Showing papers on "Distance transform published in 2023"
07 Jul 2023
TL;DR: In this paper , the authors proposed a novel Hausdorff distance loss function for image segmentation, particularly in the field of medical imaging, based on the Felzenszwalb distance transform algorithm.
Abstract: <p>This paper presents a novel Hausdorff distance loss function for image segmentation, particularly in the field of medical imaging. The proposed Hausdorff distance loss function, based on the Felzenszwalb distance transform algorithm, addresses the computational complexity associated with previous Hausdorff distance loss function implementations, bringing it closer to the efficiency of the Dice loss function. The new method significantly reduces the computational time, making it only 11.2% slower than the Dice loss function, compared to the 125.9% slower rate of previous implementations. Furthermore, the proposed Hausdorff distance loss function improves the Dice similarity coefficient from 0.918 to 0.925 and reduces the Hausdorff distance from 4.583 to 3.464, demonstrating enhanced segmentation accuracy. The study's findings suggest that the proposed Hausdorff distance loss function can be a valuable tool for medical image segmentation, providing a balance between computational efficiency and segmentation precision. The code for the new Hausdorff distance loss function is publicly available for use.</p>
TL;DR: In this article , the Euclidean distance transform on non-binary images has been proposed to propagate distance information according to a predetermined pattern, which does not work well for binary images.
Abstract: The problem of distance mapping is a thoroughly studied topic in the field of image processing. Much focus has been spent on perfecting algorithms, calculating the Euclidean Distance Transform on binary images, in order to achieve the highest accuracy as efficiently as possible. Less focus has been spent on perfecting algorithms calculating the Euclidean Distance Transform on non-binary images where a much higher degree of precision is made possible by increasing the granularity of the input image. This study focuses on taking three different established methods for calculating distance mapping on binary images and applying them to non-binary images. We have found that the simpler algorithms, which simply propagate distance information according to a predetermined pattern work well with our new transform. The more analytical and in the case of the binary image, exact method, however, does not work as well with our new transform. Its way of propagating distance information is making too strict assumptions about the geometry of an image, which hold true for binary images, but not for non-binary images.
TL;DR: In this article , the thickness of individual cell walls of closed-cell foams in micro-CT images was measured using a distance transform on CT images to obtain thickness information of cell walls, a watershed transform on the distance matrix to locate the midlines of cells, identifying the intersections of midlines, and extracting the distance values of the pixels on the midline (or midplanes) of cells.
Abstract: Characterising the microstructure of foams is an important task for improving foam manufacturing processes and building foam numerical models. This study proposed a method for measuring the thickness of individual cell walls of closed‐cell foams in micro‐CT images. It comprises a distance transform on CT images to obtain thickness information of cell walls, a watershed transform on the distance matrix to locate the midlines of cell walls, identifying the intersections of midlines of cell walls by examining how many regions each pixel on the midlines of cell walls connects with, disconnecting and numbering the midlines of cell walls, extracting the distance values of the pixels on the midlines (or midplanes) of cell walls, and calculating the thickness of individual cell walls by multiplying the extracted distance values by two. Using this method, the thickness of cell walls of a polymeric closed‐cell foam was measured. It was found that cell wall thickness measured in 2D images shows larger average values (around 1.5 times) and dispersion compared to that measured in volumetric images.
07 Jul 2023
TL;DR: In this paper , the authors proposed a novel Hausdorff distance loss function for image segmentation, particularly in the field of medical imaging, based on the Felzenszwalb distance transform algorithm.
Abstract: <p>This paper presents a novel Hausdorff distance loss function for image segmentation, particularly in the field of medical imaging. The proposed Hausdorff distance loss function, based on the Felzenszwalb distance transform algorithm, addresses the computational complexity associated with previous Hausdorff distance loss function implementations, bringing it closer to the efficiency of the Dice loss function. The new method significantly reduces the computational time, making it only 11.2% slower than the Dice loss function, compared to the 125.9% slower rate of previous implementations. Furthermore, the proposed Hausdorff distance loss function improves the Dice similarity coefficient from 0.918 to 0.925 and reduces the Hausdorff distance from 4.583 to 3.464, demonstrating enhanced segmentation accuracy. The study's findings suggest that the proposed Hausdorff distance loss function can be a valuable tool for medical image segmentation, providing a balance between computational efficiency and segmentation precision. The code for the new Hausdorff distance loss function is publicly available for use.</p>
TL;DR: The medial axis transform (MAT) as discussed by the authors is an important geometric model description tool that provides a simplified representation of complex geometric shapes while ensuring accurate descriptions of geometric shape and topology, which can meet the requirements of many modern research fields, including geometric modeling, pattern recognition, model segmentation, model deformation, physical simulation, path planning, and more.
Abstract: Geometric shape representation algorithms are key technologies in the fields of computer graphics and geometric modeling. The Medial Axis Transform (MAT) is an important geometric model description tool that provides a simplified representation of complex geometric shapes while ensuring accurate descriptions of geometric shape and topology. Therefore, it can meet the requirements of many modern research fields, including geometric modeling, pattern recognition, model segmentation, model deformation, physical simulation, path planning, and more. This paper first introduces the basic concept of the medial axis transform, including the definition of the medial axis transform and the concept of medial axis primitives. It then describes the extraction algorithms for the medial axis transform, specific research on the medial axis transform in computer vision and computer graphics, potential applications of the medial axis transform, and medial axis transform datasets. Finally, the disadvantages and advantages of the medial axis transform are discussed, and some suggestions on possible future research directions are presented.