Maximum Confidence Research Articles

Greenhouse capsicum detection in thermal imaging: A comparative analysis of a single-shot and a novel zero-shot detector

Accurate detection of capsicum in thermal images plays a pivotal role in greenhouse optimization and agricultural efficiency. However, it poses distinctive challenges stemming from subtle temperature variations and the imperative for precise localization. This study presents a comprehensive approach to thermal capsicum detection in greenhouse environments, addressing significant research gaps through the construction of a meticulously curated dataset comprising 300 thermal images. These images, capturing various occlusion levels, underwent extensive preprocessing and augmentation to enhance diversity and quality. The custom-trained single-shot YOLOv10 model demonstrated impressive performance metrics, achieving precision, recall, F1-score, and mean Average Precision (mAP) of 0.979, 0.712, 0.824, and 0.883, respectively, with a detection speed of 53.92 FPS and the highest confidence score of 91 %. In contrast, the zero-shot Grounding DINO model, hypertuned for capsicum detection in thermal scenes, achieved a maximum confidence score of 60 %, leading to the proposal of a novel SAM-DINO model. SAM-DINO, combining the Segment Anything Model (SAM) for pixel-level segmentation and Grounding self-distillation with no labels (Grounding DINO) for detection, exhibited enhanced confidence scores of 65 %. Comparative testing of YOLOv10 and SAM-DINO on 50 test images revealed YOLOv10′s superiority with precision, recall, and F1-scores of 0.95, 0.87, and 0.91, respectively, compared to SAM-DINO's scores of 0.90, 0.72, and 0.80. Additionally, YOLOv10 demonstrated an inference speed of 19 ms, approximately ten times faster than SAM-DINO. These findings underscore the efficacy of both custom-trained and zero-shot detection models in thermal capsicum detection, offering valuable insights into their respective strengths and limitations in greenhouse agricultural applications.

Joint utilization of positive and negative pseudo-labels in semi-supervised facial expression recognition

Facial expression recognition has obtained significant attention due to the abundance of unlabeled expressions, and semi-supervised learning aims to leverage unlabeled samples sufficiently. Recent approaches primarily focus on combining an adaptive margin and pseudo-labels to extract hard samples and boost performance. However, the instability of pseudo-labels and the utilization of the rest unlabeled samples remain critical challenges. We introduce a stable-positive-single and negative-multiple pseudo-labels (SPS-NM) method to solve the above two challenges. All unlabeled samples are categorized into three groups properly by adaptive confidence margins. When the maximum confidence scores are high and stable enough, the unlabeled samples are attached with positive pseudo-labels. On the contrary, when the confidence scores of unlabeled samples are low enough, the related negative-multi pseudo-labels are attached to these samples. The quality and quantity of classes in negative pseudo-labels are balanced by top-k. Eventually, the remaining unlabeled samples are ambiguous and fail to match their pseudo-labels, but they can still be used to extract valuable features by contrastive learning. We conduct comparative experiments and ablation study on RAF-DB, AffectNet and SFEW datasets to demonstrate that SPS-NM achieves improvement and becomes the state-of-the-art method in facial expression recognition.

Entanglement witness and nonlocality in confidence of measurement from multipartite quantum state discrimination

We consider multipartite quantum state discrimination and provide a specific relation between the properties of entanglement witness and quantum nonlocality inherent in the confidence of measurement. We first provide the definition of the confidence of measurement as well as its useful properties for various types of multipartite measurement. We show that globally-maximum confidence that cannot be achieved by local operations and classical communication strongly depends on the existence of a certain type of entanglement witness. We also provide conditions for an upper bound on maximum of locally-achievable confidence. Finally, we establish a method in terms of entanglement witness to construct quantum state ensembles with nonlocal maximum confidence.

Estimation of Stress–Strength Reliability for Power Transformed Perks Distribution with Applications

The power transformation of the Perks distribution, its survival, and hazard functions are computed. The shapes of hazard function are obtained analytically which shows that this model has increasing and decreasing hazard rates. In the stress–strength reliability estimation, point and interval estimates are computed using maximum likelihood and asymptotic confidence interval methods, respectively. The performance of the estimates is evaluated through the Monte Carlo simulation and the applications of the proposed distribution to three real datasets have been demonstrated.

Maximum information measurement for qubit states

We determine the optimal measurement that maximizes the average information gain about the state of a qubit system. The qubit is prepared in one of two known states with known prior probabilities. To treat the problem analytically we employ the formalism developed for the maximum confidence quantum state discrimination strategy and obtain the POVM which optimizes the information gain for the entire parameter space of the system. We show that the optimal measurement coincides exactly with the minimum-error quantum measurement only for two pure states, or when the two states have the same Bloch radius or they are on the same diagonal of the Bloch disk.

Submarine cable detection using an end-to-end neural network-based magnetic data inversion

Abstract To process magnetic anomaly data, appropriate parameters for field separation, denoising, and Euler deconvolution have to be manually selected. The traditional workflow is inefficient and cannot fulfill the rapid detection of submarine cables due to the complex processing and manual parameter tuning. This study presents an end-to-end deep learning approach for the identification and positioning of submarine cables based on magnetic anomalies. The proposed approach effectively establishes a direct mapping correlation between the magnetic field data and the position of the submarine cable. Synthetic tests suggest that our method has a better performance in terms of positioning accuracy than the conventional Euler method. Our results for the field data are comparable to those obtained using conventional techniques. Furthermore, the proposed method achieves an optimal solution by employing clustering technique and selecting the solution with the maximum confidence, which avoids spurious solutions associated with traditional methods. The proposed method can directly determine the position of the submarine cables using the raw magnetic field data. Contrary to the traditional processing workflow, field separation and denoising are not necessary in this novel approach, resulting in higher processing efficiency and a simpler processing process.

Blockchain Enhanced Student Physical Performance Analysis using Machine Learning-IoT and Apriori Algorithm in Physical Education Network Teaching

In the digital era, particularly with the rise of online teaching, traditional approaches to college physical education face challenges in adequately monitoring and enhancing students’ physical fitness. This study introduces a novel approach that integrates blockchain technology with a Machine Learning-IoT framework to evaluate and improve students' physical performance. Utilizing the Apriori algorithm, enhanced with particle swarm optimization and an improved K-means methodology, this system offers a robust tool for correlating student behavior with sports performance in a secure and decentralized manner. The proposed system uses blockchain for safe data management and IoT for real-time data collection, ensuring privacy as well as efficiency. The algorithm's accuracy, recall, and F1 values on the Iris dataset are 0.947, 0.931, and 0.928, respectively, with a considerable Calinski Harabasz score of more than 240. When applied to university student behavior data, the blockchain-enhanced system successfully mined association rules with a maximum confidence level of 0.923.

Off-Road Terrain Identification And Analysis

Background: The role of the Terrain is paramount for any autonomous vehicle to drive safely on any type of surface. The Autonomous vehicles should have the capability of identifying the terrain and should adapt to the environment. With the evolution of robotics and Artificial Intelligence, and understanding diverse terrains, the techniques for terrain identification are also advancing with a major focus on safety.
 Methodology: To make Terrain Detection and Identification more reliable we used instance segmentation which is a more sophisticated type of segmentation that goes a step ahead of semantic segmentation by performing both object detection and segmentation at the same time. In order to perform Instance segmentation, we used the YOLOv8 architecture which is considered to be the state-of-the-art CNN (Convolutional Neural Network) architecture. The YOLOv8 model was trained on an Off-road Terrain Dataset.
 Results: Our findings indicate that the state-of-the-art YOLOv8 instance segmentation model provided the best results for terrain detection and segmentation with a threshold confidence of 0.60, and the results provide a maximum confidence of 0.92 which indicates an accurate segmentation model for the given terrain detection problem.
 Conclusion: The present work motivates for a more viable hardware model that makes use of trained computer vision models and cutting-edge sensors that can be tested on different soils and terrain. The results obtained can be used to study about the different Terrains and select the most suitable model, this in turn drives for further research in the subject of Terrain Identification and Detection.

Data-limited and imbalanced bladder wall segmentation with confidence map-guided residual networks via transfer learning

Purpose: To develop an algorithm using a residual base network guided by the confidence map and transfer learning for limited dataset size and imbalanced bladder wall segmentation.Methods: The geometric transformation was made to the training data for data augmentation, and a pre-trained Resnet50 model on ImageNet was also adopted for transfer learning. Three loss functions were put into the pre-trained Resnet50 network, they are the cross-entropy loss function (CELF), the generalized Dice loss function (GDLF) and the Tversky loss function (TLF). Three models were obtained through training, and three corresponding confidence maps were output after entering a new image. By selecting the point with the maximum confidence values at the corresponding position, we merged the three images into one figure, performed threshold filtering to avoid external anomalies, and finally obtained the segmentation result.Results: The average Jaccard similarity coefficient of model training based on the CELF, GDLF and TLF is 0.9173, 0.8355, 0.8757, respectively, and the average Jaccard similarity coefficient of our algorithm can be achieved at 0.9282. In contrast, the classical 2D U-Net algorithm can only achieve 0.518. We also qualitatively give the reasons for the improvement of model performance.Conclusion: Our study demonstrates that a confidence map-assisted residual base network can accurately segment bladder walls on a limited-size data set. Compared with the segmentation results of each model alone, our method originally improves the accuracy of the segmentation results by combining confidence map guidance with threshold filtering.

Review of contemporary fluorescence correlation spectroscopy method in diverse solution studies.

Fluorescence correlation spectroscopy (FCS) is a well-known and established non-invasive method for quantification of physical parameters that preside over molecular mechanisms and dynamics. It combines maximum sensitivity and statistical confidence for the analysis of speed, size, and number of fluorescent molecules and interactions with surrounding molecules by time-averaging fluctuation analysis in a well-defined volume element. The narrow compass of this study is to acquaint the basic principle of diffusion and the FCS method in general regarding variable magnitudes and standardization adjustment. In this review, we give a theoretical introduction, examples of experimental applications, and utensils in solution systems with future perspectives.

Artificial Neural Networks and Time Series of Counts: A Class of Nonlinear INGARCH Models

Abstract Time series of counts are frequently analyzed using generalized integer-valued autoregressive models with conditional heteroskedasticity (INGARCH). These models employ response functions to map a vector of past observations and past conditional expectations to the conditional expectation of the present observation. In this paper, it is shown how INGARCH models can be combined with artificial neural network (ANN) response functions to obtain a class of nonlinear INGARCH models. The ANN framework allows for the interpretation of many existing INGARCH models as a degenerate version of a corresponding neural model. Details on maximum likelihood estimation, marginal effects and confidence intervals are given. The empirical analysis of time series of bounded and unbounded counts reveals that the neural INGARCH models are able to outperform reasonable degenerate competitor models in terms of the information loss.

A Generalized Log Gamma Approach: Theoretical Contributions and an Application to Companies’ Life Expectancy

The survival of a company has been a topic of growing interest in the scientific community. Measuring the life expectancy of Portuguese telecommunications companies using generalized log-gamma (GLG) distribution is a new research endeavor. Regarding the new theoretical contributions, original expressions for the moments and mode of the GLG distribution are presented. In this empirical study, data on the entrepreneurial fabric in the Information and Communication sector from 2004 to 2018, when some companies were born or died, were used. In addition to the GLG, three other statistical distributions with two parameters are analyzed: gamma, Weibull, and log-normal. Maximum likelihood parameters and confidence intervals for survival probabilities are estimated and compared. The Akaike information criterion is used to compare the performance of the four estimated models. The results show that GLG distribution is a promising solution to assess the resilience and longevity of a firm.

Effect of scan powder and scanning technology on measured deviations of complete-arch implant supported frameworks digitized with industrial and intraoral scanners

ObjectivesTo evaluate the suitability of intraoral scanners (IOSs) to analyze the fabrication trueness of titanium complete-arch implant-supported frameworks by comparing with an industrial-grade scanner and investigate how anti-reflective scan powder affects measured deviations. MethodsTen titanium complete-arch implant-supported frameworks were milled from a reference standard tessellation language (STL) file. An industrial-grade blue light scanner (ATOS Core 80 (AT)) and three IOSs (Primescan (PS), TRIOS T3 (T3), and TRIOS T4 (T4)) with (PS-P, T3-P, and T4-P) or without (PS, T3, and T4) anti-reflective scan powder application were used to generate test STL (TSTL) files of the frameworks. Reference STL and TSTLs were imported into a metrology-grade analysis software (Geomagic Control X) and whole surface root mean square (RMS) values were calculated. Another software (Medit Link v 2.4.4) was used to virtually isolate marginal surfaces of all STL files and marginal RMS values were calculated by using the same metrology-grade analysis software. A linear mixed effects model was used to compare the transformed deviations of the scans performed by using each IOS (with or without powder) with the deviations of those performed by using the reference AT scanner within each surface, where a Box-Cox type transformation was used for variance stability. Bonferroni corrected post-hoc tests were used to compare conditions within each IOS (α=0.05). ResultsAll IOSs had significantly higher whole surface and marginal RMS values than AT, regardless of the condition (P≤.002). However, scan powder application did not affect the whole surface and marginal RMS values in scans of tested IOSs (P≥.054). ConclusionMeasured whole surface and marginal deviations in all IOS scans performed with or without the use of scan powder were higher than those in AT scans. The application of anti-reflective scan powder did not affect the deviations in scans of tested IOSs. Clinical SignificanceEven though deviations measured in the scans of tested scanners were significantly different than those in the reference scanner, the maximum raw mean difference was 37.33 µm and the maximum raw confidence interval value of estimated differences was 47.88 µm, which can be considered clinically small taking into account the size of the frameworks tested. Therefore, tested intraoral scanners may be feasible to scan prostheses similar to or smaller than tested frameworks for fabrication trueness analysis, which may facilitate potential clinical adjustments.

Product Layout Determination System Using the Association Rules Method Using the Equivalence Class Transformation Algorithm

Competition in the business world, specifically in the sales industry, requires companies to analyze the purchases made by customers during transactions in order to find effective business strategies. In the competitive fashion industry, merchants devise marketing strategies to increase sales. One strategy that can attract consumer interest is by organizing and arranging product displays, placing them in perfect layouts that align with customers' buying habits, making it easier for them to find and purchase products. Layout arrangement significantly influences customer satisfaction and purchase intent. The algorithm used in this study is Equivalence Class Transformation (ECLAT). The data used consists of transactional data from Aufco Clothing, specifically fashion products. A total of 1041 transactions were analyzed, using variables such as order number and items sold. The data was processed using JavaScript, with a minimum support of 0.2 and a minimum confidence of 0.7, resulting in 16 rules. The rules ranged from a min. confidence of 70% to a maximum confidence of 100%, forming 6 rules with 9 combinations of items.

Target tracking algorithm based on occlusion prediction

Tracking failure caused by occlusion is one of the most challenging problems in visual tracking tasks. The current Siamese network-based object tracking algorithms use the maximum confidence strategy and center cropping mechanism, which may lead to partial or complete loss of target tracking. To address this problem, we propose an occlusion prediction-based tracking method, which uses an occlusion prediction branch to evaluate the occlusion degree of the target in four directions: up, down, left and right, and combines information fusion to predict the target position, and uses dynamic background suppression to reduce the interference of similar targets. First, we use dynamic background suppression to preprocess the input image, weaken the interference around the target, then use the target location information and regression information output by the Siamese network to obtain the candidate target position, and finally use the occlusion prediction information for information fusion to obtain the final target position. Although our method brings additional computational burden, it can still achieve real-time performance. Experiments show that our proposed method can effectively improve the robustness of Siamese tracking algorithms under different occlusion conditions. In the interference test based on OTB100, our method improves the tracking success rate by about 7% and the precision by about 10% for SiamBan and SiamGat algorithms respectively; in the interference test based on VOT2018, our method can effectively reduce the tracking loss frequency of Siamban and SiamGat algorithms, improving the EAO(Expected Average Overlap) by about 10%.

Exploring New Horizons: Advancing Data Analysis in Kidney Patient Infection Rates and UEFA Champions League Scores Using Bivariate Kavya–Manoharan Transformation Family of Distributions

In survival analyses, infections at the catheter insertion site among kidney patients using portable dialysis machines pose a significant concern. Understanding the bivariate infection recurrence process is crucial for healthcare professionals to make informed decisions regarding infection management protocols. This knowledge enables the optimization of treatment strategies, reduction in complications associated with infection recurrence and improvement of patient outcomes. By analyzing the bivariate infection recurrence process in kidney patients undergoing portable dialysis, it becomes possible to predict the probability, timing, risk factors and treatment outcomes of infection recurrences. This information aids in identifying the likelihood of future infections, recognizing high-risk patients in need of close monitoring, and guiding the selection of appropriate treatment approaches. Limited bivariate distribution functions pose challenges in jointly modeling inter-correlated time between recurrences with different univariate marginal distributions. To address this, a Copula-based methodology is presented in this study. The methodology introduces the Kavya–Manoharan transformation family as the lifetime model for experimental units. The new bivariate models accurately measure dependence, demonstrate significant properties, and include special sub-models that leverage exponential, Weibull, and Pareto distributions as baseline distributions. Point and interval estimation techniques, such as maximum likelihood and Bayesian methods, where Bayesian estimation outperforms maximum likelihood estimation, are employed, and bootstrap confidence intervals are calculated. Numerical analysis is performed using the Markov chain Monte Carlo method. The proposed methodology’s applicability is demonstrated through the analysis of two real-world data-sets. The first data-set, focusing on infection and recurrence time in kidney patients, indicates that the Farlie–Gumbel–Morgenstern bivariate Kavya–Manoharan–Weibull (FGMBKM-W) distribution is the best bivariate model to fit the kidney infection data-set. The second data-set, specifically that related to UEFA Champions League Scores, reveals that the Clayton Kavya–Manoharan–Weibull (CBKM-W) distribution is the most suitable bivariate model for fitting the UEFA Champions League Scores. This analysis involves examining the time elapsed since the first goal kicks and the home team’s initial goal.

The Discrete Exponentiated-Chen Model and Its Applications

A novel discrete exponentiated Chen (DEC) distribution, which is a subset of the continuous exponentiated Chen distribution, is proposed. The offered model is more adaptable to analyzing a wide range of data than traditional and recently published models. Several important statistical and reliability characteristics of the DEC model are introduced. In the presence of Type-II censored data, the maximum likelihood and asymptotic confidence interval estimators of the model parameters are acquired. Two various bootstrapping estimators of the DEC parameters are also obtained. To examine the efficacy of the adopted methods, several simulations are implemented. To further clarify the offered model in the life scenario, the two applications, based on the number of vehicle fatalities in South Carolina in 2012 and the final exam marks in 2004 at the Indian Institute of Technology at Kanpur, are analyzed. The analysis findings showed that the DEC model is the most effective model for fitting the supplied data sets compared to eleven well-known models in literature, including: Poisson, geometric, negative binomial, discrete-Weibull, discrete Burr Type XII, discrete generalized exponential, discrete-gamma, discrete Burr Hatke, discrete Nadarajah-Haghighi, discrete modified-Weibull, and exponentiated discrete-Weibull models. Ultimately, the new model is recommended to be applied in many fields of real practice.

In-silico Elucidation of the Role of ABC-Transporter Genes Expression Regulation by OncomiRs (miR-21, miR-15, and miR-let-7) in Drug Efflux and Chemoresistance in Breast Cancer

Breast cancer is the most common and aggressive malignancy in females with a high prevalence rate of 77.9 million worldwide. Chemotherapy and tyrosine kinase inhibitors have been used to treat invasive and malignant tumors; however, invasive tumors have showed resistance to conventional therapies. ABC transporters play a crucial role in breast cancer due to their chemo-resistance and drug efflux abilities. Additionally, chemo-resistant roles of ABC transporters have been reported in several cancers such as cervical cancer, colon cancer, esophageal squamous cell carcinoma, glioma and HCC. The goal of this study was to identify the tumor suppressor role of miR-21, miR-15 and miR-let-7 to chemo-resistant genes majorly ABCA1, ABCB1 and ABCC1 in breast cancer. TargetScan, miRWalk, and miRDB were employed to predict microRNA-mRNA interactions. MC-Sym and RNAComposer were utilized for the tertiary structure prediction of shortlisted miRNAs and mRNAs. For molecular docking and visualization, HDOCK and PyMOL were employed. The present study identified 10, 7 and 13 interactions between microRNAs (miR-21, miR-15, and miR-let-7) and oncogenes (ABCA1, ABCB1, ABCC1) through miRWalk, miRDB and TargetScan respectively. RNA22 predicted the binding sites of microRNAs such as 22 miR-21, 11 miR-15 and 58 miR-let-7 on ABCA1, ABCB1 and ABCC1, respectively. Out of multiple docked complexes, the top 3 were shortlisted for visualization based on maximum confidence score and least binding affinity. The present study identifies the interactions of two novel (miR-15a-5p and let-7c-5p) microRNAs with ABCA1, ABCB1 and ABCC1 regions due to their maximum interactions. The findings of this research may help in developing miRNA drugs that could target ABC transporters specifically ABCA1, ABCB1 and ABCC1 to inhibit increased drug efflux and chemoresistance in breast cancer.

NKX3.1 immunohistochemistry and methylome profiling in mesenchymal chondrosarcoma: additional diagnostic value for a well-defined disease?

Mesenchymal chondrosarcoma (MCS) is a rare and highly aggressive tumour of soft tissue and bone that is defined by an underlying and highly specific fusion transcript involving HEY1 and NCOA2. Histologically, the tumours show a biphasic appearance consisting of an undifferentiated blue and round cell component as well as islands of highly differentiated cartilage. Particularly in core needle biopsies, the chondromatous component can be missed and the non-specific morphology and immunophenotype of the round cell component can cause diagnostic challenges. We applied NKX3.1 immunohistochemistry which was recently reported as a highly specific marker as well as methylome and copy number profiling to a set of 45 well characterised MCS cases to evaluate their potential diagnostic value. Methylome profiling revealed a highly distinct cluster for MCS. Notably, the findings were reproducible also when analysing the round cell and cartilaginous component separately. Furthermore, four outliers were identified by methylome profiling for which the diagnosis had to be revised. NKX3.1 immunohistochemistry showed positivity in 36% of tumours, the majority of which was rather focal and weak. Taken together, NKX3.1 expression showed a low sensitivity but a high specificity in our analysis. Methylome profiling on the other hand represents a sensitive, specific and reliable tool to support the diagnosis of MCS, particularly if only the round cell component is obtained in a biopsy and the diagnosis is not suspected. Furthermore, it can aid in confirming the diagnosis in case RNA sequencing for the HEY1::NCOA2 fusion transcript is not available.

CONFORMITY ASSESSMENT IN INFRARED EAR THERMOMETER CALIBRATION

Body temperature is a basic physiological measurement used primarily for observation and diagnosis during surgery, pandemic diseases, intensive care, recovery or treatment. Various thermometers are clinically used for these measurements and can be divided into two categories, contact or non-contact. To have maximum confidence in the clinical performance of the temperature measuring instrument, it is important to calibrate the instrument traceable to the 1990 International Temperature Scale (ITS-90). The absence of traceable calibrations accredited to ISO/IEC 17025 can lead to insecurity in temperature measurement and in some cases can have a detrimental effect on patient care.Infrared or radiation thermometers are preferred in many different sectors from health to food, from automotive to agriculture, from iron and steel industry to chemistry due to their fast measuring capabilities, reasonable prices, wide measuring ranges and ease of use. In this study, ear thermometer calibration and measurement uncertainty calculations are investigated by using the traceability method via contact thermometer. Considering the maximum permissible error value of the device, conformity assessment is made according to the measurement result.