Background Subtraction Research Articles

Novel Deep Learning Domain Adaptation Approach for Object Detection Using Semi-Self Building Dataset and Modified YOLOv4

Moving object detection is a vital research area that plays an essential role in intelligent transportation systems (ITSs) and various applications in computer vision. Recently, researchers have utilized convolutional neural networks (CNNs) to develop new techniques in object detection and recognition. However, with the increasing number of machine learning strategies used for object detection, there has been a growing need for large datasets with accurate ground truth used for the training, usually demanding their manual labeling. Moreover, most of these deep strategies are supervised and only applicable for specific scenes with large computational resources needed. Alternatively, other object detection techniques such as classical background subtraction need low computational resources and can be used with general scenes. In this paper, we propose a new a reliable semi-automatic method that combines a modified version of the detection-based CNN You Only Look Once V4 (YOLOv4) technique and background subtraction technique to perform an unsupervised object detection for surveillance videos. In this proposed strategy, background subtraction-based low-rank decomposition is applied firstly to extract the moving objects. Then, a clustering method is adopted to refine the background subtraction (BS) result. Finally, the refined results are used to fine-tune the modified YOLO v4 before using it in the detection and classification of objects. The main contribution of this work is a new detection framework that overcomes manual labeling and creates an automatic labeler that can replace manual labeling using motion information to supply labeled training data (background and foreground) directly from the detection video. Extensive experiments using real-world object monitoring benchmarks indicate that the suggested framework obtains a considerable increase in mAP compared to state-of-the-art results on both the CDnet 2014 and UA-DETRAC datasets.

A preliminary investigation into the feasibility of laser ablation U-Pb isotope ratio measurement via all-faraday cup detection with 1011- and 1013-Ω amplifiers on the Neoma multicollector-inductively coupled plasma-mass spectrometer.

Signal detection for uranium-lead (U-Pb) dating of zircon is typically performed via ion counters. Here, we develop a preliminary understanding of the strengths and limitations of faraday-cup-based detection. A suite of zircon reference materials and the NIST-610 glass were sampled using laser ablation followed by U-Pb isotope ratio measurement on a Neoma multicollector-inductively coupled plasma-mass spectrometer. We were able to produce geologically accurate 207Pb/206Pb, 206Pb/238U, and 207Pb/235U ratios for the NIST-610 glass and the zircon standards, with ages ranging from ~2.5Ga to ~337 Ma (TanBrown A, Oracle, 91550, Mud Tank, Temora, and Plešovice). Two of the younger zircon standards examined (94-35, ~55.6Ma, and Fish Canyon, 28.6Ma) yielded accurate 206Pb/238U but not 207Pb/235U or 207Pb/206Pb ratios, whereas the youngest zircon standard (Penglai, ~4.4Ma) failed for all three ratios of interest. The accuracy and precision of the all-faraday method are directly tied to signal intensity, with reliable data capable of being produced even when both isotopes in a ratio have signals below ~0.001 V (equivalent to ~62 500 cps on an ion counter). The all-faraday cup multicollection method provides sufficient sensitivity to obtain geologically meaningful U-Pb data, with possible advantages being that laser pit depth-dependent changes in the observed interelemental fractionation behavior may be easier to correct using a static collector configuration compared to when the ion beam is swept across a single detector while also removing the need for an interdetector-type calibration. Further work is needed to refine the all-faraday cup method (e.g., application of background subtraction and common Pb corrections, outlier removal, and interelement as well as down-hole fractionation corrections), but our initial results demonstrate that the faraday detector method has sufficient sensitivity to warrant further study.

Reassessment of the Shirley and model functions within X-ray photoelectron spectroscopy analysis utilizing comprehensive search methodologies of Bayesian inference

The analysis of X-ray photoelectron spectra often faces challenges due to the lack of standardization in modeling approaches, background subtraction methods, and computational algorithms within the field of computer science. The interpretation of XPS data significantly relies on the unique expertise and judgment of individual researchers. Therefore, the objective of this study is to highlight the difficulties associated with analytical methods that depend heavily on the discretion of individual scientists, to elucidate the prevailing models and background subtraction techniques, and to suggest improvements to these methodologies. This endeavor aims to enhance the reliability and reproducibility of XPS analysis, thereby contributing to the advancement of research in this area. By utilizing the information criterion as part of a thorough search methodology in Bayesian inference, we show that our sophisticated analytical techniques significantly outperform others in the analysis of actual X-ray photoelectron spectroscopy (XPS) spectra. This improvement is evidenced through enhanced accuracy and reliability in spectral interpretation, underscoring the efficacy of our methods in practical applications of XPS.

A novel technique for leaf disease classification using Legion Kernels with parallel support vector machine (LK-PSVM) and fuzzy C means image segmentation

Detection of plant disease and classificationare being investigated in many parts of the worldto save precious medical plants from becoming extinct.Major problem in this task, include the lack of advanced and technology driven solution. Manual identification is often time-consuming and prone to inaccuracies. Therefore, there is an urgent need for an automated and efficient method that can accurately identify and classify plant diseases. This article focuses on detecting the disease through classificationthrough a new technique using leaf images for automatic classification. This paper proposes a novel segmentation technique using Fuzzy C means and Particle Swarm Optimization for effective segmentation of leaf images and feature extraction that can help in classification of disease.The approach emphasizes on the integration of techniques such as image processing, segmentation and feature extraction and finally the classification, which offers a comprehensive solution for the disease detection. The work leverages on the advantages of Legion Kernels and Parallal support vector Machine (LK-PSVM) clubbed with fuzzy C means Image segmentation to offer a framework that can handle diverse leaf images and which can effectively differentiate the type of the disease.The proposed method LK-PSVM combined with Fuzzy C means presents a novel approach that is significantly deviated from the conventional methods of leaf disease classification.The proposed wok brings an integrated framework which can synergistically combine the Legion Kernels with the PSVM technique coupled with Fuzzy C Means Image segmentation which can handle the issue of overlapped data sets and support vector machines are used to handle the situation where the number of dimensions are more than the number of samples, which is more probable in the classification problem under consideration.By integrating these components, the proposed method achieves more accuracy and robustness when compared to the existing methods in the literature. The segmentation is carried out using PSO after pre-processing of images. The Gaussian functions are used to eliminate the background subtraction. Different features of the images are then computed. A total of 55,400 images were used for the experiment consisting of various plants’ leaves spreading across 38 labels. A classifier is then proposed using Machine learning methods for the detection of disease in apple fruit leaves. The experiments prove that the proposed method have high degree of classification accuracy when compared to existing methods. The proposed method not only cater to the need in terms of accuracy but also making it scalable for different types of leaves.

Detection of gunshot residue by flash-pulse and long-pulse infrared thermography

Detection of gunshot residues (GSR) in a bullet hole area is one of the forensic investigations aiding in the reconstruction of crime scenes. Traditionally, chromogenic methods based on chemical exposure or microscopic/spectroscopic methods are used for this purpose. In this study, we explore the applicability of active excitation infrared thermography methods for GSR detection in the bullet hole area on fabric samples. A standard 9 mm full metal jacket ammunition with a nickel-plated shell and natural cotton fabric samples were used for experiments in this study. The applicability of active thermography methods based on two different light/heat excitation sources to detect the GSR was investigated. Flash-pulse and long-pulse thermography were compared through an experimental investigation. We evaluated the effectiveness of various thermographic data processing methods, including background subtraction, temperature derivative analysis, Fourier transform phase analysis, principal component analysis, and higher-order statistics for GSR evaluation. Our findings demonstrate that flash-pulse thermography and kurtosis analysis yield the highest contrast-to-noise ratio (CNR) and produce sharp, clear images of GSR, making it the optimal method for thermographic GSR detection. Our study indicates that even though the GSR particles are tiny, they can produce sufficient contrast to be detected by the thermographic methods if appropriate experimental and post-processing procedures are used. Thus, these methods could complement GSR detection as they are non-destructive and offer rapid inspection.

#1324 The antiproteinuric effect of Gliflozins is associated with the mitochondrial density of endothelial cells in diabetic patients

Abstract Background and Aims The sodium-glucose co-transporter type 2 inhibitors (SGLT2i) such as dapagliflozin and empagliflozin, are a new class of nephro- and cardio-protective drugs in diabetic and non-diabetic patients. SGLT2 is greatly expressed in the proximal tubule cells, and therefore the protective effects of gliflozins are thought to be mediated by a hemodynamic mechanism on glomerular hyperfiltration. However, they could be acting also in other cell types, particularly in diabetic patients, because hyperglycemia induces the expression of SGLT2 in endothelial cells, mesangial cells, cardiomyocytes, etc. Because SGLT2 changes the intracellular concentration of sodium and glucose, gliflozins may indirectly change mitochondrial swelling and activity. In this study, we have tested the hypothesis that gliflozins act on mitochondria in patients suffering from diabetes. Method We have analyzed kidney biopsies from diabetic patients in treatment with a gliflozin (n = 10) or without gliflozin treatment (n = 10) stained with fuchsin red, which gives a color intensity proportional to mitochondrial content. Clinical variables (albumin-creatinine ratio (ACR), blood pressure, eGFR, glycated albumin) have been measured before and after 4 months of treatment. Images have been acquired using a Zeiss Axioscope with 100x objective and analyzed using ImageJ software to quantify staining intensity (expressed as optical density) after background subtraction. Two different cell types have been considered: proximal tubule cells and endothelial cells in glomeruli. Only non-sclerotic glomeruli have been considered. Results The systolic blood pressure was linearly correlated to the intensity of staining of mitochondria in the proximal tubule cells in all patients irrespective of gliflozin treatment. In non-treated diabetic patients, the mitochondrial content of endothelial cells was not related to the extent of ACR reduction over time. In diabetic patients treated with gliflozins, the extent of ACR reduction after treatment was correlated with the intensity of mitochondrial staining in glomerular endothelial cells. This correlation was not present in proximal tubule cells. Gliflozin treatment may induce a change in the mitochondrial content of endothelial cells, particularly in patients responding with a reduction in proteinuria. Conclusion Proteinuria in diabetic patients depends on damage in both the endothelial and podocyte compartments. Gliflozins may improve proteinuria by a hemodynamic mechanism; however, our data suggest an effect of gliflozins specific on mitochondria of the endothelial cells in the glomeruli, which is proportional to their anti-proteinuric effect. Though we do not analyze the mechanism of gliflozins on mitochondria, it is possible to speculate that the changes of intracellular sodium-induced by gliflozins may induce an altered mitochondrial activity.

A Dynamic Detection Method for Railway Slope Falling Rocks Based on the Gaussian Mixture Model Segmentation Algorithm

Rockfall intrusion detection is crucial for the safety management of railway operations, and video detection methods help reduce deployment costs and improve detection efficiency. Mainstream neural network-based video detection methods have rapidly evolved in recent years but struggle to adapt to complex scenarios such as existing railway slope constructions due to weak generalization ability, low accuracy, and limited information acquisition. Therefore, this paper introduces a dynamic neural network detection model and establishes a dataset for rockfall intrusions in existing railway slope scenarios. The model initially relies on the YOLOv5 neural network, adopting an activation function suitable for the target scenario and addressing overfitting to achieve precise target recognition. Based on the neural network, the model dynamically detects rolling rockfalls by integrating a background subtraction algorithm based on the Gaussian Mixture Model and captures target dimensions using monocular vision technology, thus broadening the dimensions of detection information. Trials conducted on a railway in Shandong, China, demonstrate that the model accurately identifies moving rockfalls along the railway slopes and acquires the dimensions of moving rockfalls, successfully filtering out low-risk targets in the scene.

Anomaly Detection for Video Surveillance

Anomaly detection in video surveillance is critical for identifying abnormal behaviours or events that deviate from typical patterns. This abstract explores various techniques used to detect anomalies in surveillance videos, including traditional methods like motion detection and background subtraction, as well as advanced approaches such as deep learning and anomaly scoring algorithms. By leveraging these techniques, surveillance systems can automatically detect suspicious activities such as trespassing, theft, or violence, enabling timely intervention and improved security measures. Highlighting the challenges in anomaly detection, such as dealing with complex scenes, occlusions and discusses ongoing research efforts to enhance the accuracy and efficiency of anomaly detection systems in real-world surveillance scenarios. Key Words: anomaly, suspicious, pattern, motion, detection, security measure

The Hyper Suprime-Cam extended Point Spread Functions and applications

Abstract We present extended point spread function (PSF) models for the Hyper Suprime-Cam Subaru Strategic Program Public Data Release 3 (HSC-SSP PDR3) in all g,r,i,Z and Y-bands. Due to its 8.2m primary mirror and long exposure periods, HSC combines deep images with wide-field coverage. Both properties make HSC one of the most suitable observing facilities for low surface brightness (LSB) studies, which are particularly sensitive to the PSF. By applying a median stacking technique of point-like sources with different brightness, we show how to construct the HSC-SSP PDR3 PSF models to an extent of R ∼ 5.6 arcmin. These models are appropriate for the HSC-PDR3 intermediate-state data which do not have applied the final aggressive background subtraction. The intermediate-state data is especially stored for users interested in large extended objects, where our new PSFs provide them with a crucial tool to characterise LSB properties at large angles. We demonstrate that our HSC PSFs behave reasonably in two scenarios. In the first one, we generate 2-D models of a bright star, showing no evidence of residual structures across the five bands. In the second scenario, we recreate the PSF-scattered light on mock images with special consideration of the effect of this additional flux on LSB measurements. We find that, despite the well-behaved nature of the HSC-PDR3 PSFs, there is a non-negligible impact on the faint light present in the mock images. This impact could lead to incorrect LSB measurements if a proper star subtraction is not applied.

Cannabinoid Modulation of Monoamine Levels in Mouse Brain: Unveiling Neurochemical Dynamics through an Innovative High-performance Liquid Chromatography-Fluorescence Detection Bioanalysis

Homeostasis of monoaminergic pathways is compromised in aging and neurodegenerative/neurological diseases such as Alzheimer’s disease and depression. On the other hand, their modulation has also been linked to the mechanism of action of several drugs. Therefore, monitoring the levels of noradrenaline (NA), adrenaline (AD), dopamine (DA), and serotonin (5-HT), as well as those of tryptophan (TRYP), the precursor of 5-HT, and DA metabolites, namely 3,4-dihydroxyphenylacetic acid (DOPAC) and homovanillic acid (HVA), is fundamental for assessing disease severity and progression. This work aimed to develop and validate the first High-Performance Liquid Chromatography (HPLC) coupled with fluorescence detector (FLD) method that simultaneously and accurately quantifies NA, AD, DA, DOPAC, HVA, 5-HT and TRYP in mouse brain and prefrontal cortex (PFC) matrices. Previous sample preparation by protein precipitation was required to extract the compounds. Calibration curves were plotted using the background subtraction approach to reduce the interference of the endogenous analytes. Intra and inter-day accuracy and precision were within the ranges defined by ICH (The International Council for Harmonisation) guideline for bioanalytical method validation. Following validation, the impact of cannabidiol (CBD), cannabigerol (CBG), and cannabidivarin (CBDV) was explored in mice brains post-administration, revealing significant alterations in specific neurotransmitter levels upon cannabinoid exposure and shedding light on the complex modulation of neurochemical dynamics by cannabinoids. This research highlights the fit-for-purpose of the HPLC-FLD method and provides insights into potential mechanisms underlying phytocannabinoid actions in the central nervous system (CNS).

BSEO-Semiautomatic Method for Determination of Oil Recovery with Nanofluids in Microfluidic Devices.

Microfluidic models have become essential instruments for studying enhanced oil recovery techniques through fluid and chemical injection into micromodels to observe interactions with pore structures and resident fluids. The widespread use of cost-effective lab-on-a-chip devices, known for efficient data extraction and minimal reagent usage, has driven demand for efficient data management methods crucial for high-performance data and image analyses. This article introduces a semiautomatic method for calculating oil recovery in polymeric nanofluid flooding experiments based on the background subtraction (BSEO). It employs the background subtraction technique, generating a foreground binary mask to detect injected fluids represented as pixel areas. The pixel difference is then compared to a threshold value to determine whether the given pixel is foreground or background. Moreover, the proposed method compares its performance with two other representative methods: the ground truth (manual segmentation) and Fiji-ImageJ software. The experiments yielded promising results. Low values of mean-squared error (MSE), mean absolute error (MAE), and root-mean-squared error (RMSE) indicate minimal prediction errors, while a substantial coefficient of determination (R2) of 98% highlights the strong correlation between the method's predictions and the observed outcomes. In conclusion, the presented method emphasizes the viability of BSEO as a robust alternative, offering the advantages of reduced computational resource usage and faster processing times.

Hybrid Semiconductor Wafer Inspection Framework via Autonomous Data Annotation

Abstract In smart manufacturing, semiconductors play an indispensable role in collecting, processing, and analyzing data, ultimately enabling more agile and productive operations. Given the foundational importance of wafers, the purity of a wafer is essential to maintain the integrity of the overall semiconductor fabrication. This study proposes a novel automated visual inspection (AVI) framework for scrutinizing semiconductor wafers from scratch, capable of identifying defective wafers and pinpointing the location of defects through autonomous data annotation. Initially, this proposed methodology leveraged a texture analysis method known as gray-level co-occurrence matrix (GLCM) that categorized wafer images—captured via a stroboscopic imaging system—into distinct scenarios for high- and low-resolution wafer images. GLCM approaches further allowed for a complete separation of low-resolution wafer images into defective and normal wafer images, as well as the extraction of defect images from defective low-resolution wafer images, which were used for training a convolutional neural network (CNN) model. Consequently, the CNN model excelled in localizing defects on defective low-resolution wafer images, achieving an F1 score—the harmonic mean of precision and recall metrics—exceeding 90.1%. In high-resolution wafer images, a background subtraction technique represented defects as clusters of white points. The quantity of these white points determined the defectiveness and pinpointed locations of defects on high-resolution wafer images. Lastly, the CNN implementation further enhanced performance, robustness, and consistency irrespective of variations in the ratio of white point clusters. This technique demonstrated accuracy in localizing defects on high-resolution wafer images, yielding an F1 score greater than 99.3%.

SYSTEM DESIGN FOR CALCUALTING THE NUMBER AND DENSITY OF MOTORCYCLES IN PARKING AREA BASED ON BACKGROUND SUBTRACTION METHOD

There are still issues with a number of students parking their vehicles improperly in the parking spaces provided. This causes suboptimal parking capacity due to a lack of information about parking capacity. In recent years, at a certain library, vehicle detection has been implemented using a Gaussian mixture model algorithm using a Raspberry Pi. However, this library does not provide information about the density status of the parking area. Therefore, an information system was created to determine the level of density of the parking area based on the ratio of vehicles entering and exiting compared to the maximum capacity of the parking area. The system uses the Gaussian mixture model algorithm with the machine learning method of background subtraction MOG2, which can calculate the number of vehicles based on the difference between objects and the background of objects, using test data in the form of videos recorded using a camera positioned horizontally to the entrance and exit lanes of the parking area. This research resulted in an accuracy of 89.7% for Video1TA, precision of 93.2%, a crowded parking area density level, and a value of 129.12. Video2TA had a value of 101.08, precision of 100%, and accuracy of 90%, while Video3TA had an accuracy of 35%, precision of 56.7%, and a value of 49.48. The density levels of videos 2 and 3 are the same, indicating that the parking area is still empty. The test results show that the value can affect the system in detecting an object.

Bridging the Communication Gap - Sign Language Communication System

Now, imagine a world where words are not spoken, and emotions are not expressed. For the Deaf community, this is not a mere thought experiment but a daily reality. Indeed, they are forced to communicate primarily through sign language e, therefore, regularly feel alienated in a society where words are the primary tool for translating thoughts and feelings. Under such circumstances, most Deaf and hard-of-hearing people are struggling to communicate the most effectively. They usually opt for interpreters or their own signaling, but these alternatives are not the most efficient or effective choices. While signaling is the most intuitive and meaningful alternative, its grammar and semantic variations make it impossible to comprehend for those people who are not within the culture. Therefore, we have also developed a software prototype that could interpret sign language automatically. This paper presents a new method to recognize Indian sign language alphabets A-Z and digits (0-9) in a real- time video feed by employing the Bag of Visual Words model (BOVW). Furthermore, it not only predicts the labels of the signs but also gives the output as text and speech. The segmentation process here comprehends skin color detection and background subtraction. Speeded Up Robust FeaturesSurf features have been extracted again from the images, and histograms are built to map the signs to text labels. Furthermore, we have utilized algorithms such as Support Vector Machine (SVM) and Convolutional Neural Networks (CNN) for binary classification. Finally, we have established an interactive Graphical User Interface (GUI to make this process more user-friendly

Testosterone enhances adipose tissue-blood pressure axis in obese mice exposed to early life stress

Previously, we have shown that male and female mice exposed to maternal separation and early weaning (MSEW), a model of early life stress, display exacerbated obesity-induced hypertension in response to chronic high fat diet. However, only male mice showed higher sympathetic tone at baseline, along with increased neuronal activation in PVN and pressor response to the acute stimulation of TRVP1 channels with capsaicin, a mechanism known as adipose afferent reflex (AAR). The aim of this study was to investigate whether obese male MSEW mice show 1) increased intracellular calcium in response to capsaicin in neuronal cultures of dorsal ganglia root (DRG), and 2) a role for sex hormones in AAR reactivity in vivo. Six-months-old control and MSEW fed a high fat diet (HFD, 60% kcal fat) were injected with pAAV-CAG-tdTomato (59462-PHP.S, 1012 viral particles); 8 sites; 4 ul/site; bilateral). After 6 weeks, mice were decapitated and DRGs (L5-T12) were collected and placed in ice-cold HBSS solution, dissociated and mixed with Collagenase P (1 mg/ml in HBSS, 15 min, 37 oC). After 0.25% Trypsin-EDTA incubation, neurons were mechanically dissociated, collected by centrifugation and resuspended in DMEM-10% FBS, plated onto the Matrigel Basement Membrane Matrix covered coverslips and maintained at 37 oC for 15-16 hours. Neuronal cultures were incubated in Ca2+ imaging solution with 2 μM Fura-2 AM, 30 min in dark conditions. After de-estirification, intracellular Ca2+ transients were visualized by exciting Fura-2 at 340 ± 20 nm and 380 ± 20 nm using a high-speed filter changer to obtain a ratiometric value independent of indicator concentration. Emitted light was passed through a dichroic filter (400-nm high pass) and an emitter filter (520 ± 30 nm) and was digitized onto the sensor of an EMCCD camera. Background subtraction and normalization to baseline were used to quantify changes in fluorescence. Despite similar resting uncalibrated ratios of calcium levels between groups (~0.6 ratio units, n=3), sensory neurons from MSEW mice tended to display larger responses to capsaicin (50 uM), showing elevated calcium kinetics vs. controls (0.56±0.09 vs. 0.31±0.12 delta ratio, respectively, p=0.063) as well as longer transients, perhaps indicative of reduced clearance or buffering. A separate set of age-matched mice were gonadectomized. After 4 weeks, anesthetized mice were fitted with carotid artery catheter for direct blood pressure measurement in response to the infusion of perigonadal fat with capsaicin (1.5 pmol/μL during 2 minutes in 4 sites, bilaterally). Capsaicin-induced pressor response was exacerbated in intact MSEW male mice compared to intact controls (9.0±0.2 vs. 1.3±0.2 delta mmHg, respectively, n=6, p<0.05). Orchidectomy attenuated delta blood pressure in MSEW mice to 7.0±0.2 mmHg (p<0.05 vs. intact MSEW), while enhanced delta blood pressure in control mice to 7.5±0.2 mmHg (p<0.05 vs. intact control). MSEW did not influence the pressor response to perigonadal fat stimulation with capsaicin either in intact or ovariectomized obese female mice (n=4-6 per group, p>0.05), suggesting that estradiol may not to play a protective role in obese females AAR reactivity. This data indicates that testosterone could buffer the AAR reactivity of obese control mice, while MSEW exerts an opposite effect. Thus, investigating testosterone signaling on calcium handling mechanisms (i.e., buffers, pumps, ER-mediated calcium release) in sensory neurons innervating visceral fat will provide insights on the sex-specific contribution of afferent sympathoexitatory signals to obesity-hypertension in male MSEW mice. This study was supported by NIH R01 HL135158, and SALT pilot - University of Kentucky to ASL. This is the full abstract presented at the American Physiology Summit 2024 meeting and is only available in HTML format. There are no additional versions or additional content available for this abstract. Physiology was not involved in the peer review process.

Multidimensional Data Processing With Bayesian Inference via Structural Block Decomposition.

How to handle large multidimensional datasets, such as hyperspectral images and video information, efficiently and effectively plays a critical role in big-data processing. The characteristics of low-rank tensor decomposition in recent years demonstrate the essentials in describing the tensor rank, which often leads to promising approaches. However, most current tensor decomposition models consider the rank-1 component simply to be the vector outer product, which may not fully capture the correlated spatial information effectively for large-scale and high-order multidimensional datasets. In this article, we develop a new novel tensor decomposition model by extending it to the matrix outer product or called Bhattacharya-Mesner product, to form an effective dataset decomposition. The fundamental idea is to decompose tensors structurally in a compact manner as much as possible while retaining data spatial characteristics in a tractable way. By incorporating the framework of the Bayesian inference, a new tensor decomposition model on the subtle matrix unfolding outer product is established for both tensor completion and robust principal component analysis problems, including hyperspectral image completion and denoising, traffic data imputation, and video background subtraction. Numerical experiments on real-world datasets demonstrate the highly desirable effectiveness of the proposed approach.

QCD corrections to the Darwin coefficient in inclusive semileptonic B→Xuℓν¯ℓ decays

In this paper, we compute αs corrections to the matching coefficients of the dimension-six operators in the heavy quark expansion of the inclusive semileptonic heavy hadron decay rate and leptonic invariant mass spectrum with a massless quark and both a massive or massless lepton in the final state, analytically. The obtained results can be applied to the inclusive semileptonic B→Xuℓν¯ℓ (ℓ=e, μ, τ) and D→Xℓν¯ℓ (ℓ=e, μ) decays. The main application of our results is the background subtraction of the B→Xuℓν¯ℓ decay in the measurement of the B→Xcℓν¯ℓ decay, which is important for the precise extraction of Vcb and R(D(*)). They also play a role in the computation of lifetimes of heavy hadrons. Published by the American Physical Society 2024

Vehicle counting application utilizing background subtraction method with large-scale camera data

In modern society, people are increasingly using cameras at home, in shops, and on the streets. Traffic systems have also invested in building more surveillance camera systems. The data collected by cameras contains valuable information for traffic regulation and recording traffic violations. The challenge is how to effectively use this data. In this article, we will discuss the use of real-time data from surveillance cameras on some roads in Da Nang City for vehicle counting using background subtraction methods. Additionally, we also tested the detection of red light violations to contribute to the development of a smart traffic system. So, the use of background subtraction in analyzing real-time data from surveillance cameras can greatly improve traffic management

Research on health monitoring and damage recognition algorithm of building structures based on image processing

With the continuous deepening of the urbanization process and the progress of science and technology, people transform nature and develop nature on a larger and larger scale, among which the most iconic transformation is a variety of building structures built by people. And with the passage of time, the building structure in the perennial wind and sun, there will be signs of “illness”, if not timely treatment, it will have a huge impact on the stability and safety of the building structure. Based on this, in this paper, according to the characteristics of crack identification on the surface of concrete structure, background subtraction algorithm is selected for image noise reduction processing. Through three steps of digital image noise reduction, crack extraction and crack parameter identification, the quantitative recognition of cracks is completed and a complete system of crack parameter identification is formed. The experimental results show that the machine learning model of building structure health monitoring and damage recognition algorithm proposed in this paper has excellent statistical performance, and the relative error accuracy of recognition can be controlled within 10%.

Skin Cancer Image Segmentation Based on Midpoint Analysis Approach.

Skin cancer affects people of all ages and is a common disease. The death toll from skin cancer rises with a late diagnosis. An automated mechanism for early-stage skin cancer detection is required to diminish the mortality rate. Visual examination with scanning or imaging screening is a common mechanism for detecting this disease, but due to its similarity to other diseases, this mechanism shows the least accuracy. This article introduces an innovative segmentation mechanism that operates on the ISIC dataset to divide skin images into critical and non-critical sections. The main objective of the research is to segment lesions from dermoscopic skin images. The suggested framework is completed in two steps. The first step is to pre-process the image; for this, we have applied a bottom hat filter for hair removal and image enhancement by applying DCT and color coefficient. In the next phase, a background subtraction method with midpoint analysis is applied for segmentation to extract the region of interest and achieves an accuracy of 95.30%. The ground truth for the validation of segmentation is accomplished by comparing the segmented images with validation data provided with the ISIC dataset.