Background Subtraction Approach Research Articles

Intelligent Technique for Moving Object Detection from Problematic Video Captured through Camera Sensor

Aim: The significant aim of the proposed work is to develop an adaptive method to compute the threshold during run-time and update it adaptively for each pixel in the testing phase. It classifies motion-oriented pixels from the scene for moving objects using background subtraction and enhances using post-processing. Background: According to the huge demand for surveillance system, society is looking towards an intelligent video surveillance system that detect and track moving objects from video captured through a surveillance camera. So, it is very crucial and highly recommended throughout the globe in numerous domains such as video-based surveillance, healthcare, transportation, and many more. Practically, this research area faces lots of challenging issues such as illumination variation, cluttered background, camouflage, etc. So, this paper has developed an adaptive background subtraction method to handle such challenging problems. Objective: To focus and study the problematic video data captured through the camera sensor. To handle challenging issues available in real-time video scenes. To develop a background subtraction method and update the background model adaptively for moving object detection. Method: The proposed method has been accomplished using the following sections: Background model construction Automatic generation of threshold Background subtraction Maintenance of background model Results: The qualitative analysis of the proposed work is experimented with publicly available datasets and compared with considered state-of-the-art methods. In this work, library sequence (thermal data) of CDNET and other color video frame sequences Foreground aperture, Waving Tree and Camouflage are considered from Microsoft’s Wallflower. The quantitative values depicted in Table- 1. This work demonstrate the better performance of the proposed method as compared to state-ofthe- art methods. It also generates better outcomes and handles the problem of a dynamic environment and illumination variation. Conclusion: Currently, the world is demanding computer vision-based security and surveillancebased applications for society. This work has provided a method for the detection of moving information using an adaptive method of background subtraction approach for moving object detection in video scenes. The performance evaluation depicts better average results as compared to considered peer methods.

Latent variable modeling of gamma-ray background in repeated measurements

We propose a novel approach for background subtraction in repeated gamma-ray spectrometric measurements. This entirely data-driven method eliminates the need for Monte Carlo detector simulation. To accomplish this, we utilized the framework of Latent Variable Modeling, incorporating various matrix factorization techniques and artificial neural networks. Subsequently, we applied this method to estimate radionuclide activity through spectrum unmixing. Significant improvements in sensitivity, surpassing traditional methods, were observed for the test case scenario of aerosol filter measurements.

Iris

We propose a novel Visible Light Positioning (VLP) method, called Iris, that leverages light spectral information (LSI) to localize individuals in a completely passive manner. This means that the user does not need to carry any device, and the existing lighting infrastructure remains unchanged. Our method uses a background subtraction approach to accurately detect changes in ambient LSI caused by human movement. Furthermore, we design a Convolutional Neural Network (CNN) capable of learning and predicting user locations from the LSI change data. To validate our approach, we implemented a prototype of Iris using a commercial-off-the-shelf light spectral sensor and conducted experiments in two typical real-world indoor environments: a 25 m2 one-bedroom apartment and a 13.3m × 8.4m office space. Our results demonstrate that Iris performs effectively in both artificial lighting at night and in highly dynamic natural lighting conditions during the day. Moreover, Iris outperforms the state-of-the-art passive VLP techniques significantly in terms of localization accuracy and the required density of light sensors. To reduce the overhead associated with multi-channel spectral sensing, we develop and validate an algorithm that can minimize the required number of spectral channels for a given environment. Finally, we propose a conditional Generative Adversarial Network (cGAN) that can artificially generate LSI and reduce data collection effort by 50% without sacrificing localization accuracy.

Excitation functions of neutron-induced threshold reactions in Au, Bi, Ta measured using 30–94 MeV quasi mono-energetic neutron sources

Activation measurements of neutron-induced threshold reactions were studied employing 7Li(p,n) quasi mono-energetic neutron sources in the neutron energy range of 30–94 MeV, where scarce experimental and evaluated data exist. A novel iterative neutron background subtraction approach is presented for the first time and its successful application is demonstrated on the analysis of the excitation functions of (n,xn) reactions in Au, Bi and Ta with x up to 10. Some of them are on the High Priority Nuclear Data Request List (HPRL). A comprehensive comparison of the measured data with theoretical predictions using the TALYS code is presented. While the calculated cross-sections agree with the measurements for lower-threshold reactions, severe discrepancies are observed for higher (n,xn) reactions for any of the TALYS model parameters.

Rapid and sustainable HPLC method for the determination of uremic toxins in human plasma samples.

Protein-bound uremic toxins, mainly indoxyl sulfate (3-INDS), p-cresol sulfate (pCS), and indole-3-acetic acid (3-IAA) but also phenol (Pol) and p-cresol (pC), are progressively accumulated during chronic kidney disease (CKD). Their accurate measurement in biomatrices is demanded for timely diagnosis and adoption of appropriate therapeutic measures. Multianalyte methods allowing the establishment of a uremic metabolite profile are still missing. Hence, the aim of this work was to develop a rapid and sensitive method based on high-performance liquid chromatography with fluorescence detection for the simultaneous quantification of Pol, 3-IAA, pC, 3-INDS, and pCS in human plasma. Separation was attained in 12min, using a monolithic C18 column and isocratic elution with acetonitrile and phosphate buffer containing an ion-pairing reagent, at a flow rate of 2mLmin-1. Standards were prepared in plasma and quantification was performed using the background subtraction approach. LOQ values were ≤ 0.2µgmL-1 for all analytes except for pCS (LOQ of 2µgmL-1). The method proved to be accurate (93.5-112%) and precise (CV ≤ 14.3%). The multianalyte application of the method, associated to a reduced sample volume (50 µL), a less toxic internal standard (eugenol) in comparison to the previously applied 2,6-dimethylphenol and 4-ethylphenol, and a green extraction solvent (ethanol), resulted in the AGREE score of 0.62 which is in line with the recent trend of green and sustainable analytical chemistry. The validated method was successfully applied to the analysis of plasma samples from control subjects exhibiting normal levels of uremic toxins and CKD patients presenting significantly higher levels of 3-IAA, pC, 3-INDS, and pCS that can be further investigated as biomarkers of disease progression.

A sparse tensor optimization approach for background subtraction from compressive measurements

Background subtraction from compressive measurements (BSCM) is a fundamental and critical task in video surveillance. Existing methods have limitations for incorporating the structural information and exhibit degraded performance in dynamic background, shadow and complex natural scenes. To address this issue, we propose a new Tucker decomposition-based sparse tensor optimization problem, which makes full use of the spatio-temporal features embedded in the video. The l0-norm in the objective function is used to constrain the sparseness of the spatio-temporal structure of video foreground, which enhances the spatio-temporal continuity and improves the accuracy of foreground detection. The orthogonality constraints on factor matrices in low-rank Tucker decomposition are used to characterize the spatio-temporal correlation of video background, which enhances low-rank characterization and makes better background estimation. Optimality analysis in terms of Karush-Kuhn-Tucker (KKT) conditions is addressed for the proposed sparse tensor optimization problem and a hard-threshing based alternating direction method of multipliers (HT-ADMM) is designed. Comprehensive experiments are conducted on real-world video datasets to demonstrate the effectiveness and superiority of our approach for BSCM.

Analyzing ophthalmic suspension particle size distributions using laser diffraction: Placebo background subtraction method

The current study demonstrated that the presence of excipients can interfere with the measurement of particle size distribution (PSD), a critical quality attribute of ophthalmic suspensions, by laser diffraction (LD) and that a placebo background subtraction approach can eliminate the impact of excipients on the PSD measurement. Commercially available loteprednol etabonate and brinzolamide ophthalmic suspensions were used as model suspensions. The impact of excipients in these formulations on the LD measurements was determined using a one-factor-at-a-time experimental design approach, using National Institute of Standards and Technology (NIST) traceable polystyrene particle size standards as references. Among the evaluated excipients, polymers containing polyacrylic acid were found to interfere with the PSD analysis by creating the LD signals correspond to particles ranging from a few micrometers to a hundred micrometers in size. As a result, the measured PSD of active pharmaceutical ingredient (API) particles in the formulation overlapped with or superimposed on the excipient PSD signal, leading to erroneous interpretation of the API particle size. Additionally, dispersion of brinzolamide particles in unsaturated solutions led to rapid dissolution of brinzolamide particles during the measurement, resulting in underestimation of the particle size range. Here, a placebo background subtraction approach was developed to eliminate the interference of the excipients. This newly developed LD method was also evaluated using orthogonal methods, including polarized light microscopy and scanning electron microscopy (SEM). The strategy used in this study to eliminate the interference of excipients may also be useful for other heterogeneous dispersions where excipient interference may be of concern.

Improved cellulose X-ray diffraction analysis using Fourier series modeling

This paper addresses two fundamental issues in the peak deconvolution method of cellulose XRD data analysis: there is no standard model for amorphous cellulose and common peak functions such as Gauss, Lorentz and Voigt functions do not fit the amorphous profile well. It first examines the effects of ball milling on three types of cellulose and results show that ball milling transforms all samples into a highly amorphous phase exhibiting nearly identical powder X-ray diffraction (XRD) profiles. It is hypothesized that short range order within a glucose unit and between adjacent units survives ball milling and generates the characteristic amorphous XRD profiles. This agrees well with cellulose I d-spacing measurements and oligosaccharide XRD analysis. The amorphous XRD profile is modeled using a Fourier series equation where the coefficients are determined using the nonlinear least squares method. A new peak deconvolution method then is proposed to analyze cellulose XRD data with the amorphous Fourier model function in conjunction with standard Voigt functions representing the crystalline peaks. The impact of background subtraction method has also been assessed. Analysis of several cellulose samples was then performed and compared to the conventional peak deconvolution methods with common peak fitting functions and background subtraction approach. Results suggest that prior peak deconvolution methods overestimate cellulose crystallinity.

A background subtraction approach for determination of endogenous cortisol and 6β-hydroxycortisol in urine by UPLC-MS/MS with application in a within-day variability study in HIV-infected pregnant women

Different methods have been used for CYP3A phenotyping, such as probe drugs or the urinary index 6β-hydroxycortisol/cortisol ratio (6β-OHF:C). This work describes a simple and affordable method for the simultaneous determination of the endogenous compounds cortisol and 6β-hydroxycortisol in urine using a background subtraction approach. The method was applied to investigate the CYP3A activity in HIV-infected pregnant women (n = 9) in the third trimester and postpartum periods. Also, the within-day variability in the 6β-OHF:C index was also evaluated. The sample preparation consists of a pre-cleanup with acetonitrile followed by liquid-liquid extraction with ethyl acetate. The analytes were resolved by employing an Acquity UPLC®BEH C18 column with a mobile phase that consisted of a mixture of acetonitrile containing 0.1% formic acid and 0.1% formic acid in gradient mode. The method presented linearities of 1–1.000 ng/mL and 2–1.000 ng/mL for C and 6β-OHF, respectively, and presented acceptable precision and accuracy. Qualitative and quantitative matrix effects tests were also performed. A high 6β-OHF:C within-day variability was observed in both phases. In the third trimester period, the 6β-OHF:C ranged from 2.57 to 51.69, with a mean ± standard deviation (SD) of 15.12 ± 5.41 (n = 9). Similar values were obtained in the postpartum period, with 6β-OHF:C ranging from 3.48 to 44.54 with a mean ± SD of 14.37 ± 5.73 (n = 7). Even though the 6β-OHF:C is a non-invasive index for CYP3A phenotyping, its use is susceptible to high within-day variability.

Background subtraction in diffraction X-ray images using deep CNN

Diffraction X-ray images provide molecular level information in tissue under hydrated, physiological conditions at the physiologically relevant millisecond time scale. When processing diffraction x-ray images there is a need to subtract background produced during the capture process prior to making measurements. This is a non-uniform background that is strongest at the diffraction center and decays with increased distance from the center. Existing methods require careful parameter selection or assume a specific background model. In this paper we propose a novel approach for background subtraction in which we learn to subtract background based on labeled examples. The labeled examples are image pairs where in each pair one of the images has diffraction background and the second has the background removed. Using a deep convolutional neural network (CNN) we learn to map an image with background to an image without it. Experimental results demonstrate that the proposed approach is capable of learning background removal with results close to ground truth data (PSNR > 68, SSIM > 0.99) and without having to manually select background parameters.

FOREGROUND DETECTION IN DYNAMIC SCENES OF INTELLIGENT TRANSPORT SYSTEMS

У статті розглядаються аспекти виявлення рухомих об'єктів у динамічному просторі інтелектуальної транспортної системи на основі комп'ютерного зору і штучного інтелекту. Були розглянуті традиційні і нові моделі фонового моделювання. Непараметричний підхід для визначення фону, що базується на ядерному оцінюванні, був використаний як найбільш придатний для використання в динамічному середовищі. Значення оціночної функції щільності ядра для кожного пікселя вихідного зображення порівнювалося із граничним значенням, оціненим за методом Оцу. Запропонований метод побудови ядерної оцінки щільності був перевірений на відеопотоці, що містить рухомі об'єкти, і показав прийнятну ефективність для застосування в безпілотних літальних апаратах.

Moving Object Detection Using Adaptive Blind Update and RGB-D Camera

A novel background subtraction approach using an RGB-D camera and an adaptive blind updating policy is introduced. This method in the initialization creates a model to store background pixels to compare each pixel of the new frame with the model in the same location to identify background pixels. The background-model update presented in this paper uses regular and blind updates which also has different criteria from existing methods. In particular, blind update frequently changes based on the background changes and the speed of moving object. This will allow the scene model to adapt to the changes in the background, detecting the stationary moving object and reducing the ghost phenomenon. In addition, the proposed bootstrapping segmentation and shadow detection are added to the system to improve the accuracy of the algorithm in shadow and depth camouflage scenarios. The proposed method is compared with the original method and the other state of the art algorithms. The experimental results show significant improvement in those videos that stationary object appears. In addition, the benchmark results also indicate strong and stable results compared to the other state of the art algorithms.

An Effective Subsuperpixel-Based Approach for Background Subtraction

How to achieve competitive accuracy and less computation time simultaneously for background estimation is still an intractable task. In this paper, an effective background subtraction approach for video sequences is proposed based on a subsuperpixel model. In our algorithm, the superpixels of the first frame are constructed using a simple linear iterative clustering method. After transforming the frame from a color format to gray level, the initial superpixels are divided into K smaller units, i.e., subsuperpixels, via the k -means clustering algorithm. The background model is then initialized by representing each subsuperpixel as a multidimensional feature vector. For the subsequent frames, moving objects are detected by the subsuperpixel representation and a weighting measure. In order to deal with ghost artifacts, a background model updating strategy is devised, based on the number of pixels represented by each cluster center. As each superpixel is refined via the subsuperpixel representation, the proposed method is more efficient and achieves a competitive accuracy for background subtraction. Experimental results demonstrate the effectiveness of the proposed method.

EFF-ViBE: An Efficient and Improved Background Subtraction Approach based on ViBE

EFF-ViBE: An Efficient and Improved Background Subtraction Approach based on ViBE

Dynamic Background Subtraction Using Histograms Based on Fuzzy C-Means Clustering and Fuzzy Nearness Degree

Background subtraction has been widely used in the detection of a moving object from a still scene. Due to the uncertainty in the classification of the pixels in the foreground and background, we propose a novel fuzzy approach for background subtraction using fuzzy histograms based on fuzzy c-means clustering and the fuzzy nearness degree, called FCFN. In this method, the temporal characteristics of the pixels are described by a fuzzy histogram using the fuzzy c-means algorithm. The segmentation threshold is adaptively calculated according to the distribution of the fuzzy nearness degree of the individual pixel. Fuzzy adaptive background maintenance is adopted in the background update framework. The performance of the FCFN is evaluated against several state-of-the-art methods in the complex dynamic scenes. The experimental results demonstrate that the proposed method doubles the improvements in performance than the classic fuzzy background modeling methods and outperforms most state-of-the-art methods.

Cut set-based Dynamic Key frame selection and Adaptive Layer-based Background Modeling for background subtraction

Background subtraction has been widely discussed in video surveillance, but it still has open challenges such as dynamic background, illumination variation. To address these challenges a novel Cut set-based Dynamic Key frame selection (CDK) and Adaptive Layer-based Background Modeling (ALBM) approach for background subtraction is proposed which adaptively changes layers in the background model for each scenario such as static, dynamic background and high illumination.The concept of key frame is used to choose representative frames from the video. In order to capture the invariant directional codes of each spatio-temporal patch symmetric operators such as line and rotational symmetry are used. The proposed method identifies highly similar static spatio-temporal patches and sets it as background there by reducing the computational complexity in the foreground detection step. Both qualitative and quantitative evaluations on challenging video sequences demonstrate that the proposed algorithm performs background subtraction more favorably than the state-of-the-art methods.

Efficient background subtraction for thermal images using reflectional symmetry pattern (RSP)

Nowadays, thermal image processing has gained more attention. Thermal camera’s cost is decreasing, and so many real-time applications use thermal cameras since they have an ability to detect objects in darkness and track objects in the video. A background subtraction approach using Reflectional Symmetry Pattern (RSP) for thermal image background subtraction is proposed based on the assumption that the geometric reflectional symmetrical pattern of each of the objects (person) is much lower than the surrounding background. Reflectional symmetrical texture pattern can be used to create a subspace from the result of frame differencing approach. Statistical parameters such as VP (Vector Product), VD (Vector Direction) can be used to create an accurate background model. As a result, the proposed scheme can provide a high precision and less error rate to meet the requirements of object detection from real-time thermal videos.

Object Detection and Tracking using Modified Diamond Search Block Matching Motion Estimation Algorithm

Object tracking is one of the main fields within computer vision. Amongst various methods/ approaches for object detection and tracking, the background subtraction approach makes the detection of object easier. To the detected object, apply the proposed block matching algorithm for generating the motion vectors. The existing diamond search (DS) and cross diamond search algorithms (CDS) are studied and experiments are carried out on various standard video data sets and user defined data sets. Based on the study and analysis of these two existing algorithms a modified diamond search pattern (MDS) algorithm is proposed using small diamond shape search pattern in initial step and large diamond shape (LDS) in further steps for motion estimation. The initial search pattern consists of five points in small diamond shape pattern and gradually grows into a large diamond shape pattern, based on the point with minimum cost function. The algorithm ends with the small shape pattern at last. The proposed MDS algorithm finds the smaller motion vectors and fewer searching points than the existing DS and CDS algorithms. Further, object detection is carried out by using background subtraction approach and finally, MDS motion estimation algorithm is used for tracking the object in color video sequences. The experiments are carried out by using different video data sets containing a single object. The results are evaluated and compared by using the evaluation parameters like average searching points per frame and average computational time per frame. The experimental results show that the MDS performs better than DS and CDS on average search point and average computation time.

Untargeted analysis of protein modification by drug-related chemicals using background subtraction and proteomic approach

Untargeted analysis of protein modification by drug-related chemicals using background subtraction and proteomic approach

Camera-trap images segmentation using multi-layer robust principal component analysis

The segmentation of animals from camera-trap images is a difficult task. To illustrate, there are various challenges due to environmental conditions and hardware limitation in these images. We proposed a multi-layer robust principal component analysis (multi-layer RPCA) approach for background subtraction. Our method computes sparse and low-rank images from a weighted sum of descriptors, using color and texture features as case of study for camera-trap images segmentation. The segmentation algorithm is composed of histogram equalization or Gaussian filtering as pre-processing, and morphological filters with active contour as post-processing. The parameters of our multi-layer RPCA were optimized with an exhaustive search. The database consists of camera-trap images from the Colombian forest taken by the Instituto de Investigaci\'on de Recursos Biol\'ogicos Alexander von Humboldt. We analyzed the performance of our method in inherent and therefore challenging situations of camera-trap images. Furthermore, we compared our method with some state-of-the-art algorithms of background subtraction, where our multi-layer RPCA outperformed these other methods. Our multi-layer RPCA reached 76.17 and 69.97% of average fine-grained F-measure for color and infrared sequences, respectively. To our best knowledge, this paper is the first work proposing multi-layer RPCA and using it for camera-trap images segmentation.