An HMM-based segmentation method for traffic monitoring movies

Kato, Jien; Watanabe, Toshio; Joga, S.; Rittscher, Jens; Blake, A.

doi:10.1109/tpami.2002.1033221

Cited by 86 publications

(49 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In [13], [14], an autoregressive process was used to model the pixel value distribution over time. In [15], a Hidden Markov Model (HMM) approach was adopted.…”

Section: Related Workmentioning

confidence: 99%

A texture-based method for modeling the background and detecting moving objects

Heikkilä

Pietikäinen

2006

IEEE Trans. Pattern Anal. Machine Intell.

1,171

636

View full text Add to dashboard Cite

This paper presents a novel and efficient texture-based method for modeling the background and detecting moving objects from a video sequence. Each pixel is modeled as a group of adaptive local binary pattern histograms that are calculated over a circular region around the pixel. The approach provides us with many advantages compared to the state-of-the-art. Experimental results clearly justify our model. Index TermsMotion, texture, background subtraction, local binary pattern.

show abstract

“…In [13], [14], an autoregressive process was used to model the pixel value distribution over time. In [15], a Hidden Markov Model (HMM) approach was adopted.…”

Section: Related Workmentioning

confidence: 99%

A texture-based method for modeling the background and detecting moving objects

Heikkilä

Pietikäinen

2006

IEEE Trans. Pattern Anal. Machine Intell.

1,171

636

View full text Add to dashboard Cite

show abstract

“…The first variation of B(ϕ(x)) (with respect to ϕ(x)) can be easily shown to be given by (7) Differentiating (3) and (4) with respect to ϕ(x), one obtains (8) and (9) where δ(·) is the delta function, and A − and A + are the areas of Ω − and Ω + given by ∫ Ω χ − (x) dx and ∫ Ω χ + (x) dx, respectively. By substituting (8) and (9) into (7) and combining the corresponding terms, one can arrive at (10) where (11) Assuming the same kernel K(z) is used for computing the last two terms in (11), i.e., K(z) = K − (z) = K + (z), the latter can be further simplified to the following form: (12) where (13) Finally, introducing an artificial time parameter t, the gradient flow of that minimizes (6) is given by (14) where the subscript denotes the corresponding partial derivative, and V(x) is defined as given by either (11) or (12).…”

Section: B Gradient Flowmentioning

confidence: 99%

“…The latter is usually distinguished on a semantic basis and assumed to be either an object or a background. Thus, for example, the object may be associated with a diseased organ in medical imaging [1], [2], an intruder in surveillance video [3], [4], a moving part of a machine in robotics [5], [6], a maneuvering vehicle in traffic control [7], [8], or a target in navigation and military applications [9], [10].…”

Section: Introductionmentioning

confidence: 99%

Image Segmentation Using Active Contours Driven by the Bhattacharyya Gradient Flow

Michailovich

Rathi

Tannenbaum

2007

IEEE Trans. on Image Process.

298

242

View full text Add to dashboard Cite

This paper addresses the problem of image segmentation by means of active contours, whose evolution is driven by the gradient flow derived from an energy functional that is based on the Bhattacharyya distance. In particular, given the values of a photometric variable (or of a set thereof), which is to be used for classifying the image pixels, the active contours are designed to converge to the shape that results in maximal discrepancy between the empirical distributions of the photometric variable inside and outside of the contours. The above discrepancy is measured by means of the Bhattacharyya distance that proves to be an extremely useful tool for solving the problem at hand. The proposed methodology can be viewed as a generalization of the segmentation methods, in which active contours maximize the difference between a finite number of empirical moments of the "inside" and "outside" distributions. Furthermore, it is shown that the proposed methodology is very versatile and flexible in the sense that it allows one to easily accommodate a diversity of the image features based on which the segmentation should be performed. As an additional contribution, a method for automatically adjusting the smoothness properties of the empirical distributions is proposed. Such a procedure is crucial in situations when the number of data samples (supporting a certain segmentation class) varies considerably in the course of the evolution of the active contour. In this case, the smoothness properties of the empirical distributions have to be properly adjusted to avoid either over-or underestimation artifacts. Finally, a number of relevant segmentation results are demonstrated and some further research directions are discussed.

show abstract

“…method moved time of light waving camouflage bootstrap foreground mean object day switch trees aperture Frame difference 0 (1) 1358 (12) 2565 (3) 6789 (16) 10070 (12) 2175 (4) 4354 (9) 3902 (8) Mean + threshold 0 (1) 2593 (15) 16232 (11) 3285 (13) 1832 (3) 3236 (9) 2818 (5) 4285 (9) Mixture of Gaussians 0 (1) 1028 (10) 15802 (8) 1664 (8) 3496 (6) 2091 (3) 2972 (6) 3865 (7) Block correlation 1200 (11) 1165 (11) 3802 (4) 3771 (15) 6670 (11) 2673 (8) 2402 (4) 3098 (5) Eigen-background 1065 (10) 895 (7) 1324 (2) 3084 (12) 1898 (4) 6433 (11) 2978 (7) 2525 (3) Toyama [1] 0 (1) 986 (8) 1322 (1) 2876 (11) 2935 (5) 2390 (6) 969 (1) 1640 (2) Maddalena [12] 453 (2) 293 (3) Wren [30] 654 (6) 2...…”

Section: Methodsmentioning

confidence: 99%

Background Subtraction with Dirichlet Processes

Haines

Xiang

2012

Computer Vision – ECCV 2012

View full text Add to dashboard Cite

Abstract. Background subtraction is an important first step for video analysis, where it is used to discover the objects of interest for further processing. Such an algorithm often consists of a background model and a regularisation scheme. The background model determines a perpixel measure of if a pixel belongs to the background or the foreground, whilst the regularisation brings in information from adjacent pixels. A new method is presented that uses a Dirichlet process Gaussian mixture model to estimate a per-pixel background distribution, which is followed by probabilistic regularisation. Key advantages include inferring the perpixel mode count, such that it accurately models dynamic backgrounds, and that it updates its model continuously in a principled way.

show abstract

An HMM-based segmentation method for traffic monitoring movies

Cited by 86 publications

References 16 publications

A texture-based method for modeling the background and detecting moving objects

A texture-based method for modeling the background and detecting moving objects

Image Segmentation Using Active Contours Driven by the Bhattacharyya Gradient Flow

Background Subtraction with Dirichlet Processes

Contact Info

Product

Resources

About