Automated Microwave Tomography (MWT) Image Segmentation: State-of-the-Art Implementation and Evaluation

Abstract

Inspired by the high performance in image-based medical analysis, this paper explores the use of advanced segmentation techniques for industrial Microwave Tomography (MWT). Our context is the visual analysis of moisture levels in porous foams undergoing microwave drying. We propose an automatic segmentation technique-MWT Segmentation based on K-means (MWTS-KM) and demonstrate its efficiency and accuracy for industrial use. MWTS-KM consists of three stages: image augmentation, grayscale conversion, and K-means implementation. To estimate the performance of this technique, we empirically benchmark its efficiency and accuracy against two well-established alternatives: Otsu and K-means. To elicit performance data, three metrics (Jaccard index, Dice coefficient and false positive) are used. Our results indicate that MWTS-KM outperforms the well-established Otsu and K-means, both in visually observable and objectively quantitative evaluation.

Affective Colormap Design for Accurate Visual Comprehension in Industrial Tomography

“…were liable to result in divergent outcomes in segmentation tasks [ 16 ]. The segmentation of each image was conducted by an automatic method MWTS-KM [ 28 , 29 ] containing three key steps: image augmentation, grayscale conversion and k-means implementation. The automatic nature of this method constructed the link between the colormaps and the segmentation task we carried out.…”

“…After initial observations ( Figure 4 ), we first inferred that parula , viridis , cool , hot and autumn schemes are capable of visualizing the blue parts on the source image in segmentation. However, a systematic quantitative evaluation was then followed by adopting a data-driven approach [ 8 , 29 ], where three indexes (Jaccard index, Dice coefficient and false positive, as illustrated in Equations (1)–(3), with Source denoting the source MWT image to be segmented while Segmentation represents the segmented image) were used to compare segmented and corresponding ground-truth images presented in the respective 11 colormaps. The explanation of these three indicators is narrated as follows [ 29 ]: …”

Supporting Visualization Analysis in Industrial Process Tomography by Using Augmented Reality—A Case Study of an Industrial Microwave Drying System

“…To comprehensively help people to understand the outputted results, as well as to increase the accountability regarding the visualization, we deployed an entirely automatic segmentation method to visualize the high moisture areas on the reconstructed MWT images. Following our previously published work [ 12 ], we harness the MWT Segmentation based on K-means (MWTS-KM) segmentation algorithm to highlight the desired high moisture areas of the post-dried polymer foams on the reconstructed MWT images engaged in our study. As shown in Figure 5 , we implement the segmentation on the corresponding nine MWT images originating from different timestamps of the drying process.…”

“…IPT underpins and facilitates the extraction of qualitative and quantitative data regarding the related industrial processes, which is usually visualized in various ways for people to understand its nature, measure the critical process characteristics, and implement process control in a complete feedback network [ 7 ]. Some typical representatives of IPT, such as microwave tomography (MWT) [ 8 , 9 , 10 , 11 , 12 , 13 , 14 , 15 ], electrical resistance tomography (ERT) [ 16 ], and electrical capacitance tomography (ECT) [ 17 ] are widely used for industrial purposes such as moisture detection [ 8 , 9 ], crack detection and powder flow in pipes, and flow pattern detection of granules. In our study, we concentrate on a unique industrial microwave drying process [ 10 ] which uses precise drying and heating equipment for polymer foams with the aid of MWT, as displayed in Figure 1 .…”

A Modified Exact Reconstruction Algorithm to Determine the Complex Permittivity Perturbations of a Cancer Affected Biological Target using Microwave Tomography Technique

“…The MTT can detect the presence of tumors based on certain cell properties and is considered to be one of the best among the existing techniques such as X-ray computed tomography, MR imaging (MRI), positron emission, ultrasound, C-ray imaging, etc. [3,4].…”