Image Segmentation by Student's-t Mixture Models Based on Markov Random Field and Weighted Mean Template

Xu, Peng; Zhu, Hongqing; Xie, Qunyi

doi:10.14257/ijsip.2016.9.2.27

Cited by 5 publications

(2 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Analyses conducted on simulated images, as well as on real clinical images, reveal an improvement in segmentation and accurate object boundary results. Pan et al [34] proposed a new mixture model, based on the Student's t-distribution [35] and the MRF. The main advantage of the Student's t-distribution is that it is heavily tailed, and hence provides a much more robust approach than the standard GMM.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Unsupervised Local Spatial Mixture Segmentation of Underwater Objects in Sonar Images

Abu

Diamant

2019

IEEE J. Oceanic Eng.

View full text Add to dashboard Cite

We focus on the segmentation of sonar images to achieve underwater object detection and classification. Our goal is to achieve accurate segmentation of the object's highlight and shadow regions. We target a robust solution that can manage different seabed backgrounds. Segmentation of sonar images is a challenging task. Speckle noise and intensity inhomogeneity may cause false detections, and complex seabed textures, like sand-ripples and sea-grass, often leading to false segmentation. In this paper,we propose our local spatial mixture (LSM) method for image segmentation of side-scan deployed sonar systems of any type. This new method estimates pixel labels in sonar images by incorporating the possible spatial correlation between neighboring pixels for improved segmentation. LSM modifies the expectation-maximization (EM) algorithm by adding an intermediate step (I-step) between the expectation (E-step) and maximization (M-step) steps. To combat intensity inhomogeneity, we employ a new initialization algorithm, one whose thresholds are set automatically to achieve and maintain robustness in various underwater environments. Using multiple evaluation indexes that measure the geometrical features of the segmented objects, we tested LSM using synthetic and real sonar images, one of which obtained from our own sea experiment. Our results show that LSM achieves improved segmentation performance over the state-of-the-art methods of four different approaches; LSM is also robust to different seabed textures and intensity inhomogeneity. We share the sonar images from our sea experiments.

show abstract

Section: Related Workmentioning

confidence: 99%

“…4) Quantitative Analysis: a) MCR and PR Analysis: We start by showing the performance in terms of MCR (33) and 1-PR, where PR is given in (34). The MCR and PR measures are indicators of the image segmentation's accuracy.…”

Section: ) Parameters Study Of γ and δmentioning

confidence: 99%

Unsupervised Local Spatial Mixture Segmentation of Underwater Objects in Sonar Images

Abu

Diamant

2019

IEEE J. Oceanic Eng.

View full text Add to dashboard Cite

show abstract

Mixture model for image segmentation using Gaussian, Student’s t, and Laplacian distribution with spatial dependence

Iriawan¹,

Pravitasari²,

Irhamah³

et al. 2019

AIP Conference Proceedings

View full text Add to dashboard Cite

Bayesian MSTBurr mixture model in the construction of 3D-MRI brain tumor images

Pravitasari

Iriawan

Fithriasari

et al. 2021

J. Phys.: Conf. Ser.

View full text Add to dashboard Cite

Detection of a brain tumor could be done with the serial of MRI images. The location and size of the tumor should be determined by viewing the 2D images individually. This kind of analysis is inefficient and error-prone. For better visualization, this study reconstructs a 3D structure from 2D MRI images. In recognizing the brain tumors, image segmentation is performed using the clustering analysis via Bayesian MSTBurr Mixture Model. The optimum cluster is selected by calculating the Correct Classification Ratio. The segmentation results for each image slice are performed in 3D rendering with the Matlab Volume Viewer. This study succeeded in creating a 3D model with a segmentation accuracy of 93.66% and an estimation of the tumor volume of about 33,556 mm3.

show abstract

Image Segmentation by Student's-t Mixture Models Based on Markov Random Field and Weighted Mean Template

Cited by 5 publications

References 8 publications

Unsupervised Local Spatial Mixture Segmentation of Underwater Objects in Sonar Images

Unsupervised Local Spatial Mixture Segmentation of Underwater Objects in Sonar Images

Mixture model for image segmentation using Gaussian, Student’s t, and Laplacian distribution with spatial dependence

Bayesian MSTBurr mixture model in the construction of 3D-MRI brain tumor images

Contact Info

Product

Resources

About