Motion Detection Technique Research Articles

Development and Implementation of a New Dynamic Face Detection Operator

work is part of a project dedicated to the development of a real-time multimodal biometric identification system. In this kind of applications, face detection is an important step before any identification process. This paper presents the development and the implementation of a dynamic face detector based on both skin color and motion detection techniques. Our choice will be guided by the computational complexity and timing constraints for a real-time processing on a low cost Virtex FPGA platform.

Mitigating motion artifacts in FDK based 3D Cone-beam Brain Imaging System using markers

Abstract Head motion during Computed Tomographic (CT) brain imaging studies can adversely affect the reconstructed image through distortion, loss of resolution and other related artifacts. In this paper, we propose a marker based innovative approach to detect and mitigate motion artifacts in three dimensional cone-beam brain CT systems without using any external motion tracking sensor. Motion is detected using correlations between the adjacent projections. Once motion is detected, motion parameters (i.e. six degrees-of-freedom of motions) are estimated using a numerical optimization technique. Artifacts, caused by motions, are mitigated by using a modified form Feldkemp-Davis-Kress (FDK) algorithm which uses the estimated motion parameters in back-projection stage. The proposed approach has been evaluated on a modified three-dimensional Shepp-Logan phantom with a range of simulated motions. Simulation results demonstrate a quantitative and qualitative validation of motion detection and artifacts mitigation technique.

A debris image tracking using optical flow algorithm

This study proposes a motion detection and object tracking technique for GEO debris in a sequence of images. A couple of techniques (called the “stacking method” and “line-identifying technique”) were recently proposed to address the same problem. Although these techniques are effective at detecting the debris position and motion in the image sequences, there are some issues concerned with computational load and assumed debris motion. This study derives a method to estimate motion vectors of objects in image sequence and finally detect the debris locations by using a computer vision technique called an optical flow algorithm. The new method detects these parameters in low computational time in a serial manner, which implies that it has an advantage to track not only linear but also nonlinear motion of GEO debris more easily than the previous methods. The feasibility of the proposed methods is validated using real and synthesized image sequences which contain some typical debris motions.

Color rolling suppression algorithm considering luminance and color constancy

When a digital image sequence is sampled with a periodically emitting light source, the color rolling (CR) phenomenon occurs, which is shown by periodical variations of color and luminance values. In conventional CR suppression (CRS) methods, color variation has been reduced by using auto white balance methods. However, the CR phenomenon still appears in the resulting image sequences due to interfield illuminant intensity variation. In the proposed CRS method, the interfield luminance and color variations are simultaneously suppressed by estimating the illuminant change between the current and the target fields. In order to consider the object motions, a motion detection technique is used to estimate the luminance changes that occurred due to the CR phenomenon. Moreover, the illuminant color is estimated using the CR achromatic color distribution in the chromaticity space which is founded on the periodicity of the CR phenomenon. Based on the motion detection and the achromatic color detection techniques, the illuminant is estimated using the obtained color components in a common area of both static and achromatic regions. The experimental results demonstrate that our strategy efficiently suppresses the CR phenomenon without being affected by moving objects and produces luminance and color constant image sequences.

Integrating location tracking, traffic monitoring and semantics in a layered ITS architecture

The new generation of Intelligent Transportation Systems (ITSs) should be able both to monitor in real-time the current user location and traffic conditions, and to access the heterogeneous data sources maintained by different organisations and useful for ITS applications. For this reason, a layered ITS architecture is proposed, which consists of a Wi-Fi communication layer for data exchange between users and control devices, a monitoring layer provided with fast algorithms to locate the users and to compute the travel times of a traffic network, an application layer to supervise the monitoring layer and to map the proprietary data sources into a standard format, i.e., the Resource Description Framework Schema (RDFS), and a semantic layer to retrieve and process RDFS data for supporting mobile applications. Implementation details are also given. In particular, a novel motion detection technique based on the Poisson distribution to evaluate the traffic parameters and a technique based on neural networks to locate the users are presented, whereas a suitable topology of the overall system to timely inform the mobile users is outlined. The architecture supports the internetworking of different monitoring and communication technologies to best fit a variety of users and use scenarios.

ViBe: A Universal Background Subtraction Algorithm for Video Sequences

This paper presents a technique for motion detection that incorporates several innovative mechanisms. For example, our proposed technique stores, for each pixel, a set of values taken in the past at the same location or in the neighborhood. It then compares this set to the current pixel value in order to determine whether that pixel belongs to the background, and adapts the model by choosing randomly which values to substitute from the background model. This approach differs from those based upon the classical belief that the oldest values should be replaced first. Finally, when the pixel is found to be part of the background, its value is propagated into the background model of a neighboring pixel. We describe our method in full details (including pseudo-code and the parameter values used) and compare it to other background subtraction techniques. Efficiency figures show that our method outperforms recent and proven state-of-the-art methods in terms of both computation speed and detection rate. We also analyze the performance of a downscaled version of our algorithm to the absolute minimum of one comparison and one byte of memory per pixel. It appears that even such a simplified version of our algorithm performs better than mainstream techniques.

Block-wise motion detection using compressive imaging system

A block-wise motion detection strategy based on compressive imaging, also referred to as feature-specific imaging (FSI), is described in this work. A mixture of Gaussian distributions is used to model the background in a scene. Motion is detected in individual object blocks using feature measurements. Gabor, Hadamard binary and random binary features are studied. Performance of motion detection methods using pixel-wise measurements is analyzed and serves as a baseline for comparison with motion detection techniques based on compressive imaging. ROC (Receiver Operation Characteristic) curves and AUC (Area Under Curve) metrics are used to quantify the algorithm performance. Because a FSI system yields a larger measurement SNR(Signal-to-Noise Ratio) than a traditional system, motion detection methods based on the FSI system have better performance. We show that motion detection algorithms using Hadamard and random binary features in a FSI system yields AUC values of 0.978 and 0.969 respectively. The pixel-based methods are only able to achieve a lower AUC value of 0.627. ► We studied pixel-wise and block-wise motion detection using Gaussian mixture model. ► Block-wise motion is also detected using extracted features from pixel measurements. ► Features directly measured in FSI system is also implemented for motion detection. ► Using features in FSI, motion detection method has significantly improved performance.

A Local Algorithm for the Computation of Image Velocity via Constructive Interference of Global Fourier Components

A novel Fourier-based technique for local motion detection from image sequences is proposed. In this method, the instantaneous velocities of local image points are inferred directly from the global 3D Fourier components of the image sequence. This is done by selecting those velocities for which the superposition of the corresponding Fourier gratings leads to constructive interference at the image point. Hence, image velocities can be assigned locally even though position is computed from the phases and amplitudes of global Fourier components (spanning the whole image sequence) that have been filtered based on the motion-constraint equation, reducing certain aperture effects typically arising from windowing in other methods. Regularization is introduced for sequences having smooth flow fields. Aperture effects and their effect on optic-flow regularization are investigated in this context. The algorithm is tested on both synthetic and real image sequences and the results are compared to those of other local methods. Finally, we show that other motion features, i.e. motion direction, can be computed using the same algorithmic framework without requiring an intermediate representation of local velocity, which is an important characteristic of the proposed method.

Probabilistic model-based silhouette refinement for gait recognition

An algorithm to refine and clean gait silhouette noises generated by imperfect motion detection techniques is developed, and a relatively complete and high quality silhouette is obtained. The silhouettes are sequentially refined in two levels according to two different probabilistic models. The first level is within-sequence refinement. Each silhouette in a particular sequence is refined by an individual model trained by the gait images from current sequence. The second level is between-sequence refinement. All the silhouettes that need further refinement are modified by a population model trained by the gait images chosen from a certain amount of pedestrians. The intention is to preserve the within-class similarity and to decrease the interaction between one class and others. Comparative experimental results indicate that the proposed algorithm is simple and quite effective, and it helps the existing recognition methods achieve a higher recognition performance.

Spatially Adaptive Subsampling for Motion Detection

Many video surveillance applications rely on efficient motion detection. However, the algorithms are usually costly since they compute a background model at every pixel of the frame. This paper shows that, in the case of a planar scene with a fixed calibrated camera, a set of pixels can be selected to compute the background model while ignoring the other pixels for accurate but less costly motion detection. The calibration is used to first define a volume of interest in the real world and to project the volume of interest onto the image, and to define a spatial adaptive subsampling of this region of interest with a subsampling density that depends on the camera distance. Indeed, farther objects need to be analyzed with more precision than closer objects. Tests on many video sequences have integrated this adaptive subsampling to various motion detection techniques.

Development and evaluation of a new fully automatic motion detection and correction technique in cardiac SPECT imaging

BackgroundIn cardiac SPECT perfusion imaging, motion correction of the data is critical to the minimization of motion introduced artifacts in the reconstructed images. Software-based (data-driven) motion correction techniques are the most convenient and economical approaches to fulfill this purpose. However, the accuracy is significantly affected by how the data complexities, such as activity overlap, non-uniform tissue attenuation, and noise are handled. MethodsWe developed STASYS, a new, fully automatic technique, for motion detection and correction in cardiac SPECT. We evaluated the performance of STASYS by comparing its effectiveness of motion correcting patient studies with the current industry standard software (Cedars-Sinai MoCo) through blind readings by two readers independently. ResultsFor 204 patient studies from multiple clinical sites, the first reader identified (1) 69 studies with medium to large axial motion, of which STASYS perfectly or significantly corrected 86.9% and MoCo 72.5%; and (2) 20 studies with medium to large lateral motion, of which STASYS perfectly or significantly corrected 80.0% and MoCo 60.0%. The second reader identified (1) 84 studies with medium to large axial motion, of which STASYS perfectly or significantly corrected 82.2% and MoCo 76.2%; and (2) 34 studies with medium to large lateral motion, of which STASYS perfectly or significantly corrected 58.9% and MoCo 50.0%. ConclusionsWe developed a fully automatic software-based motion correction technique, STASYS, for cardiac SPECT. Clinical studies showed that STASYS was effective and corrected a larger percent of cardiac SPECT studies than the current industrial standard software.

A Reliable FPGA-based Real-time Optical-flow Estimation

Optical flow is a research topic of interest for many years. It has, until recently, been largely inapplicable to real-time applications due to its computationally expensive nature. This paper presents a new reliable flow technique which is combined with a motion detection algorithm, from stationary camera image streams, to allow flow-based analyses of moving entities, such as rigidity, in real-time. The combination of the optical flow analysis with motion detection technique greatly reduces the expensive computation of flow vectors as compared with standard approaches, rendering the method to be applicable in real-time implementation. This paper describes also the hardware implementation of a proposed pipelined system to estimate the flow vectors from image sequences in real time. This design can process 768 × 576 images at a very high frame rate that reaches to 156 fps in a single low cost FPGA chip, which is adequate for most real-time vision applications.

Noise-robustness improvement of the H.264 video coder

When sequence noise increases, the prediction process in the H.264 advanced video coding (AVC) encoder becomes less efficient. The residue resulting from the difference between the matching macroblocks in the frame coded in predictive mode and the best reference frame increases dramatically. To solve this problem, we propose to embed in the H.264/AVC encoder a motion detection technique based on the Markov random fields algorithm, which relies on robust moving pixel segmentation and reduces the luminance variation that results more often from noise rather than motion. This algorithm needs three reference frames to detect real motion between matching macroblocks, and it allows a decrease in bit rate while maintaining the compressed sequence quality.

Multifrequency ultrasound radiation force excitation and motion detection of harmonically vibrating scatterers

Elasticity imaging is a medical imaging modality that creates images based on the mechanical response of tissue. In some techniques, ultrasound radiation force has been used to deform tissue and ultrasonic or other techniques detect the tissue displacement. The focus of this dissertation was development of methods utilizing multiple ultrasound frequencies simultaneously to produce multifrequency harmonic radiation force and development of harmonic motion detection techniques adapted for vibrometry with simultaneous radiation force excitation. Beamforming for multifrequency radiation force was developed analytically, numerically, and experimentally. Sample applications in the fields of vibrometry and vibro-acoustic breast imaging are shown. Numerical models for harmonic motion detection of a vibrating reflective target and a vibrating scattering medium were developed. Parametric analyses were performed to extensively analyze the error in the vibration amplitude and phase measurements. An experimental method using multifrequency radiation force and harmonic motion detection with a single transducer is described and experimental results for a reflective target and scattering medium are shown. This work produced techniques for excitation and measurement of small amplitude harmonic motion for performing vibro-acoustography and vibrometry in a multiplexed manner. Models of vibration responses were developed to further understanding of current and future applications in vibro-acoustography and vibrometry.

Active Video Surveillance Based on Stereo and Infrared Imaging

Video-surveillance is a very actual and critical issue at the present time. Within this topic we address the problem of firstly identifying moving people in a scene through motion detection techniques, and subsequently categorising them in order to identify humans for tracking their movements. The use of stereo cameras, coupled with infrared vision, allows to apply this technique to images acquired through different and variable condition, and allows an a priori filtering based on the characteristics of such images to give evidence to objects emitting an higher radiance (i.e. higher temperature).

Motion detection and object tracking with discrete leaky integrate-and-fire neurons

A biologically inspired visual system capable of motion detection and pursuit motion is implemented using a Discrete Leaky Integrate-and-Fire (DLIF) neuron model. The system consists of a visual world, a virtual retina, the neural network circuitry (DLIF) to process the information, and a set of virtual eye muscles that serve to move the input area (visual field) of the retina within the visual world. Temporal aspects of the DLIF model are heavily exploited including: spike propagation latency, relative spike timing, and leaky potential integration. A novel technique for motion detection is employed utilizing coincidence detection aspects of the DLIF and relative spike timing. The system as a whole encodes information using relative spike timing of individual action potentials as well as rate coded spike trains. Experimental results are presented in which the motion of objects is detected and tracked in real and animated video. Pursuit motion is successful using linear and also sinusoidal paths which include object velocity changes. The visual system exhibits dynamic overshoot correction heavily exploiting neural network characteristics. System performance is within the bounds of real-time applications.

Advances in Coronary MRA from Vessel Wall to Whole Heart Imaging

Since its introduction, magnetic resonance (MR) imaging has undergone continued technical and methodological development and found numerous practical clinical applications. Cardiac MR imaging is one of the more sophisticated applications of MR, owing to the inherent presence of flow and motion and specific anatomy. Among the different categories of cardiac MR imaging, coronary MR angiography (MRA) places particularly high demands on planning, spatial resolution, high signal-to-noise ratio (SNR), and precise cardiac and respiratory motion correction. However, recent advances in hardware, MR sequences, and motion detection techniques have made it possible to perform coronary MRA that includes volumetric acquisition of the entire heart as well as imaging of the vessel walls on a submillimeter scale within a clinically acceptable scan time. We discuss from a technical perspective some of the milestones leading to the current state of coronary MR imaging and outline recent developments that will further advance coronary MR imaging. We discuss planning procedure, contrast preparation mechanisms and MR sequences, motion correction, high-resolution coronary artery and vessel wall imaging, and fast volumetric scanning techniques. Although MR imaging has certain limitations in providing simultaneous speed, resolution, and high SNR, it nonetheless offers a dedicated scanning procedure that addresses most clinically relevant questions in the diagnosis of ischemic heart disease.

An adaptive real-time skin detector based on Hue thresholding: A comparison on two motion tracking methods

Various applications like face and hand tracking and image retrieval have made skin detection an important area of research. However, currently available algorithms are based on static features of the skin colour, or require a significant amount of computation. On the other hand, skin detection algorithms are not robust to deal with real-world conditions, like background noise, change of intensity and lighting effects. This situation can be improved by using dynamic features of the skin colour in a sequence of images. This article proposes a skin detection algorithm based on adaptive Hue thresholding and its evaluation using two motion detection technique. The skin classifier is based on the Hue histogram of skin pixels, and adapts itself to the colour of the skin of the persons in the video sequence. This algorithm has demonstrated improvement in comparison to the static skin detection method.

Generalized self‐navigated motion detection technique: Preliminary investigation in abdominal imaging

Patient motion remains a primary obstacle to diagnostic image quality, especially in the abdomen, despite the existence of various motion artifact reduction techniques. This work presents a self-navigated motion detection method that can be generalized for most pulse sequences and k-space trajectories. Motion information is extracted directly from raw MR data, requiring no additional gradient or RF pulses, no physiologic monitoring equipment, and minimal--if any--impact on scan time. Initial feasibility results with a two-dimensional gradient echo sequence are shown in phantom studies and in vivo volunteer abdominal studies, demonstrating the sensitivity of the method to both respiratory motion and cardiovascular pulsatility. Prospectively gated images were acquired using the self-navigated data to synchronize image acquisition with motion. These preliminary results suggest that the self-navigated method is a promising technique for reducing motion artifacts in clinical abdominal and cardiac applications.

Generalized mutual information tests applied to fMRI analysis

To analyze functional magnetic resonance imaging (fMRI) data sets, the physiologically induced signals have to be separated from noise or from artifacts as a result of involuntary patient motion or MRI detection techniques. In the past years various methods of analysis of fMRI data were proposed, however one of the unsolved problems is related to dependence of the results with the threshold applied within each analysis. An alternative fMRI time series analysis method based on information theory and Tsallis entropy is proposed to decrease this dependency. The method applies the generalized mutual information (GMI) concept. Within the fMRI framework the GMI between the hemodynamic response function of a given voxel and the paradigm slope is evaluated as a measure of the amount of information that the paradigm could be encoded by a given brain region or voxel. This value represented the activation of the voxels. However, the analysis relies heavily on the Tsallis q parameter, allowing for a broad spectrum of results. As a test of applicability of the method real fMRI data were obtained in a simple motor paradigm and subsequently analyzed with GMI for several values of a q as well as with classical methods (cross-correlation and Student's t -test). The simulated data were used to compare quantitatively the presented method with the classical ones by means of the construction of receiver operating characteristic (ROC) curve. The results showed the applicability of the GMI method as well as the range of optimum q -values.