Markov random field modeled range image segmentation

Wang, Xiao; Wang, Han

doi:10.1016/j.patrec.2003.10.017

Cited by 27 publications

(11 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The improved Laplacian operators reduce noise in range images and have a higher segmentation rate than the traditional Laplacian operator. Xiao and Han [5] proposed a range image segmentation based on Bayes inference and Markov random field modelling. The authors introduced the surface function parameters to group distance pixels into planar regions.…”

Section: Related Workmentioning

confidence: 99%

Scene Segmentation and Pedestrian Classification from 3-D Range and Intensity Images

Xue

Phung

Bouzerdoum

2012

2012 IEEE International Conference on Multimedia and Expo

View full text Add to dashboard Cite

. (2012). Scene segmentation and pedestrian classification from 3-D range and intensity images. IEEE International Conference on . Australia: IEEE. Scene segmentation and pedestrian classification from 3-D range and intensity images AbstractThis paper proposes a new approach to classify obstacles using a time-of-flight camera, for applications in assistive navigation of the visually impaired. Combining range and intensity images enables fast and accurate object segmentation, and provides useful navigation cues such as distances to the nearby obstacles and obstacle types. In the proposed approach, a 3-D range image is first segmented using histogram thresholding and mean-shift grouping. Then Fourier and GIST descriptors are applied on each segmented object to extract shape and texture features. Finally, support vector machines are used to recognize the obstacles. This paper focuses on classifying pedestrian and non-pedestrian obstacles. Evaluated on an image data set acquired using a time-of-flight camera, the proposed approach achieves a classification rate of 99.5%. ABSTRACTThis paper proposes a new approach to classify obstacles using a time-of-flight camera, for applications in assistive navigation of the visually impaired. Combining range and intensity images enables fast and accurate object segmentation, and provides useful navigation cues such as distances to the nearby obstacles and obstacle types. In the proposed approach, a 3-D range image is first segmented using histogram thresholding and mean-shift grouping. Then Fourier and GIST descriptors are applied on each segmented object to extract shape and texture features. Finally, support vector machines are used to recognize the obstacles. This paper focuses on classifying pedestrian and non-pedestrian obstacles. Evaluated on an image data set acquired using a time-of-flight camera, the proposed approach achieves a classification rate of 99.5%.

show abstract

Section: Related Workmentioning

confidence: 99%

Scene Segmentation and Pedestrian Classification from 3-D Range and Intensity Images

Xue

Phung

Bouzerdoum

2012

2012 IEEE International Conference on Multimedia and Expo

View full text Add to dashboard Cite

show abstract

“…They can either be parametric or nonparametric. Both are extensively used in segmentation of brain MR images such as Markov random models (MRFs) and Bayesian network classifiers [6,7].…”

Section: Introductionmentioning

confidence: 99%

Fast FCM with Spatial Neighborhood Information for Brain Mr Image Segmentation

Biniaz

Abbasi

2013

Journal of Artificial Intelligence and Soft Computing Research

View full text Add to dashboard Cite

Among different segmentation approaches Fuzzy c-Means clustering (FCM) is a welldeveloped algorithm for medical image segmentation. In emergency medical applications quick convergence of FCM is necessary. On the other hand spatial information is seldom exploited in standard FCM; therefore nuisance factors can simply affect it and cause misclassification. This paper aims to introduce a Fast FCM (FFCM) technique by incorporation of spatial neighborhood information which is exploited by a linear function on fuzzy membership. Applying proposed spatial Fast FCM (sFFCM), elapsed time is decreased and neighborhood spatial information is exploited in FFCM. Moreover, iteration numbers by proposed FFCM/sFFCM techniques are decreased efficiently. The FCM/FFCM techniques are examined on both simulated and real MR images. Furthermore, to considerably decrease of convergence time and iterations number, cluster centroids are initialized by an algorithm. Accuracy of the new approach is same as standard FCM. The quantitative assessments of presented FCM/FFCM techniques are evaluated by conventional validity functions. Experimental results demonstrate that sFFCM techniques efficiently handle noise interference and significantly decrease elapsed time.

show abstract

“…Wang and Wang [12] have presented a hybrid scheme for range image segmentation. First, they proposed a joint Bayesian estimation of both pixel labels and surface patches.…”

Section: Introductionmentioning

confidence: 99%

“…Moreover, the proposed methods proceed by assigning pixels to clusters without ensuring the continuity of the resulting clusters. Typically, in the approach proposed by Wang and Wang [12] , the pixels belonging to coplanar regions may be assigned equally to any of these regions. The spatial continuity constraint of resulting regions seems that it was not taken into account.…”

Section: Introductionmentioning

confidence: 99%

Range Image Segmentation by Randomized Region Growing and Bayesian Edge Regularization

Mazouzi¹,

Batouche²

2007

J. of Computer Science

View full text Add to dashboard Cite

Abstract:We presented and evaluated a new Bayesian method for range image segmentation. The method proceeds in to stages. First, an initial segmentation was produced by a randomized region growing technique. The produced segmentation was considered as a degraded version of the ideal segmentation, which should be then refined. In the second stage, image pixels not labeled in the first stage were labeled by using a Bayesian estimation, based on some prior assumptions on the regions in the image. Image priors were modeled by a new Markov Random Field (MRF) model. Contrary to most of the authors in range image segmentation, who use only surface smoothness MRF models, our MRF model takes into account also the smoothness of region boundaries. Tests performed with real images from the ABW database show the great potential of the proposed method for significantly improving the segmentation results.

show abstract

Markov random field modeled range image segmentation

Cited by 27 publications

References 18 publications

Scene Segmentation and Pedestrian Classification from 3-D Range and Intensity Images

Scene Segmentation and Pedestrian Classification from 3-D Range and Intensity Images

Fast FCM with Spatial Neighborhood Information for Brain Mr Image Segmentation

Range Image Segmentation by Randomized Region Growing and Bayesian Edge Regularization

Contact Info

Product

Resources

About