Abstract
Thresholding is an efficient step to extract quantitative information since the potential artefacts are often introduced by the point-spread effect of tomographic imaging. The thresholding value was previously selected only relying on engineering experience or histogram of tomographic image, which often presents a great challenge to determine an accurate thresholding value for various applications. As the tomographic image features often do not provide sufficient information to choose the best thresholding value, the information implicit in the measured boundary data is introduced into the thresholding process in this paper. A projection error, the relative difference between the computed boundary data of current segmentation and the measured boundary data, is proposed as a quantitative measure of such image segmentation quality. Then, a new multistep image segmentation process, called size projection algorithm (SPA), is proposed to automatically determine an optimal thresholding value by minimising the projection error. Results of simulation and experiment demonstrate the improved performance of the SPA-based tomographic image segmentation. An application of size projection algorithm for gas-water two-phase flow visualisation is also reported in this paper.
Highlights
Tomography is a highly promising technique for providing 2D/3D internal configurations of physical objects utilising the feature of “seeing through” [1]
For a large object, the electrical impedance tomography (EIT) image was reconstructed by the sensitivity back-projection (SBP) algorithm, where only the object shape and location are revealed and the object size is hardly determined. e empirical value method and the histogram-based method (Otsu’s method) extract but underestimate the object in the segmented image, while the size projection algorithm (SPA) method gives a better estimation of the extracted object with relative image error of 5.21%. e jagged boundary in the SPA result is a product of pixel error of image discretization as the setup cannot be represented by a series of pixels
Focusing on the thresholding value selection in tomographic image postprocessing, a multistep image segmentation process, size projection algorithm (SPA), is proposed to extract the reconstructed objects in EIT images. is method determines an optimal thresholding value for EIT image segmentation by comparing the computed boundary voltages corresponding to each segmentation image with the measured boundary voltages
Summary
Tomography is a highly promising technique for providing 2D/3D internal configurations of physical objects utilising the feature of “seeing through” [1]. To the authors’ knowledge, the previous thresholding value selection methods only trust the information of tomographic reconstruction, while abandoning the information implicit in the measured boundary data. These tomographic images suffer from serious reconstruction artefacts. Is yields a quantitative measure of such image segmentation quality, that is, the relative difference between the segmented image and actual image of objects From this point, a new multistep image segmentation process, called size projection algorithm (SPA), is proposed for automatically determining an optimal thresholding value by minimising the projection error.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
