model is constructed to limit the non-zero force and punish the force value which approaching to zero, so that it can adapt to the identification of various types of forces. Then the optimal force solution is obtained by using the Primal-dual interior point method. Finally, the effectiveness and accuracy of the proposed method are verified by numerical simulation and experiment of asymmetric plate and double-layer plate structures. Different from previous studies, the L1+2-norm method in this work has high robustness, considers the characteristics of sparse regularization and L2-norm regularization, and can adapt to the identification of continuous forces and sparse forces at the same time, which has potential engineering application value.

Species identification of Panax ginseng throughout the entire industrial chain: From the ginseng field to highly processed products

Panax ginseng has been highly treasured in Asian countries due to its long history, more types of pharmaceutically active ginsenosides, and prominent pharmacological effects. However, the potential adulteration of P. ginseng with P. quinquefolius or P. notoginseng poses serious health risks to consumers and disrupts market stability within the ginseng industry. To address this challenge, a simple real-time PCR assay for rapid screening of field P. ginseng and a reliable multiplex PCR method for species identification of highly processed ginseng products were respectively established, based on the InDel markers derived from nuclear and mitochondrial intron sequences. The real-time PCR assay, coupled with a simple NaOH-Tris DNA extraction method, facilitates high-throughput screening of field ginseng samples based on the Tm values of PCR products with a single pair of primers. The multiplex PCR assay, targeting the mitochondrial nad1 and nad5 InDel markers, could detect exogenous adulteration of P. quinquefolius or P. notoginseng at a remarkable detection limit of 0.001 ng and successfully identified the original species in various forms of highly processed ginseng products. The results indicate that mitochondrial InDel markers resist DNA degradation better than nuclear ones, making them more suitable for species identification in highly processed ginseng products that contain trace amounts of degraded DNA. The present study provides robust InDel marker-based methods for species identification across the entire ginseng industrial chain, from the ginseng field to final processed products, thereby serving as reliable tools for ensuring the authenticity of P. ginseng products.

RDR100: A Robust Computational Method for Identification of Krüppel-like Factors

Background: Krüppel-like factors (KLFs) are a family of transcription factors containing zinc fingers that regulate various cellular processes. KLF proteins are associated with human diseases, such as cancer, cardiovascular diseases, and metabolic disorders. The KLF family consists of 18 members with diverse expression profiles across numerous tissues. Accurate identification and annotation of KLF proteins is crucial, given their involvement in important biological functions. Although experimental approaches can identify KLF proteins precisely, large-scale identification is complicated, slow, and expensive. Methods: In this study, we developed RDR100, a novel random forest (RF)-based framework for predicting KLF proteins based on their primary sequences. First, we identified the optimal encodings for ten different features using a recursive feature elimination approach, and then trained their respective model using five distinct machine learning (ML) classifiers. Results: The performance of all models was assessed using independent datasets, and RDR100 was selected as the final model based on its consistent performance in cross-validation and independent evaluation. Conclusion: Our results demonstrate that RDR100 is a robust predictor of KLF proteins. RDR100 web server is available at https://procarb.org/RDR100/.

Photovoltaic Model Parameters Estimation Via the Fully Informed Search Algorithm

Effective parameter estimation for photovoltaic (PV) systems holds significant importance for both researchers and industry professionals. An accurate understanding of PV models, achieved through modeling and simulation, plays a pivotal role in optimizing the design, control, testing, and forecasting of PV system performance. Developing a precise and robust parameter identification method significantly contributes to enhancing the modeling, control, and optimization of photovoltaic systems. In this context, our research contribution introduces a novel version of Rao metaheuristic algorithm named the Fully Informed Search Algorithm (FISA). Which demonstrate acceptable performance to solving optimization problems in several applied fields. While, maintaining the simplicity and non-parametric nature of the original algorithm. The proposed algorithm holds promise for various industrial applications, particularly in optimizing complex systems such as photovoltaic (PV) systems. For which, we used it to efficiently identifying the parameters of the single-diode model (SDM). Thus, we demonstrate its effectiveness through the application in two distinct case studies within our simulation research. in the end, we compared the results of FISA algorithm to seven other well-known algorithms, the obtained results indicate the superiority of the proposed algorithm in term of the stability of the system, a faster convergence and higher accuracy.

Shiba: A unified computational method for robust identification of differential RNA splicing across platforms.

Alternative pre-mRNA splicing (AS) is a fundamental regulatory process that generates transcript diversity and cell type variation. We developed Shiba, a robust method integrating transcript assembly, splicing event identification, read counting, and statistical analysis, to efficiently quantify exon splicing levels across various types of RNA-seq datasets. Compared to existing pipelines, Shiba excels in capturing both annotated and unannotated or cryptic differential splicing events with superior accuracy, sensitivity, and reproducibility. Furthermore, Shiba's unique consideration of junction read imbalance and exon-body read coverage reduces false positives, essential for downstream functional analyses. We have further developed scShiba for single-cell/nucleus (sc/sn) RNA-seq data, enabling the exploration of splicing variations in heterogeneous cell populations. Both simulated and real data demonstrate Shiba's robustness across multiple sample sizes, including n=1 datasets and individual cell clusters from scRNA-seq. Application of Shiba on single replicates of RNA-seq identified new AS-NMD targets, and scShiba on snRNA-seq revealed intricate temporal AS regulation in dopaminergic neurons. Both Shiba and scShiba are provided in Docker/Singularity containers and Snakemake pipeline, enhancing accessibility and reproducibility. The comprehensive capabilities of Shiba and scShiba allow systematic and robust quantification of alternative splicing events, laying a solid foundation for mechanistic exploration of functional complexity in RNA splicing.

Parameter identification of Yoshida–Uemori combined hardening model by using a variable step size firefly algorithm

Abstract The material behavior under cyclic loading is more complex than under monotonic loading and the usage of the sophisticated constitutive models is required to accurately define the elastoplastic behaviors of the advanced high-strength steels and aluminum alloys. These models involve the numerous material parameters that are determined from cyclic tests and accurate calibration of the variables has a great influence on the description of the material response. Therefore, the development of a precise and robust identification method is needed to obtain reliable results. In this study, a systematic methodology depending upon the firefly algorithm (FA) with variable step size has been developed and Yoshida–Uemori combined hardening model parameters of a dual-phase steel (DP980) and an aluminum alloy (AA6XXX-T4) are determined. The identified parameters are verified based on comparisons between the finite element simulations of the cyclic uniaxial tension-compression tests and experimental data and also the search performance of the variable FA is evaluated by comparing it with the standard FA. It is seen from these comparisons that variable FA can easily find and rapidly converge to the global optimum solutions.

Prediction of Nili-Ravi buffalo bull fertility through Fourier harmonic analysis of sperm

Fourier harmonic analysis (FHA) is a robust method for identification of minute changes in sperm nuclear shape that are indicative of reduced fertility. The current study was designed to develop a fertility prediction model for Nili-Ravi buffalo bulls through FHA of sperm. In experiment I, FHA technique was standardized, average sperm nuclear perimeter was measured and sperm nuclear shape plot of buffalo bull was constructed. Sperm of buffalo bulls (n = 10) were stained with YOYO-1 and Hoechst-33342 to differentiate live and dead, and digital images were captured using phase contrast and fluorescent microscopy. The images were analyzed by ImageJ software and 100 sperm/bull were evaluated. The results are described as mean ± SEM values of mean harmonic amplitude (mharm), skewness harmonic amplitude (skharm), kurtosis harmonic amplitude (kurharm) and variance harmonic amplitude (varharm) at Fourier frequencies 0–5 along with the cartesian and polar coordinate plots of buffalo bull sperm. In experiment II, a fertility prediction model was developed based on FHA of buffalo bull sperm. Semen samples of low (n = 6), medium (n = 3) and high (n = 8) fertility bulls were investigated for FHA of sperm and harmonic amplitudes (HA) were generated. Firstly, to determine if live and dead sperm population have unique nuclear shape distribution; the mean, skewness, kurtosis and variance HA 0–5 of 1700 live and 1294 dead spermatozoa of 17 bulls were evaluated. T-test signified a difference in the mharm0 (2.363 ± 0.01 vs. 2.439 ± 0.02), skharm0 (−0.0002 ± 0.07 vs. −0.266 ± 0.09), kurharm0 (−0.156 ± 0.07 vs. 0.260 ± 0.18), kurharm2 (0.142 ± 0.11 vs. 1.031 ± 0.32) and varharm4 (0.109 ± 0.00 vs. 0.082 ± 0.00) of live vs. dead sperm population (p < 0.05). Therefore, 100 live sperm/bull were further evaluated for mean, skewness, kurtosis and variance HA 0–5 values among high (n = 6) and low-fertility (n = 6) groups. Results of T-test showed higher values of mharm2 (0.739 ± 0.01 vs. 0.686 ± 0.00), mharm4 (0.105 ± 0.001 vs. 0.007 ± 0.001), and skharm0 (0.214 ± 0.109 vs. −0.244 ± 0.097) in high vs. low-fertility group (p < 0.05). In next step, five significantly different combinations of discriminant measures between high and low-fertility groups were obtained by discriminant analysis. In conclusion, mharm4, skharm0 and varharm2 correctly identified 91.7 % of bulls into their respective fertility groups, and upon cross validation the value of the canonical correlation was 0.928.

Robust deep learning method for fruit decay detection and plant identification: enhancing food security and quality control.

This paper presents a robust deep learning method for fruit decay detection and plant identification. By addressing the limitations of previous studies that primarily focused on model accuracy, our approach aims to provide a more comprehensive solution that considers the challenges of robustness and limited data scenarios. The proposed method achieves exceptional accuracy of 99.93%, surpassing established models. In addition to its exceptional accuracy, the proposed method highlights the significance of robustness and adaptability in limited data scenarios. The proposed model exhibits strong performance even under the challenging conditions, such as intense lighting variations and partial image obstructions. Extensive evaluations demonstrate its robust performance, generalization ability, and minimal misclassifications. The inclusion of Class Activation Maps enhances the model's capability to identify distinguishing features between fresh and rotten fruits. This research has significant implications for fruit quality control, economic loss reduction, and applications in agriculture, transportation, and scientific research. The proposed method serves as a valuable resource for fruit and plant-related industries. It offers precise adaptation to specific data, customization of the network architecture, and effective training even with limited data. Overall, this research contributes to fruit quality control, economic loss reduction, and waste minimization.

A novel AI-powered method for robust identification of operational phases in refrigerators

Refrigerators are essential appliances in our daily lives, while the accurate identification of their operational phase is crucial for various applications, such as energy management and maintenance. This paper introduces a novel method to predict the operational phase of refrigerators based on power consumption data, exploiting artificial intelligence (AI) and pattern recognition techniques. Specifically, the proposed method initially identifies the driving speed of the compressor (fixed or variable speed) utilizing an Adaptive Boosting, (AdaBoost) classifier. Following that, a Long Short-Term Memory (LSTM) classifier was designed for each type of compressor based on extracted statistical features. This classifier is used to categorize daily power consumption into the different operational phases. Moreover, the proposed AI-driven diagnostic tool facilitates operational-based prescriptive maintenance by addressing periodic and phase-specific maintenance actions while providing homeowners with informed suggestions on alternative refrigerator models that offer greater cost savings and reduced annual energy consumption. The proposed method was tested on real and simulated data for over 20 years, with the AdaBoost classifier achieving an accuracy close to 97% and an F1 score close to 97% and the LSTM nodes achieving an accuracy of approximately 95% in all cases. Finally, the method demonstrated robustness when applied to new datasets, showing comparable performance on unknown data, rendering it a powerful tool for refrigerator management and maintenance.

Identification of partial differential equations from noisy data with integrated knowledge discovery and embedding using evolutionary neural networks

Identification of underlying partial differential equations (PDEs) for complex systems remains a formidable challenge. In the present study, a robust PDE identification method is proposed, demonstrating the ability to extract accurate governing equations under noisy conditions without prior knowledge. Specifically, the proposed method combines gene expression programming, one type of evolutionary algorithm capable of generating unseen terms based solely on basic operators and functional terms, with symbolic regression neural networks. These networks are designed to represent explicit functional expressions and optimize them with data gradients. In particular, the specifically designed neural networks can be easily transformed to physical constraints for the training data, embedding the discovered PDEs to further optimize the metadata used for iterative PDE identification. The proposed method has been tested in four canonical PDE cases, validating its effectiveness without preliminary information and confirming its suitability for practical applications across various noise levels.

Bluetooth Device Identification Using RF Fingerprinting and Jensen-Shannon Divergence.

The proliferation of radio frequency (RF) devices in contemporary society, especially in the fields of smart homes, Internet of Things (IoT) gadgets, and smartphones, underscores the urgent need for robust identification methods to strengthen cybersecurity. This paper delves into the realms of RF fingerprint (RFF) based on applying the Jensen-Shannon divergence (JSD) to the statistical distribution of noise in RF signals to identify Bluetooth devices. Thus, through a detailed case study, Bluetooth RF noise taken at 5 Gsps from different devices is explored. A noise model is considered to extract a unique, universal, permanent, permanent, collectable, and robust statistical RFF that identifies each Bluetooth device. Then, the different JSD noise signals provided by Bluetooth devices are contrasted with the statistical RFF of all devices and a membership resolution is declared. The study shows that this way of identifying Bluetooth devices based on RFF allows one to discern between devices of the same make and model, achieving 99.5% identification effectiveness. By leveraging statistical RFFs extracted from noise in RF signals emitted by devices, this research not only contributes to the advancement of the field of implicit device authentication systems based on wireless communication but also provides valuable insights into the practical implementation of RF identification techniques, which could be useful in forensic processes.

State estimation-based parameter identification for a class of nonlinear fractional-order systems

Parametric identification is an important part of system theory since knowledge of the parameters allows the analysis and control of the system. The aim of this paper is to propose a novel robust (against measurement noise) parameter identification method for a class of nonlinear fractional-order systems. In order to solve the parametric identification we carry out this problem to a state estimation problem, we introduce a Fractional Algebraic Identifiability (FAI) property which allows to represent the system parameters as a function of the inputs and outputs of the system, this parameter identification method provides an on-line identification process (while the system is operating), we also propose a fractional-order differentiator which allows to reduce the effect of measurement noise as well as to provide the estimation of a fractional-order derivative of the system output. Moreover, we use the Mittag–Leffler boundedness to demonstrate the convergence of this method, a different approach for this stability analysis method is given in this paper. Finally, we illustrate the accuracy and robustness of our proposed method by means of the parametric identification of two nonlinear fractional-order systems: a time-varying nonlinear fractional-order system and a nonlinear fractional-order mathematical model of a simple pendulum.

Automatic Identification and Tracking Method of Case-Related Vehicles Based on Computer Vision Algorithm

Abstract Amidst the rapid advancement of artificial intelligence, computer vision algorithms have found extensive applications across various societal sectors. This paper presents the development of an automatic vehicle identification algorithm for crime-related scenarios, leveraging enhancements in computer vision technology. Initially, we refine the multi-scale feature fusion within the YOLOv4 architecture, subsequently substituting the standard convolution in the feature extraction network with depth-separable convolution to minimize parameter computation. Furthermore, we replace the conventional CIOU target localization loss function with EIOU to expedite model convergence. To address the issue of target vehicle detection failures, we incorporate the Kalman filter algorithm, ensuring precise tracking. Our experimental analysis, which utilizes both target detection and multi-target tracking evaluation indices, demonstrates that the modified YOLOv4 algorithm excels in recall, precision, and average IOU metrics compared to other algorithms, with a mean Average Precision (mAP) of 95.68% and an average detection speed of 0.039 seconds per image, satisfying real-time operational criteria. Vehicle tracking efficacy, compared using the YOLOv4-based detector before and after the modifications, shows significant improvements, indicated by a reduction in Identification Switches (IDS) across all video samples. This study introduces a robust method for the accurate identification and localization of vehicles involved in criminal activities, significantly enhancing case resolution efficiency.

Proteomic Analysis of Dictyostelium discoideum by Mass Spectrometry.

The large-scale proteomic analysis of Dictyostelium discoideum has contributed to our understanding of intracellular as well as secreted proteins in this versatile model eukaryote. Mass spectrometry-based proteomic analysis is a robust, sensitive, and rapid analytical method for identification and characterization of proteins extracted from tissues, cells, cell fractions, or pull-down assays. The availability of core facilities which make proteomics inexpensive and easy to do has facilitated a wide range of research projects. In this chapter, we present a simple standard methodology to extract proteins and prepare samples from D. discoideum for mass spectrometry and methods to analyze the identified proteins.

A Robust Specific Emitter Identification Method for CPS Devices Based on Deep Residual Shrinkage Network

A Robust Specific Emitter Identification Method for CPS Devices Based on Deep Residual Shrinkage Network

Robust Recursive Identification of ARX Models Using Beta Divergence

The robust recursive identification method of ARX models is proposed using the beta divergence. The proposed parameter update law suppresses the effect of outliers using a weight function that is automatically determined by minimizing the beta divergence. A numerical example illustrates the efficacy of the proposed method.