Object Recognition with Class Conditional Gaussian Mixture Model - A Statistical Learning Approach

High viscosity of quench oil is a critical problem of quench system in ethylene cracking furnaces in petrochemical plant, due to its influences on the safety and stability of equipments, meanwhile, the variety of viscosity of quench oil has a negative impact on yield of ethylene and other chemical products. This paper presents a new statistical learning model to forecast the real-time variety of quench oil viscosity based on statistical algorithm and machine learning method. Firstly, statistical algorithm is applied to reduce dimension of parameters, secondly, fitting real-time predictive model through machine learning method. The simulation results shows that this model can monitor the variety of viscosity per hour according to identified controllable parameters which are highly correlated with viscosity of quench oil.

Setting up workable budgets symbolizes the competence of state highway agencies (SHAs) in fulfilling their responsibilities, and unreliable cost estimates can cause economic and political complications. The unclear scope definition and scarcity of project information available at early stages make it hard to generate reliable preliminary estimates. Hence, based on the 1,249 projects retrieved from the Florida Department of Transportation (FDOT) database, this research aimed to develop a cost estimation model using statistical learning methods for SHAs to forecast preliminary costs during the early stages of a transportation project to fulfill different cost control and managerial functions. However, the currently used methods have serious limitations. This study introduced alternative statistical learning approaches to the currently most used methods: least absolute shrinkage and selection operator (LASSO) and general regression neural network (GRNN). LASSO regression, for instance, has proved in other areas of science to be remarkably better in terms of variable selection, interpretability, and numerical stability. In addition, this study also accounted for economic factors in model development because economic conditions are influential on highway construction costs but have received limited attention. Using the same dataset, LASSO and GRNN models were developed, and then their performances were evaluated based on a set of criteria, e.g., the mean absolute error and mean absolute percentage error. In comparison to the current practice with state DOTs, this research contributes to the body of knowledge by introducing a series of objective modeling approaches that can prevent human errors, requiring no substantial experience in preliminary estimating. Besides the introduction of statistical learning methods, this study took economic indicators into account when developing the models because they are important factors but have been ignored in previous studies. In addition, these statistical learning methods can produce reliable estimates in a much faster and more consistent fashion, which is critical, particularly considering the massive workload faced by most SHAs and the allowable time to make a preliminary estimate.

There is growing momentum to develop statistical learning (SL) methods as an alternative to conventional genome-wide association studies (GWAS). Methods such as random forests (RF) and gradient boosting machine (GBM) result in variable importance measures that indicate how well each single-nucleotide polymorphism (SNP) predicts the phenotype. For RF, it has been shown that variable importance measures are systematically affected by minor allele frequency (MAF) and linkage disequilibrium (LD). To establish RF and GBM as viable alternatives for analyzing genome-wide data, it is necessary to address this potential bias and show that SL methods do not significantly under-perform conventional GWAS methods. Both LD and MAF have a significant impact on the variable importance measures commonly used in RF and GBM. Dividing SNPs into overlapping subsets with approximate linkage equilibrium and applying SL methods to each subset successfully reduces the impact of LD. A welcome side effect of this approach is a dramatic reduction in parallel computing time, increasing the feasibility of applying SL methods to large datasets. The created subsets also facilitate a potential correction for the effect of MAF using pseudocovariates. Simulations using simulated SNPs embedded in empirical data-assessing varying effect sizes, minor allele frequencies and LD patterns-suggest that the sensitivity to detect effects is often improved by subsetting and does not significantly under-perform the Armitage trend test, even under ideal conditions for the trend test. Code for the LD subsetting algorithm and pseudocovariate correction is available at http://www.nd.edu/~glubke/code.html.

The purpose of this work is to explore the design principles for a Real-Time Robotic Multi Camera Vision System, in a case study involving a real world competition of autonomous driving. Design practices from vision and real-time research areas are applied into a Real-Time Robotic Vision application, thus exemplifying good algorithm design practices, the advantages of employing the “zero copy one pass” methodology and associated trade-offs leading to the selection of a controller platform. The vision tasks under study are: (i) recognition of a “flat” signal; and (ii) track following, requiring 3D reconstruction. This research firstly improves the used algorithms for the mentioned tasks and finally selects the controller hardware. Optimization for the shown algorithms yielded from 1.5 times to 190 times improvements, always with acceptable quality for the target application, with algorithm optimization being more important on lower computing power platforms. Results also include a 3-cm and five-degree accuracy for lane tracking and 100% accuracy for signalling panel recognition, which are better than most results found in the literature for this application. Clear results comparing different PC platforms for the mentioned Robotic Vision tasks are also shown, demonstrating trade-offs between accuracy and computing power, leading to the proper choice of control platform. The presented design principles are portable to other applications, where Real-Time constraints exist.

A robot vision system can be called humanoid if it possesses an oculomotor system similar to human eyes and if it is capable to simultaneously acquire and process images of varying resolution. Designers of a number of humanoid robots attempted to mimic the foveated structure of the human eye. Foveation is useful because, firstly, it enables the robot to monitor and explore its surroundings in images of low resolution, thereby increasing the efficiency of the search process, and secondly, it makes it possible to simultaneously extract additional information – once the area of interest is determined – from higher resolution foveal images that contain more detail. There are several visual tasks that can benefit from foveated vision. One of the most prominent among them is object recognition. General object recognition on a humanoid robot is difficult because it requires the robot to detect objects in dynamic environments and to control the eye gaze to get the objects into the fovea and to keep them there. Once these tasks are accomplished, the robot can determine the identity of the object by processing foveal views. Approaches proposed to mimic the foveated structure of biological vision systems include the use of two cameras per eye (Atkeson et al., 2000; Breazeal et al., 2001; Kozima & Yano, 2001; Scassellati, 1998) (Cog, DB, Infanoid, Kismet, respectively), i. e. a narrow-angle foveal camera and a wide-angle camera for peripheral vision; lenses with space-variant resolution (Rougeaux & Kuniyoshi, 1998) (humanoid head ESCHeR), i. e. a very high definition area in the fovea and a coarse resolution in the periphery; and space-variant log-polar sensors with retina-like distribution of photo-receptors (Sandini &Metta, 2003) (Babybot). It is also possible to implement log-polar sensors by transforming standard images into log-polar ones (Engel et al., 1994), but this approach requires the use of high definition cameras to get the benefit of varying resolution. Systems with zoom lenses have some of the advantages of foveated vision, but cannot simultaneously acquire wide angle and high resolution images. Our work follows the first approach (see Fig. 1) and explores the advantage of foveated vision for object recognition over standard approaches, which use equal resolution across the visual field. While log-polar sensors are a closer match to biology, we note that using two cameras per eye can be advantageous because cameras with standard chips can be utilized. This makes it possible to equip a humanoid robot with miniature cameras (lipstick size and 1

The Saudi population is at high risk of multimorbidity. The risk of these morbidities can be reduced by identifying common modifiable behavioural risk factors. This study uses statistical and machine learning methods to predict factors for multimorbidity in the Saudi population. Data from 23,098 Saudi residents were extracted from the "Sharik" Health Indicators Surveillance System 2021. Participants were asked about their demographics and health indicators. Binary logistic models were used to determine predictors of multimorbidity. A backpropagation neural network model was further run using the predictors from the logistic regression model. Accuracy measures were checked using training, validation, and testing data. Females and smokers had the highest likelihood of experiencing multimorbidity. Age and fruit consumption also played a significant role in predicting multimorbidity. Regarding model accuracy, both logistic regression and backpropagation algorithms yielded comparable outcomes. The backpropagation method (accuracy 80.7%) was more accurate than the logistic regression model (77%). Machine learning algorithms can be used to predict multimorbidity among adults, particularly in the Middle East region. Different testing methods later validated the common predicting factors identified in this study. These factors are helpful and can be translated by policymakers to consider improvements in the public health domain.

This study investigates the mechanical properties of aluminum-lithium (Al-Li) preforms via statistical and machine learning (ML) methods, with a focus on optimizing processing parameters to achieve enhanced strength and conducts a comparative analysis between the best performing ML approach. Al-Li alloys are recognized for their high strength-to-weight ratio, making them highly suitable for aerospace applications where weight reduction is a critical design consideration. However, the challenge lies in refining processing conditions to enhance mechanical performance while maintaining manufacturability due to the inherent anisotropic nature of Al-Li alloys. The study employs response surface methodology (RSM) to analyze the effects of lithium content (0–2 wt%), aspect ratio (0.2–0.6), relative density (82%–86%), and lubrication conditions (none versus zinc stearate) on key mechanical properties. The results indicate that an intermediate lithium content of 1 wt% optimizes mechanical strength, including yield strength, tensile strength, and elastic modulus, whereas pure aluminum (0 wt% lithium) exhibits superior ductility and workability. Furthermore, higher aspect ratios (0.6) contribute to increased strength, while lower aspect ratios (0.2) enhance strain hardening and densification. The study also reveals that forging without lubrication enhances densification and mechanical strength due to increased internal stresses. Predictive models are also developed using machine learning approach whereby artificial neural networks (ANN) demonstrates superior predictive accuracy, achieving up to 92.97% accuracy for fractional theoretical density, owing to its capacity to capture complex nonlinear relationships. However, ANN showed susceptibility to overfitting, particularly when predicting ductility, where its performance on unseen data is less reliable. Conversely, linear regression, while a simpler approach, showed greater stability by mitigating overfitting risks and providing a robust baseline for evaluating linear dependencies. The integration of these methodologies offers a comprehensive framework for optimizing cold forging conditions, thereby contributing to the advancement of high-performance Al–Li alloy manufacturing.

Clinical prediction models estimate an individual's risk of a particular health outcome. A developed model is a consequence of the development dataset and model‐building strategy, including the sample size, number of predictors, and analysis method (e.g., regression or machine learning). We raise the concern that many models are developed using small datasets that lead to instability in the model and its predictions (estimated risks). We define four levels of model stability in estimated risks moving from the overall mean to the individual level. Through simulation and case studies of statistical and machine learning approaches, we show instability in a model's estimated risks is often considerable, and ultimately manifests itself as miscalibration of predictions in new data. Therefore, we recommend researchers always examine instability at the model development stage and propose instability plots and measures to do so. This entails repeating the model‐building steps (those used to develop the original prediction model) in each of multiple (e.g., 1000) bootstrap samples, to produce multiple bootstrap models, and deriving (i) a prediction instability plot of bootstrap model versus original model predictions; (ii) the mean absolute prediction error (mean absolute difference between individuals’ original and bootstrap model predictions), and (iii) calibration, classification, and decision curve instability plots of bootstrap models applied in the original sample. A case study illustrates how these instability assessments help reassure (or not) whether model predictions are likely to be reliable (or not), while informing a model's critical appraisal (risk of bias rating), fairness, and further validation requirements.

PurposeThe purpose of this paper was exploring and comparing different deterioration models based on statistical and machine learning approaches. These models were chosen from their successful results in other case studies. The deterioration models were developing considering two scenarios: (i) only the age as covariate (Scenario 1); and (ii) the age together with other available sewer characteristics as covariates (Scenario 2). Both were evaluated to achieve two different management objectives related to the prediction of the critical condition of sewers: at the network and the sewer levels.Design/methodology/approachSix statistical and machine learning methods [logistic regression (LR), random forest (RF), multinomial logistic regression, ordinal logistic regression, linear discriminant analysis and support vector machine] were explored considering two kinds of predictor variables (independent variables in the model). The main propose of these models was predicting the structural condition at network and pipe level evaluated from deviation analysis and performance curve techniques. Further, the deterioration models were exploring for two case studies: the sewer systems of Bogota and Medellin. These case studies were considered because of both counts with their own assessment standards and low inspection rate.FindingsThe results indicate that LR models for both case studies show higher prediction capacity under Scenario 1 (considering only the age) for the management objective related to the network, such as annual budget plans; and RF shows the highest success percentage of sewers in critical condition (sewer level) considering Scenario 2 for both case studies.Practical implicationsThere is not a deterioration method whose predictions are adaptable for achieving different management objectives; it is important to explore different approaches to find which one could support a sewer asset management objective for a specific case study.Originality/valueThe originality of this paper consists of there is not a paper in which the prediction of several statistical and machine learning-based deterioration models has been compared for case studies with different local assessment standard. The above to find which is adaptable for each one and which model is adaptable for each management objective.

In this article we introduce modern statistical machine learning and bioinformatics approaches that have been used in learning statistical relationships from big data in medicine and behavioral science that typically include clinical, genomic (and proteomic) and environmental variables. Every year, data collected from biomedical and behavioral science is getting larger and more complicated. Thus, in medicine, we also need to be aware of this trend and understand the statistical tools that are available to analyze these datasets. Many statistical analyses that are aimed to analyze such big datasets have been introduced recently. However, given many different types of clinical, genomic, and environmental data, it is rather uncommon to see statistical methods that combine knowledge resulting from those different data types. To this extent, we will introduce big data in terms of clinical data, single nucleotide polymorphism and gene expression studies and their interactions with environment. In this article, we will introduce the concept of well-known regression analyses such as linear and logistic regressions that has been widely used in clinical data analyses and modern statistical models such as Bayesian networks that has been introduced to analyze more complicated data. Also we will discuss how to represent the interaction among clinical, genomic, and environmental data in using modern statistical models. We conclude this article with a promising modern statistical method called Bayesian networks that is suitable in analyzing big data sets that consists with different type of large data from clinical, genomic, and environmental data. Such statistical model form big data will provide us with more comprehensive understanding of human physiology and disease.

Traditional analytical methods use aggregate metrics, which fail to show the fine-grained patterns that emerge from trial-level fluctuations of cognitive performance due to dynamic internal states like attention, fatigue, and learning processes. This research examines how statistical and machine learning (ML) methods can analyse trial-level behavioural and physiological data to improve understanding of cognitive dynamics. A total of 9,276 trials were obtained from 112 participants who completed Stroop, N-back, and Go/No-Go tasks. The annotation of each trial included reaction time measurements alongside accuracy data, task condition information, and EEG-derived alpha power measurements. Our analysis incorporated Bayesian hierarchical models, generalized linear mixed models, state-space models, Random Forest, XGBoost, deep neural networks, and Long Short-Term Memory (LSTM) networks to forecast both reaction times and task accuracy. The LSTM model demonstrated the best predictive power by achieving R² = 0.862 for RT prediction and AUC-ROC = 0.925 for accuracy classification. The AUC-ROC score reached 0.925 for classification, while R² reached 0.862 for reaction time prediction, which proved superior to all other techniques. The predictive features of trial number, task congruency, and EEG alpha power emerged through Shapley Additive Explanations (SHAP) and LSTM saliency maps. The research demonstrates how combining statistical transparency with ML flexibility helps reveal personalized and time-dependent cognitive patterns. The proposed method provides a powerful structure for modelling trials while creating potential applications for individualized cognitive assessment systems in educational and mental health settings.

Extreme warm water events, known as marine heatwaves, cause a variety of adverse impacts on the marine ecosystem. They are occurring more and more frequently across the global ocean. Yet monitoring marine heatwaves below the sea surface is still challenging due to the sparsity of in situ temperature observations. Here, we propose a statistical learning method guided by ocean dynamics and optimal prediction theory, to detect subsurface marine heatwaves based on the observable sea surface temperature and sea surface height. This dynamics-guided statistical learning method shows good skills in detecting subsurface marine heatwaves in the oceanic epipelagic zone over many parts of the global ocean. It outperforms both the classical ordinary least square regression and popular deep learning methods that do not effectively exploit ocean dynamics, with clear dynamical interpretation for its outperformance. Our study provides a useful statistical learning method for near real-time monitoring of subsurface marine heatwaves at a global scale and highlights the importance of exploiting ocean dynamics for enhancing the efficiency and interpretability of statistical learning.

This paper describes a statistical learning approach for modeling the performance of single stage missile systems with the speed and fidelity required for conceptual and early preliminary design. Developing a Response Surface Model (RSM) using statistical learning approaches not only enables design from first principles, but it also facilitates the determination of design parameters given system performance. This “two-way” predictive capability further speeds conceptual design while providing the observer with potentially valuable information about a missile using only limited information such as telemetry. For the purposes of demonstrating the overall approach for this paper, models are developed for a single stage liquid propellant system. For this paper, the authors identify the framework necessary to build the RSM consisting of characteristics used to define the liquid propellant missile design and launch settings. These inputs form the essence of a conventional design process. The most important design variables are missile diameter and length, nozzle expansion ratio, nozzle throat diameter, and initial launch angle. In tandem, these inputs are fed into the statistical learning algorithm, which in turn is trained to predict telemetry parameters that include range, apogee, max thrust, and the time of flight (TOF). A physics model is used to generate training data for the statistical learning exercise. This model is described in the paper and is the “glue” that connects the input and output necessary for statistical learning. Furthermore, it is necessary to use design of experiments to develop an appropriate database without agglomerating an excessively massive database. The AU Liquid Rocket Code (AULRC) uses a Latin hypercube design to randomly generate missile designs (inputs) and then simulate the missile thrust, time, and its trajectory. For this effort, the output is transformed into max thrust, TOF, range, and apogee. The bulk of the paper is methodology adopted for the Model development. Multiple software packages can implement foundational elements of the modeling process and models are developed with TensorFlow using Python, Flux using JULIA, Matlab, and SAS. The authors compare and contrast the platforms and their suitability for this problem, but ultimately TensorFlow is the preferred approach and the one used to develop the results for the paper. The performance capabilities of the RSM's from each software program are training time, error history, and the general mean absolute percentage error computed on each output characteristic and are used as the basis for choosing the best tools and models. All models are validated and/or cross-validated, then evaluated based on their performance on independent test data. Statistical/machine learning methods include, among others, deep learning neural networks, linear and non-linear regression (through basis expansion) and ensemble methods. This paper shows that the statistical learning RSM approach can yield results with sufficient fidelity to support conceptual and early preliminary design. Ongoing efforts address occasionally spuriously large errors while most of the output has error not exceeding 1%-2% of accepted values. Finally, the data set requirements and constraints required to produce results of this quality are discussed in the paper.

The objective of this paper is to build an Object Recognition model using ‘TINY YOLO’ and FPGA. Object Recognition is a CV (i. e., Computer Vision) technique for identifying and detecting objects in images or videos. Object recognition is a key output of DL (i. e., Deep Learning) and ML (i. e., Machine Learning). When we humans look at an image or video, we can readily spot people, object, scene and visual details. The aim of this is to teach a computer to do what comes innately to a human being. Object recognition is done with Image Classification, Object Localization and Object Detection that combines these two tasks and localizes and classifies an object in an image. Object Recognition has become a pivotal technology for driverless cars, security purpose, traffic surveillance etc., It is also used in a various application such as disease detection in bio imaging, industrial examination, and robotic vision. Hence, in a nutshell, ‘Object Recognition is the staple of automation industry’.

Object detection and recognition are fundamental capabilities for a mobile robot. Objects are a powerful representation for a variety of tasks including mobile manipulation and inventory tracking. As a result, object-based world representations have seen a great deal of research interest in the last several years. However, these systems usually assume that object recognition is well-solved: they require that accurate recognition be available for every object they might encounter. Despite steady advances, object recognition remains a difficult, open problem. Existing object recognition algorithms rely on high-resolution three-dimensional object models or on extensive hand-labeled training data. The sheer variety of objects that occur in natural environments makes manually training a recognizer for every possible object infeasible. In this work, we present a robotic system for unsupervised object and class discovery, in which objects are first discovered, and then grouped into classes in an unsupervised fashion. At each step, we approach the problem as one of robotics, not disembodied computer vision. On a very large robotic dataset, we discover object classes with 98.7% precision while achieving 71.8% recall. The scale and quality of these results demonstrate the merit of our approach, and prove the practicality of longterm large-scale object discovery. To our knowledge, no other authors have investigated robotic object discovery at this scale, making direct quantitative comparison impossible. We make our implementation and ground-truth labelings available, and evaluate our technique on a very large dataset. As a result, this work is a baseline against which future work can be compared.