This article describes our extended and generalized approach to detection of periodic signals in image sequences. These signals appear in a small number of pixels of an image sequence as periodic fluctuations in the temporal domain. Neither the shape of a signal, nor its fundamental frequency is assumed to be known, but the fundamental frequency is assumed to be localized in some narrow range. The frame sequences cover only a few periods of each signal under discussion. We consider groups of these signals relative to our sampling operator, that is defined by its sampling frequency and integration (exposure) time. For each group the appropriate coherent basis is used: Fourier basis or periodized gaussians. Not unusually, under the sampling operator the signals and basic functions loose periodicity, and the bases loose orthogonality. The problems that arise are treated by some version of matching pursuit. Our approach to signal accumulation from adjacent pixels by spectrum-specific version of principal components is generalized by using projection onto more general class of subspaces. Normally, the computationally expensive processing sketched above is performed for less than 1% of pixels only. The remaining 99% are rejected by simple and fast procedures. The algorithm was tested by processing simulated image sequences, as well as several real ones.
KEYWORDS: Signal detection, Image processing, Motion models, Data modeling, Detection and tracking algorithms, Distortion, Image analysis, Statistical analysis, Video, Digital signal processing
An algorithm is reported for estimation and suppression of small vibration effects in image sequences. Such effects, even of sub-pixel magnitude, may critically degrade power spectrum of temporal-domain signals. The algorithm consists of the following steps: (1) We perform preliminary detection of the presence of vibration and localize its fundamental frequency by estimating and analyzing the two-dimensional signal, composed of micro-displacements caused by vibrations; (2) We approximate this two-dimensional signal by a two-dimensional periodic function, treating it basically the same way as periodic signals. This model depends on a small number of coefficients. These coefficients are determined by direct LS fitting of the data. (3) We eliminate the effects of the vibration using this model function, for each pixel separately. With this algorithm, several image sequences were processed. The vibration image motions were reconstructed with sub-pixel accuracy and were not, usually, reducible to one-dimensional sinusoidal motion. The algorithm appears to be useful for improving detection of periodic signals
in image sequences and reducing false alarms. This article continues our work on detection of periodic signals in image sequences.
Gamma-Ray Resonant Absorption (GRA) is an automatic-decision radiographic screening technique that combines high radiation penetration with very good sensitivity and specificity to nitrogenous explosives. The method is particularly well-suited to inspection of large, massive objects (since the resonant γ-ray probe is at 9.17 MeV) such as aviation and marine containers, heavy vehicles and railroad cars. Two kinds of γ-ray detectors have been employed to date in GRA systems: 1) Resonant-response nitrogen-rich liquid scintillators and 2) BGO detectors. This paper analyses and compares
the response of these detector-types to the resonant radiation, in terms of single-pixel figures of merit. The latter are sensitive not only to detector response, but also to accelerator-beam quality, via the properties of the nuclear reaction that produces the resonant-γ-rays. Generally, resonant detectors give rise to much higher nitrogen-contrast sensitivity in the radiographic image than their non-resonant detector counterparts and furthermore, do not require proton beams of high energy-resolution. By comparison, the non-resonant detectors have higher γ-detection efficiency, but their contrast sensitivity is very sensitive to the quality of the accelerator beam. Implications of these detector/accelerator
characteristics for eventual GRA field systems are discussed.
KEYWORDS: Signal detection, Francium, Data modeling, Image processing, Video, Signal to noise ratio, Electro optical modeling, Signal processing, Computer simulations, Analog electronics
The article describes a new, improved and fast version of our method and algorithm1 for detection of periodic signals in image sequences, i.e. signals that appear in a small number of adjacent pixels of an image sequence and are periodic in the temporal domain. The signal information is accumulated from adjacent pixels with the spectrum-specific version of Principal Components1. For this uniformly-sampled accumulated signal, a model dependent on few parameters is used for signal fitting. In this new version: 1) the sampling frequency may be below the Nyquist rate, and the model includes fold-over frequencies as well. 2) The general linear LS fit with pre-computed inverse matrixes was used for the model parameter estimation. It speeds-up the procedure. 3) The procedure is also speeded-up by preliminary pixel selection based on coarse estimation of the signal energy and SNR by the cross-power spectrum (CPS) method ith small data sub-frames. Our spectrum-specific covariance matrix estimate, employed in Spectrum-Specific Principal Components, is made more robust by utilizing the CPS method with small data sub-frames. The algorithm was tested by processing simulated image sequences as well as some real ones.
This article is a continuation of our work on signal detection in image sequences 1, 2. An algorithm has been developed for detection, localization and tracking of signals that appear in a small number of pixels of an image sequence and are periodic in the temporal domain. For such a signal, uniformly sampled, a model dependent on few parameters is derived and used for fitting a power spectrum and the signal itself. To enhance such a signal, a generalization of the Principal Component Method is proposed. A simulation based on the model is described. The pre-processing procedure is basically the same as before 1, 2. The algorithm was tested by processing simulated image sequences, as well as some real ones.
KEYWORDS: Signal detection, Detection and tracking algorithms, Algorithm development, Binary data, Image processing, Signal processing, Blob detection, Cameras, Statistical analysis, Signal to noise ratio
An algorithm developed for detection, localization and tracking of periodic signals, that appear as few-pixel blobs in image sequences and have a characteristic binary pattern in the temporal domain, is described. It is the further development of our original algorithm that extends its capabilities to the detection of moving targets and also improves its performance in some cases. We sketch those stages of the algorithm that were already described and discussed in. Then we describe in full detail the new features of our algorithm, namely: 1) A search for signal sources that move with a relatively-slowly-varying velocity. This velocity is adaptively estimated and re-estimated in order to maximize the correlation with the time-domain pattern. While the first estimate is performed at pixel level, the re-estimation (when tracking detected signals) is performed at blob level. 2) Re-estimation of the spatial domain background around the already-detected blob, yielding a more precise estimate of the blob. The new algorithm was tested by processing simulated, as well as real, image sequences. The results are discussed. The principal conclusion is that all good features of our original algorithm, namely: efficient detection of visible signals, reasonable detection of invisible signals, insensitivity to local motions in a video, to camera motion, to intensity changes and to any weak flickering of background, remain valid. However, we are required to increase slightly the minimal length of the time-domain correlation (thus, the delay in detection) at the same false alarm probability. Finally, we also consider the problem of the most suitable temporal-domain pattern for signals to be detected.
An algorithm for real-time detection, localization and tracking of periodic signals, which appear as few-pixel blobs in video image sequences and have a characteristic binary pattern in the temporal domain, was developed, implemented and tested. All stages of the algorithm are described and discussed: (1) initial estimation and continuous re-estimation of global motion; (2) local 'signal augmentation' (the crucial step for detecting small-sized signals in image sequences); (3) temporal domain -- followed by spatial domain -- background estimation and subtraction; (4) binarization; (5) preliminary temporal domain pattern matching and subsequent pixel tracking; (6) signal detection and localization by post- processing and clustering, applied to the set obtained in (5); (7) temporal and spatial tracking of the located signals. The algorithm was tested by processing simulated as well as real image sequences. The results are discussed. The algorithm enables for efficient detection of visible signals, provides for reasonable detection even of invisible signals and yields an acceptable false alarm probability. It has also proved to be insensitive to local motions in a video film, to camera motion, to intensity changes and to any weak flickering of background. Strong flickering of background can, however, decrease the probability of detection.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.