Subtraction Approach Research Articles

Algorithm for particle detection and parameterization in high-frame-rate thermal video

Over the last decade, feature tracking and recognition in infrared (IR) video has become an important strategy used in many applications. To achieve such a capability, we developed a method based on the top-hat transform, hybridized with refinement by thresholding. Our algorithm uses two different but correlated background subtraction approaches to clean the image. A mathematical-morphology-based method was then applied to enhance the contrast between particles and background. The algorithm was tested using images acquired during a controlled experiment and was compared with another particle tracking velocimetry method. We demonstrate that our algorithm can detect dim IR targets and enables computation of a local velocity field that can be used for the tracking step. Using this method, we were able to obtain both the distribution of particle sizes, volumes (or masses), and velocities. We also apply our algorithm to images recorded during ballistic emitting explosive events at Stromboli volcano (Italy) and favorably compare our results with other volcanologic data sets. Experimental results demonstrate that our algorithm achieves a high recognition accuracy with a low-computational cost.

Arbitrary cylinder color model for the codebook based background subtraction.

The codebook background subtraction approach is widely used in computer vision applications. One of its distinguished features is the cylinder color model used to cope with illumination changes. The performances of this approach depends strongly on the color model. However, we have found this color model is valid only if the spectrum components of the light source change in the same proportion. In fact, this is not true in many practical cases. In these cases, the performances of the approach would be degraded significantly. To tackle this problem, we propose an arbitrary cylinder color model with a highly efficient updating strategy. This model uses cylinders whose axes need not going through the origin, so that the cylinder color model is extended to much more general cases. Experimental results show that, with no loss of real-time performance, the proposed model reduces the wrong classification rate of the cylinder color model by more than fifty percent.

Rapid and robust pulmonary proton ZTE imaging in the mouse.

Pulmonary MRI is challenging because of the low proton density and rapid transverse relaxation in the lung associated with microscopic magnetic field inhomogeneities caused by tissue-air interfaces. Therefore, low signal is obtained in gradient and spin echo proton images. Alternatively, non-proton MRI using hyperpolarized gases or radial techniques with ultrashort or zero TE have been proposed to image the lung. Also with the latter approach, the general challenge remains to provide full coverage of the lung at sufficient spatial resolution, signal-to-noise ratio (SNR) and image quality within a reasonable scan time. This task is further aggravated by physiological motion and is particularly demanding in small animals, such as mice. In this work, three-dimensional (3D) zero echo time (ZTE) imaging is employed for efficient pulmonary MRI. Four protocols with different averaging and respiratory triggering schemes are developed and compared with respect to image quality and SNR. To address the critical issue of background signal in ZTE images, a subtraction approach is proposed, providing images virtually free of disturbing signal from nearby hardware parts. The protocols are tested for pulmonary MRI in six mice at 4.7 T, consistently providing images of high quality with a 3D isotropic resolution of 313 µm and SNR values in the lung between 8.0 and 18.5 within scan times between 1 min 21 s and 4 min 44 s. A generally high robustness of the ZTE approach against motion is observed, whilst respiratory triggering further improves the SNR and visibility of image details. The developed techniques are expected to enable efficient preclinical animal studies in the lung and will also be of importance for human applications. Further improvements are expected from radiofrequency (RF) coils with increased SNR and reduced background signal.

Conteo de personas con un sensor RGBD comercial

In this paper, we prove that depth information provided by a consumer depth camera is a reliable data source to perform ro- bust people counting. The adoption of a top view configuration reduces the space problem complexity for this task, while pre- serving privacy. Two different background subtraction approaches for color images are transferred to this context and tested in real video to perform detection, tracking, and behavioral pat- terns analysis of subjects crossing the field of view. The results achieved in an experimental setup with real video reported a TPR over 95%, beating robust GMM background subtraction based only on the visual cue. The results suggest the benefits of the depth cue for this particular task.

Neural Background Subtraction for Pan-Tilt-Zoom Cameras

We propose an extension of a neural-based background subtraction approach to moving object detection to the case of image sequences taken from pan-tilt-zoom (PTZ) cameras. The background model automatically adapts in a self-organizing way to changes in the scene background. Background variations arising in a usual stationary camera setting, such as those due to gradual illumination changes, to waving trees, or to shadows cast by moving objects, are accurately handled by the neural self-organizing background model originally proposed for this type of setting. Handling of variations due to the PTZ camera movement is ensured by a novel registration mechanism that allows the neural background model to automatically compensate the eventual ego-motion, estimated at each time instant. Experimental results on several real image sequences and comparisons with seven state-of-the-art methods demonstrate the accuracy of the proposed approach.

A subtraction pipeline for automatic detection of new appearing multiple sclerosis lesions in longitudinal studies

Time-series analysis of magnetic resonance images (MRI) is of great value for multiple sclerosis (MS) diagnosis and follow-up. In this paper, we present an unsupervised subtraction approach which incorporates multisequence information to deal with the detection of new MS lesions in longitudinal studies. The proposed pipeline for detecting new lesions consists of the following steps: skull stripping, bias field correction, histogram matching, registration, white matter masking, image subtraction, automated thresholding, and postprocessing. We also combine the results of PD-w and T2-w images to reduce false positive detections. Experimental tests are performed in 20 MS patients with two temporal studies separated 12 (12M) or 48 (48M) months in time. The pipeline achieves very good performance obtaining an overall sensitivity of 0.83 and 0.77 with a false discovery rate (FDR) of 0.14 and 0.18 for the 12M and 48M datasets, respectively. The most difficult situation for the pipeline is the detection of very small lesions where the obtained sensitivity is lower and the FDR higher. Our fully automated approach is robust and accurate, allowing detection of new appearing MS lesions. We believe that the pipeline can be applied to large collections of images and also be easily adapted to monitor other brain pathologies.

Vessel segmentation in MRI using a variational image subtraction approach

Vessel segmentation is important for many clinical applications, such as the diagnosis of vascular diseases, the planning of surgery, or the monitoring of the progress of disease. Although various approaches have been proposed to segment vessel structures from 3-dimensional medical images, to the best of our knowledge, there has been no known technique that uses magnetic resonance imaging (MRI) as prior information within the vessel segmentation of magnetic resonance angiography (MRA) or magnetic resonance venography (MRV) images. In this study, we propose a novel method that uses MRI images as an atlas, assuming that the patient has an MRI image in addition to MRA/MRV images. The proposed approach intends to increase vessel segmentation accuracy by using the available MRI image as prior information. We use a rigid mutual information registration of the MRA/MRV to the MRI, which provides subvoxel accurate multimodal image registration. On the other hand, vessel segmentation methods tend to mostly suer from imaging artifacts, such as Rician noise, radio frequency (RF) inhomogeneity, or partial volume eects that are generated by imaging devices. Therefore, this proposed method aims to extract all of the vascular structures from MRA/MRI or MRV/MRI pairs at the same time, while minimizing the combined eects of noise and RF inhomogeneity. Our method is validated both quantitatively and visually using BrainWeb phantom images and clinical MRI, MRA, and MRV images. Comparison and observer studies are also realized using the BrainWeb database and clinical images. The computation time is markedly reduced by developing a parallel implementation using the Nvidia compute unied device architecture and OpenMP frameworks in order to allow the use of the method in clinical settings.

Robust Background Subtraction for Network Surveillance in H.264 Streaming Video

The H.264/Advanced Video Coding (AVC) is the industry standard in network surveillance offering the lowest bitrate for a given perceptual quality among any MPEG or proprietary codecs. This paper presents a novel approach for background subtraction in bitstreams encoded in the Baseline profile of H.264/AVC. Temporal statistics of the proposed feature vectors, describing macroblock units in each frame, are used to select potential candidates containing moving objects. From the candidate macroblocks, foreground pixels are determined by comparing the colors of corresponding pixels pair-wise with a background model. The basic contribution of the current work compared to the related approaches is that, it allows each macroblock to have a different quantization parameter, in view of the requirements in variable as well as constant bit-rate applications. Additionally, a low-complexity technique for color comparison is proposed which enables us to obtain pixel-resolution segmentation at a negligible computational cost as compared to those of classical pixel-based approaches. Results showing striking comparison against those of proven state-of-the-art pixel domain algorithms are presented over a diverse set of standardized surveillance sequences.

Single-trial time–frequency analysis of electrocortical signals: Baseline correction and beyond

Event-related desynchronization (ERD) and synchronization (ERS) of electrocortical signals (e.g., electroencephalogram [EEG] and magnetoencephalogram) reflect important aspects of sensory, motor, and cognitive cortical processing. The detection of ERD and ERS relies on time–frequency decomposition of single-trial electrocortical signals, to identify significant stimulus-induced changes in power within specific frequency bands. Typically, these changes are quantified by expressing post-stimulus EEG power as a percentage of change relative to pre-stimulus EEG power. However, expressing post-stimulus EEG power relative to pre-stimulus EEG power entails two important and surprisingly neglected issues. First, it can introduce a significant bias in the estimation of ERD/ERS magnitude. Second, it confuses the contribution of pre- and post-stimulus EEG power. Taking the human electrocortical responses elicited by transient nociceptive stimuli as an example, we demonstrate that expressing ERD/ERS as the average percentage of change calculated at single-trial level introduces a positive bias, resulting in an overestimation of ERS and an underestimation of ERD. This bias can be avoided using a single-trial baseline subtraction approach. Furthermore, given that the variability in ERD/ERS is not only dependent on the variability in post-stimulus power but also on the variability in pre-stimulus power, an estimation of the respective contribution of pre- and post-stimulus EEG variability is needed. This can be achieved using a multivariate linear regression (MVLR) model, which could be optimally estimated using partial least square (PLS) regression, to dissect and quantify the relationship between behavioral variables and pre- and post-stimulus EEG activities. In summary, combining single-trial baseline subtraction approach with PLS regression can be used to achieve a correct detection and quantification of ERD/ERS.

An effective method for determining the ingredients of Shuanghuanglian formula in blood samples using high‐resolution LC–MS coupled with background subtraction and a multiple data processing approach

Shuanghuanglian formula (SF) is a combination of Flos lonicerae japonicae, Radix scutellariae, and Fructus forsythiae, commonly used to treat viral or bacterial infections. However, the constituents absorbed into the blood after oral administration of SF are difficult to determine and thus remain unclear. Here, we report the application of an accurate background subtraction and multiple data processing approach (Bs-Mpa) for the comprehensive detection of compounds of SF in vivo. A sensitive and reliable ultra-performance LC coupled with ESI quadrupole TOF MS (UPLC-ESI-Q-TOF-MS) approach coupled with Bs-Mpa, which is implemented in the Strip tool from UPLC to remove nonrelated ion signals from accurate mass LC-MS data, was established to characterize the chemical constituents and rat metabolites of SF. In the loading plot of the principal component analysis, 68 ions of interest were extracted from blood samples, among them, 39 absorbed prototype components of SF and 29 metabolites were identified in vivo. It is concluded that the integrative Bs-Mpa method can be successfully applied for the rapid discovery of multiple components from a traditional Chinese medicine. The above challenge was addressed by using the proposed Bs-Mpa method and it was particularly suitable for applying to the global characterization of the constituents or metabolites in rat blood after oral administration of other well-known formulae.

An Efficient Hierarchical Approach for Background Subtraction and Shadow Removal using Adaptive GMM and Color Discrimination

paper presents an efficient approach for moving objects detection and shadow removal from color videos obtained using stationary camera. A background subtraction technique based on modified adaptive GMM has been proposed for detecting moving objects. Speed-up techniques have also been applied to enhance the computational efficiency of the algorithm. Then, a robust algorithm for shadow removal is used to remove cast shadows and ghosts. Foreground is reconstructed using graph cut based cleaning and non-recursive blob finding. Comparative experimental results demonstrate that proposed approach performs better in comparison to other state-of-the-art algorithms.

Automated Quantitative Spectroscopic Analysis Combining Background Subtraction, Cosmic Ray Removal, and Peak Fitting

An integrated concept for post-acquisition spectrum analysis was developed for in-line (real-time) and off-line applications that preserves absolute spectral quantification; after the initializing parameter setup, only minimal user intervention is required. This spectral evaluation suite is composed of a sequence of tasks specifically addressing cosmic ray removal, background subtraction, and peak analysis and fitting, together with the treatment of two-dimensional charge-coupled device array data. One may use any of the individual steps on their own, or may exclude steps from the chain if so desired. For the background treatment, the canonical rolling-circle filter (RCF) algorithm was adopted, but it was coupled with a Savitzky-Golay filtering step on the locus-array generated from a single RCF pass. This novel only-two-parameter procedure vastly improves on the RCF's deficiency to overestimate the baseline level in spectra with broad peak features. The peak analysis routine developed here is an only-two-parameter (amplitude and position) fitting algorithm that relies on numerical line shape profiles rather than on analytical functions. The overall analysis chain was programmed in National Instrument's LabVIEW; this software allows for easy incorporation of this spectrum analysis suite into any LabVIEW-managed instrument control, data-acquisition environment, or both. The strength of the individual tasks and the integrated program sequence are demonstrated for the analysis of a wide range of (although not necessarily limited to) Raman spectra of varying complexity and exhibiting nonanalytical line profiles. In comparison to other analysis algorithms and functions, our new approach for background subtraction, peak analysis, and fitting returned vastly improved quantitative results, even for "hidden" details in the spectra, in particular, for nonanalytical line profiles. All software is available for download.

Background subtraction using hybrid feature coding in the bag-of-features framework

Although numerous algorithms have been proposed for background subtraction with demonstrated success, it remains a challenging problem. One of the main reasons is the lack of effective background model to account for the complex variations of backgrounds. Although researchers have strived to obtain a background model effectively attenuating false positives from dynamic background variations, their methods are still sensitive to structured motion patterns of background (e.g., waving leaves, rippling water, spouting fountain, etc.). In this paper, inspired by the bag-of-features framework, we present a simple, novel, yet powerful approach for background subtraction. It relies on the hypothesis that texture variations in the background scenes can be well attenuated by effectively encoding the local color and texture information. Specifically, the proposed method adopts joint domain-range features, which are encoded in the soft-assignment coding procedure. We also propose a novel method for deciding the appropriate kernel variances in the soft-assignment coding, which result in strong adaptability and robustness to dynamic scenes compared to employing fixed kernel variances. Experimental results demonstrate that our proposed method is able to handle severe textural variations of backgrounds and perform favorably against the state-of-the-art methods.

Analysis of QRS-T subtraction in unipolar atrial fibrillation electrograms

This paper presents a QRS-T subtraction approach for atrial fibrillation (AF) intracardiac atrial electrograms (AEG). It also presents a comparison between the proposed method and two alternative ventricular subtraction techniques: average beat subtraction (ABS) using a fixed length window and an approach based on flat interpolation for QRS cancellation. Areas of the atrium close to the mitral valve showed stronger ventricular influence on the AEGs when compared with the remaining atrial regions. Ventricular influence affects the spectral power distribution of the AEG and can also affect the estimation of the dominant frequency unless the whole ventricular activity influence (QRS-T) is removed. The average power after QRS-T subtraction is significantly reduced for frequencies above 10Hz (mostly associated with QRS complexes), as well as for frequencies between 3 and 5.5Hz, (mostly related to T waves). The results indicate that the proposed approach removes ventricular influence on the AF AEGs better than the QRS cancellation method. Spectral analysis showed that both the ABS and the proposed method do well and no method should be preferred to the other. In the time domain, the proposed approach is matched to the lengths and timings of onset and offset for individual QRS-T segments while the ABS approach uses an arbitrary length around the QRS for the pattern used for QRS-T removal.

A System based on Adaptive Background Subtraction Approach for Moving Object Detection and Tracking in Videos

Video surveillance systems are based on video and image processing research areas in the scope of computer science. Video processing covers various methods which are used to browse the changes in existing scene for specific video. Nowadays, video processing is one of the important areas of computer science. Two-dimensional videos are used to apply various segmentation and object detection and tracking processes which exists in multimedia content-based indexing, information retrieval, visual and distributed cross-camera surveillance systems, people tracking, traffic tracking and similar applications. Background subtraction (BS) approach is a frequently used method for moving object detection and tracking. In the literature, there exist similar methods for this issue. In this research study, it is proposed to provide a more efficient method which is an addition to existing methods. According to model which is produced by using adaptive background subtraction (ABS), an object detection and tracking system’s software is implemented in computer environment. The performance of developed system is tested via experimental works with related video datasets. The experimental results and discussion are given in the study

Evaluation of Whole Atmosphere Community Climate Model simulations of ozone during Arctic winter 2004–2005

The work presented here evaluates polar stratospheric ozone simulations from the Whole Atmosphere Community Climate Model (WACCM) for the Arctic winter of 2004–2005. We use the Specified Dynamics version of WACCM (SD‐WACCM), in which temperatures and winds are nudged to meteorological assimilation analysis results. Model simulations of ozone and related constituents generally compare well to observations from the Earth Observing System Microwave Limb Sounder (MLS). At most times, modeled ozone agrees with MLS data to within ~10%. However, a systematic high bias in ozone in the model of ~18% is found in the lowermost stratosphere in March. We attribute most of this ozone bias to too little heterogeneous processing of halogens late in the winter. We suggest that the model under‐predicts ClONO2 early in the winter, which leads to less heterogeneous processing and too little activated chlorine. Model HCl could also be overestimated due to an underestimation of HCl uptake into supercooled ternary solution (STS) particles. In late winter, the model overestimates gas‐phase HNO3, and thus NOy, which leads to an over‐prediction of ClONO2 (under‐prediction of activated chlorine). A sensitivity study, in which temperatures for heterogeneous chemistry reactions were reduced by 1.5 K, shows significant improvement of modeled ozone. Chemical ozone loss is inferred from the MLS observations using the pseudo‐passive subtraction approach. The inferred ozone loss using this method is in agreement with or less than previous independent results for the Arctic winter of 2004–2005, reaching 1.0 ppmv on average and up to 1.6 ppmv locally in the polar vortex.

Quantitative measurements of glutathione in yeast cell lysate using 1H NMR

Methods for quantifying the level of glutathione (GSH) in yeast cell lysate are described using (1)H NMR analysis. For quantification purposes, the (1)H resonances corresponding to the Cys βCH2 of GSH were identified as having the fewest overlapping spectral interferences from lysate matrix components using GSH spiked yeast lysate samples. Two methods, standard addition based on peak integration and a spectral subtraction approach, were evaluated for quantifying GSH in lysate samples. The peak integration procedure required baseline estimation and a peak fitting step to correct for background interferences while the spectral subtraction procedure was comparatively straightforward. The level of GSH measured by (1)H NMR was in good agreement with the concentration measured by the DTNB-GSSG reductase recycling assay. The proposed NMR method can lead to a reliable quantitation of GSH and could be applicable to a variety of other analytes of interest in complex biological matrices.

Spatio-Temporal Traffic Scene Modeling for Object Motion Detection

Moving object detection is an important component of a traffic surveillance system. Usual background subtraction approaches often poorly perform on a long outdoor traffic video due to vehicles waiting at an intersection and gradual changes of illumination and background shadow position. We present a fast and robust background subtraction algorithm based on unified spatio-temporal background and foreground modeling. The correlation between neighboring pixels provides high levels of detection accuracy in the dynamic background scene. Our Bayesian fusion method, which establishes the traffic scene model, combines both background and foreground models and considers prior probabilities to adapt changes of background in each frame. We explicitly model both temporal and spatial information based on the kernel density estimation (KDE) formulation for background modeling. Then, we use a Gaussian formulation to describe the spatial correlation of moving objects for foreground modeling. In the updating step, a fusion background frame is generated, and reasonable updating rates are also proposed for the traffic scene. The experimental results show that the proposed method outperforms the previous work with less computation and is better suited for the traffic scenes.

Dual-tracer background subtraction approach for fluorescent molecular tomography

Diffuse fluorescence tomography requires high contrast-to-background ratios to accurately reconstruct inclusions of interest. This is a problem when imaging the uptake of fluorescently labeled molecularly targeted tracers in tissue, which can result in high levels of heterogeneously distributed background uptake. We present a dual-tracer background subtraction approach, wherein signal from the uptake of an untargeted tracer is subtracted from targeted tracer signal prior to image reconstruction, resulting in maps of targeted tracer binding. The approach is demonstrated in simulations, a phantom study, and in a mouse glioma imaging study, demonstrating substantial improvement over conventional and homogenous background subtraction image reconstruction approaches.

A robust approach for background subtraction with shadow removal for moving object detection

Background subtraction is the primary step that can be applied in numerous applications, including object tracking, surveillance, robot vision, object-based coding, image database and video teleconferencing. However, background subtraction approaches are very sensitive to dynamic scene changes. This paper aims to investigate ways to exploit the characteristics of the wavelet domain to improve moving object segmentation using background subtraction for object tracking applications. In this paper, we also innovate by using a high frequency sub-band to neutralise the negative influences caused by cast shadow, without using any colour information. The state-of-theart performance of the proposed approach is illustrated by using both indoor and outdoor video sequences.