Introduction

Maragos, Petros; Schafer, Ronald W.; Butt, Muhammad Akmal

doi:10.1007/978-1-4613-0469-2_1

Cited by 5 publications

(6 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is one of the oldest and most successful areas of digital image processing; see e.g. the textbooks [10,15,21,27,28,29,34] and the proceedings volumes [14,16,20,24,30,31,36] for an overview. Its fundamental operations are called dilation and erosion.…”

Section: Introductionmentioning

confidence: 99%

A Shock-Capturing Algorithm for the Differential Equations of Dilation and Erosion

Breuß

Weickert

2006

J Math Imaging Vis

View full text Add to dashboard Cite

Dilation and erosion are the fundamental operations in morphological image processing. Algorithms that exploit the formulation of these processes in terms of partial differential equations offer advantages for non-digitally scalable structuring elements and allow sub-pixel accuracy. However, the widely-used schemes from the literature suffer from significant blurring at discontinuities. We address this problem by developing a novel, flux corrected transport (FCT) type algorithm for morphological dilation / erosion with a flat disc. It uses the viscosity form of an upwind scheme in order to quantify the undesired diffusive effects. In a subsequent corrector step we compensate for these artifacts by means of a stabilised inverse diffusion process that requires a specific nonlinear multidimensional formulation. We prove a discrete maximum-minimum principle in this multidimensional framework. Our experiments show that the method gives a very sharp resolution of moving fronts, and it approximates rotation invariance very well.

show abstract

Section: Introductionmentioning

confidence: 99%

A Shock-Capturing Algorithm for the Differential Equations of Dilation and Erosion

Breuß

Weickert

2006

J Math Imaging Vis

View full text Add to dashboard Cite

show abstract

“…The obtained regions for the kth humanlike object O k (y, z) are further smoothed by a morphological operation in order to erase small isolated regions and fill holes in certain regions [Maragos et al, 1996]. Then, we have the new segmentation image O * k (y, z).…”

Section: Human Segmentationmentioning

confidence: 99%

“…If the size of the head portion is less than h w h h /2, it is discarded. In order to extract the contour neighborhood of the kth human object, we dilate O * k (y, z) k s times as in [Maragos et al, 1996] and have its dilated version,Ô * k (y, z). The contour neighborhood of this human object is obtained by…”

Section: Human Verificationmentioning

confidence: 99%

Robust Human Detection and Identification by Using Stereo and Thermal Images in Human Robot Interaction

Guan

et al. 2007

Int. J. Inf. Acquisition

View full text Add to dashboard Cite

In this paper, robust human detection is investigated by fusing the stereo and infrared thermal images for effective interaction between humans and socially interactive robots. A scale-adaptive filter is first designed for the stereo vision system to detect human candidates. To eliminate the difficulty of the vision system in distinguishing human beings from human-like objects, the infrared thermal image is used to solve the ambiguity and reduce the illumination effect. Experimental results show that the fusion of these two types of images gives an improved vision system for robust human detection and identification, which is a most important and essential component of human robot interaction.

show abstract

“…Interest in mathematical morphology has grown enormously during the past decade and our understanding of the underlying mathematical structure has improved beyond measure. Nevertheless, it is only since 1993 that the existence of a transform that maps the basic morphological operation into a simpler "product", just as the Fourier transform maps the convolution product into a direct product, has been known [14][15][16][17][18][19]. Furthermore, this is no mere abstract development awaiting applications.…”

Section: Multi-resolution Transformsmentioning

confidence: 99%