Comparison of image segmentation methods in simulated 2D and 3D microtomographic images of soil aggregates

Advances in X-ray microtomography (μCT) are opening new opportunities for examining soil pore structures. However, usefulness of μCT data for pore structure characterization depends on how accurately the grayscale images are segmented into pore and solid components. Multiple segmentation algorithms have been developed; however, one of the difficulties in comparing the accuracy of segmentation algorithms is the lack of ground-truth information in the soil samples subjected to μCT. This means that only the criteria that do not depend on the availability of the ground-truth data can be used in assessing accuracy of the segmentation methods, yet the reliability of such criteria in soil images is unclear. In this study, we simulated 2D and 3D soil images to resolve the problem of the lack of ground-truth information. The objectives of the study were (i) to explore optimal parameter selection for indicator kriging (IK) segmentation; (ii) to compare the accuracy of several commonly used segmentation methods, namely, entropy based method, iterative method, Otsu's method, and IK method; and (iii) to evaluate performance of the region non-uniformity measure (NU), the criterion that does not depend on presence of the ground-truth image, in segmentation method selection for soil images. We found that though there was no single segmentation method that preserved pore characteristics in all the cases, IK method yielded segmented images most similar to the ground-truth in most of the cases when the histogram of image grayscale values had clearly distinguishable peaks. For the image with poorly distinguishable histogram peaks, IK did not perform well, while Otsu's method produced acceptable segmentation results. The results indicated that selecting the segmentation method based on NU did not always produce optimal representation of pore characteristics. However, overall, the NU was found to be an acceptable criterion for segmentation method selection in μCT soil images.

[1]  Rainer Horn,et al.  Three-dimensional quantification of intra-aggregate pore-space features using synchrotron-radiation-based microtomography , 2008 .

[2]  M. Tuller,et al.  Segmentation of X‐ray computed tomography images of porous materials: A crucial step for characterization and quantitative analysis of pore structures , 2009 .

[3]  P. Monestiez,et al.  3D skeleton reconstructions of natural earthworm burrow systems using CAT scan images of soil cores , 1998, Biology and Fertility of Soils.

[4]  Anders Kaestner,et al.  Imaging and image processing in porous media research , 2008 .

[5]  Paola Campadelli,et al.  Quantitative evaluation of color image segmentation results , 1998, Pattern Recognit. Lett..

[6]  Bülent Sankur,et al.  Survey over image thresholding techniques and quantitative performance evaluation , 2004, J. Electronic Imaging.

[7]  Y. J. Zhang,et al.  A survey on evaluation methods for image segmentation , 1996, Pattern Recognit..

[8]  N. Otsu.  A threshold selection method from gray level histograms , 1979 .

[9]  Rafael C. González,et al.  Digital image processing using MATLAB , 2006 .

[10]  Donald J. Merchant,et al.  Cell cultures for virus vaccine production : held November 6-8, 1967, Clinical Center, National Institutes of Health, Bethesda, Maryland , 1968 .

[11]  Yu Jin Zhang,et al.  A review of recent evaluation methods for image segmentation , 2001, Proceedings of the Sixth International Symposium on Signal Processing and its Applications (Cat.No.01EX467).

[12]  Peter J. Eng,et al.  Geoscience applications of x-ray computed microtomography , 1999, Optics & Photonics.

[13]  Sacha J. Mooney,et al.  A reliable method for preserving soil structure in the field for subsequent morphological examinations , 2006 .

[14]  Markus Tuller,et al.  Application of Segmentation for Correction of Intensity Bias in X‐Ray Computed Tomography Images , 2010 .

[15]  W. Brent Lindquist,et al.  Image Thresholding by Indicator Kriging , 1999, IEEE Trans. Pattern Anal. Mach. Intell..

[16]  Hans-Jörg Vogel,et al.  Quantification of soil structure based on Minkowski functions , 2010, Comput. Geosci..

[17]  Jason I. Gerhard,et al.  Measurement and prediction of the relationship between capillary pressure, saturation, and interfacial area in a NAPL‐water‐glass bead system , 2010 .

[18]  R. Ketcham,et al.  Acquisition, optimization and interpretation of X-ray computed tomographic imagery: applications to the geosciences , 2001 .

[19]  Sacha J. Mooney,et al.  Quantification of soil structural changes induced by cereal anchorage failure: Image analysis of thin sections , 2007 .

[20]  Hans-Jörg Vogel,et al.  Topological characterization of pore space in soil — sample preparation and digital image-processing , 1996 .

[21]  Clayton V. Deutsch,et al.  Geostatistical Software Library and User's Guide , 1998 .

[22]  George Christakos,et al.  Random Field Models in Earth Sciences , 1992 .

[23]  Sankar K. Pal,et al.  A review on image segmentation techniques , 1993, Pattern Recognit..

[24]  Martin D. Levine,et al.  Dynamic Measurement of Computer Generated Image Segmentations , 1985, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[25]  T. W. Ridler,et al.  Picture thresholding using an iterative selection method. , 1978 .

[26]  William A. Yasnoff,et al.  Error measures for scene segmentation , 1977, Pattern Recognit..

[27]  Andrew P. Whitmore,et al.  Combining Spatial Resolutions in the Multiscale Analysis of Soil Pore‐Size Distributions , 2009 .

[28]  Prasanna K. Sahoo,et al.  Threshold selection using Renyi's entropy , 1997, Pattern Recognit..

[29]  Manfred Krafczyk,et al.  Tomographical Imaging and Mathematical Description of Porous Media Used for the Prediction of Fluid Distribution , 2006 .

[30]  Ws. Rasband.  ImageJ, U.S. National Institutes of Health, Bethesda, Maryland, USA , 2011 .

[31]  D. Rubin,et al.  Maximum likelihood from incomplete data via the EM - algorithm plus discussions on the paper , 1977 .

[32]  Wei Wang,et al.  Observer-dependent variability of the thresholding step in the quantitative analysis of soil images and X-ray microtomography data , 2010 .

[33]  J. Nellesen,et al.  Non-invasive 3D analysis of local soil deformation under mechanical and hydraulic stresses by μCT and digital image correlation , 2010 .

[34]  Alvin J. M. Smucker,et al.  Saturated Hydraulic Conductivity and Porosity within Macroaggregates Modified by Tillage , 2005 .

[35]  W. Brent Lindquist.  Quantitative analysis of three-dimensional x-ray tomographic images , 2002, Optics + Photonics.

[36]  Andrew P. Whitmore,et al.  Fractal Analysis of Pore Roughness in Images of Soil Using the Slit Island Method , 2008 .

[37]  Yee-Hong Yang,et al.  Multiresolution Color Image Segmentation , 1994, IEEE Trans. Pattern Anal. Mach. Intell..

[38]  Hans-Jörg Vogel,et al.  Segmentation of X-ray microtomography images of soil using gradient masks , 2010, Comput. Geosci..

[39]  Bärbel Tiemeyer,et al.  Artificially Drained Catchments—From Monitoring Studies towards Management Approaches , 2010 .

[40]  Harold E. Garrett,et al.  Agroforestry and Grass Buffer Effects on Pore Characteristics Measured by High‐Resolution X‐ray Computed Tomography , 2008 .

[41]  H. Vogel,et al.  From local hydraulic properties to effective transport in soil , 2000 .

[42]  R. Heck,et al.  A comparison of 2D vs. 3D thresholding of X-ray CT imagery , 2007 .

[43]  P.K Sahoo,et al.  A survey of thresholding techniques , 1988, Comput. Vis. Graph. Image Process..

[44]  Martial Hebert,et al.  A Measure for Objective Evaluation of Image Segmentation Algorithms , 2005, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05) - Workshops.

[45]  Diego Andina,et al.  Quantifying a soil pore distribution from 3D images: Multifractal spectrum through wavelet approach , 2010 .