Search

This paper presents a novel segmentation algorithm specially developed for applications in 3D point clouds with high variability and noise, particularly suitable for heritage building 3D data. The method can be categorized within the segmentation procedures based on edge detection. In addition, it uses a graph-based topological structure generated from the supervoxelization of the 3D point clouds, which is used to make the closure of the edge points and to define the different segments. The algorithm provides a valuable tool for generating results that can be used in subsequent classification tasks and broader computer applications dealing with 3D point clouds. One of the characteristics of this segmentation method is that it is unsupervised, which makes it particularly advantageous for heritage applications where labelled data is scarce. It is also easily adaptable to different edge point detection and supervoxelization algorithms. Finally, the results show that the 3D data can be segmented into different architectural elements, which is important for further classification or recognition. Extensive testing on real data from historic buildings demonstrated the effectiveness of the method. The results show superior performance compared to three other segmentation methods, both globally and in the segmentation of planar and curved zones of historic buildings. [ABSTRACT FROM AUTHOR]

This paper presents a novel segmentation algorithm specially developed for applications in 3D point clouds with high variability and noise, particularly suitable for heritage building 3D data. The method can be categorized within the segmentation procedures based on edge detection. In addition, it uses a graph-based topological structure generated from the supervoxelization of the 3D point clouds, which is used to make the closure of the edge points and to define the different segments. The algorithm provides a valuable tool for generating results that can be used in subsequent classification tasks and broader computer applications dealing with 3D point clouds. One of the characteristics of this segmentation method is that it is unsupervised, which makes it particularly advantageous for heritage applications where labelled data is scarce. It is also easily adaptable to different edge point detection and supervoxelization algorithms. Finally, the results show that the 3D data can be segmented into different architectural elements, which is important for further classification or recognition. Extensive testing on real data from historic buildings demonstrated the effectiveness of the method. The results show superior performance compared to three other segmentation methods, both globally and in the segmentation of planar and curved zones of historic buildings.

Reality capture technologies such as Structure-from-Motion (SfM) photogrammetry have become a state-of-the-art practice within landslide research workflows in recent years. Such technology has been predominantly utilized to provide detailed digital products in landslide assessment where often, for thorough mapping, significant accessibility restrictions must be overcome. UAV photogrammetry produces a set of multi-dimensional digital models to support landslide management, including orthomosaic, digital surface model (DSM), and 3D point cloud. At the same time, the recognition of objects depicted in images has become increasingly possible with the development of various methodologies. Among those, Geographic Object-Based Image Analysis (GEOBIA) has been established as a new paradigm in the geospatial data domain and has also recently found applications in landslide research. However, most of the landslide-related GEOBIA applications focus on large scales based on satellite imagery. In this work, we examine the potential of different UAV photogrammetry product combinations to be used as inputs to image segmentation techniques for the automated extraction of landslide elements at site-specific scales. Image segmentation is the core process within GEOBIA workflows. The objective of this work is to investigate the incorporation of fully 3D data into GEOBIA workflows for the delineation of landslide elements that are often challenging to be identified within typical rasterized models due to the steepness of the terrain. Here, we apply a common unsupervised image segmentation pipeline to 3D grids based on the superpixel/supervoxel and graph cut algorithms. The products of UAV photogrammetry for two landslide cases in Greece are combined and used as 2D (orthomosaic), 2.5D (orthomosaic + DSM), and 3D (point cloud) terrain representations in this research. We provide a detailed quantitative comparative analysis of the different models based on expert-based annotations of the landscapes and conclude that using fully 3D terrain representations as inputs to segmentation algorithms provides consistently better landslide segments. [ABSTRACT FROM AUTHOR]

Supervoxels find applications as a pre-processing step in many image processing problems due to their ability to present a regional representation of points by correlating them into a set of clusters. Besides reducing the overall computational time for subsequent algorithms, the desirable properties in supervoxels are adherence to object boundaries and compactness. Existing supervoxel segmentation methods define the size of a supervoxel based on a user inputted resolution value. A fixed resolution results in poor performance in point clouds with non-uniform density. Whereas, other methods, in their quest for better boundary adherence, produce supervoxels with irregular shapes and elongated boundaries. In this article, we propose a new supervoxel segmentation method, based on k-means algorithm, with dynamic cluster seed initialization to ensure uniform distribution of cluster seeds in point clouds with variable densities. We also propose a new cluster seed initialization strategy, based on histogram binning of surface normals, for better boundary adherence. Our algorithm is parameter-free and gives equal importance to the color, spatial location and orientation of the points resulting in compact supervoxels with tight boundaries. We test the efficacy of our algorithm on a publicly available point cloud dataset consisting of 1449 pairs of indoor RGB-D images, i.e., color (RGB) images coupled with depth information (D) mapped per pixel. Results are compared against three state-of-the-art algorithms based on four quality metrics. Results show that our method provides significant improvement over other methods in the undersegmentation error and compactness metrics and, performs equally well in the boundary recall and contour density metrics.

Reality capture technologies such as Structure-from-Motion (SfM) photogrammetry have become a state-of-the-art practice within landslide research workflows in recent years. Such technology has been predominantly utilized to provide detailed digital products in landslide assessment where often, for thorough mapping, significant accessibility restrictions must be overcome. UAV photogrammetry produces a set of multi-dimensional digital models to support landslide management, including orthomosaic, digital surface model (DSM), and 3D point cloud. At the same time, the recognition of objects depicted in images has become increasingly possible with the development of various methodologies. Among those, Geographic Object-Based Image Analysis (GEOBIA) has been established as a new paradigm in the geospatial data domain and has also recently found applications in landslide research. However, most of the landslide-related GEOBIA applications focus on large scales based on satellite imagery. In this work, we examine the potential of different UAV photogrammetry product combinations to be used as inputs to image segmentation techniques for the automated extraction of landslide elements at site-specific scales. Image segmentation is the core process within GEOBIA workflows. The objective of this work is to investigate the incorporation of fully 3D data into GEOBIA workflows for the delineation of landslide elements that are often challenging to be identified within typical rasterized models due to the steepness of the terrain. Here, we apply a common unsupervised image segmentation pipeline to 3D grids based on the superpixel/supervoxel and graph cut algorithms. The products of UAV photogrammetry for two landslide cases in Greece are combined and used as 2D (orthomosaic), 2.5D (orthomosaic + DSM), and 3D (point cloud) terrain representations in this research. We provide a detailed quantitative comparative analysis of the different models based on expert-based annotations of the landscapes and conclude that using fully 3D terrain representations as inputs to segmentation algorithms provides consistently better landslide segments.

In this paper, we propose a graph self-construction and fusion network (GSCFN) for semi-supervised brain tissue segmentation in Magnetic Resonance Imaging (MRI) by fusing multiple types of image features. Compared to the use of a single feature, various features bring complementary information and can contribute to a better graph representation with a great discriminative power increase. But to do so, two problems need to be solved. The first one consists in effectively inferring a graph from a Magnetic Resonance (MR) image so as to implicitly encode the segmentation information and the second in fully leveraging various features. To solve both problems, we propose an original brain MR image semi-supervised segmentation framework, called graph self-construction and fusion network. This one relies on two parts. In the first one, a graph self-construction network is utilized to obtain various graph representations of an MR image depending on different features. In the second, a multi-graph convolution network is proposed for the fusion of multiple graphs and features as well as for the classification of supervoxels which are treated as graph nodes. Experiments on the BrainWeb18 dataset and the Internet Brain Segmentation Repository 18 dataset validate the superiority of our scheme compared with approaches based on a single feature type, and some other state-of-the-art methods. The ablation experiment indicates that the proposed GSCFN can produce more accurate and reliable segmentation by seamlessly integrates multiple types of features. [ABSTRACT FROM AUTHOR]

Over-segmenting a video into supervoxels has strong potential to reduce the complexity of downstream computer vision applications. Content-sensitive supervoxels (CSSs) are typically smaller in content-dense regions (i.e., with high variation of appearance and/or motion) and larger in content-sparse regions. In this paper, we propose to compute feature-aware CSSs (FCSSs) that are regularly shaped 3D primitive volumes well aligned with local object/region/motion boundaries in video. To compute FCSSs, we map a video to a 3D manifold embedded in a combined color and spatiotemporal space, in which the volume elements of video manifold give a good measure of the video content density. Then any uniform tessellation on video manifold can induce CSS in the video. Our idea is that among all possible uniform tessellations on the video manifold, FCSS finds one whose cell boundaries well align with local video boundaries. To achieve this goal, we propose a novel restricted centroidal Voronoi tessellation method that simultaneously minimizes the tessellation energy (leading to uniform cells in the tessellation) and maximizes the average boundary distance (leading to good local feature alignment). Theoretically our method has an optimal competitive ratio O(1), and its time and space complexities are O(NK) and O(N + K) for computing K supervoxels in an N-voxel video. We also present a simple extension of FCSS to streaming FCSS for processing long videos that cannot be loaded into main memory at once. We evaluate FCSS, streaming FCSS and ten representative supervoxel methods on four video datasets and two novel video applications. The results show that our method simultaneously achieves state-of-the-art performance with respect to various evaluation criteria. [ABSTRACT FROM AUTHOR]

The issue of achieving an appropriate segmentation for indoor point cloud scenes remains difficult. Although available methods continue to improve the benchmark performance, more attentions need to be paid to deal with the drawbacks of inaccurate or incomplete segments in division. To push the research to the next level, this work proposes an learning-free algorithm for the segmentation of indoor point clouds which consists of two stages. The first stage extracts edges of RGBD point clouds and applies them in the voxel clustering process to avoid generating supervoxels which are situated across object boundaries. After this pre-segmentation, a two-phase merging procedure is presented in the second part. By conducting region growing on optimized supervoxels, a set of local regions is obtained. Then we propose to define the convexity-concavity of adjacent regions based on the observations of object structures and merge the convexly connected regions to achieve object-level segmentation. This algorithm is straightforward to implement and requires no training data. Experimental results show that it produces supervoxels with plausible boundaries and arrives at better object-level segmentation.

Point cloud feature extraction as a classification task is crucial in maximizing the efficient downstream applicability of raw point clouds. With the goal of learning optimum features for efficient classification of a multi-class point cloud for downstream applications, this letter presents a supervoxel-Fisher vector (FV)-based approach for airborne light detection and ranging (LiDAR) data classification. In our approach, FV encoding is implemented to deduce compact global descriptors from aggregated supervoxels to establish a more descriptive and discriminative representation, transforming the low-level visual features into high-level semantic features. As a result, the proposed approach combines local and global feature properties through the quantization and aggregation of higher order statistics to harnesses their combined advantages for producing good classification results. Experiments were conducted on the international society for photogrammetry and remote sensing 3-D semantic labeling benchmark data set. Results indicate that the proposed approach is robust and efficient, attained the third best position with an overall accuracy of 81.79%, and ranked first with an $\text{F}_{{1}}$ -score of 72.31%. [ABSTRACT FROM AUTHOR]

Reality capture technologies, also known as close-range sensing, have been increasingly popular within the field of engineering geology and particularly rock slope management. Such technologies provide accurate and high-resolution n-dimensional spatial representations of our physical world, known as 3D point clouds, that are mainly used for visualization and monitoring purposes. To extract knowledge from point clouds and inform decision-making within rock slope management systems, semantic injection through automated processes is necessary. In this paper, we propose a model that utilizes a segmentation procedure which delivers segments ready to classify and be retained or rejected according to complementary knowledge-based filter criteria. First, we provide relevant voxel-based features based on the local dimensionality, orientation, and topology and partition them in an assembly of homogenous segments. Subsequently, we build a decision tree that utilizes geometrical, topological, and contextual information and enables the classification of a multi-hazard railway rock slope section in British Columbia, Canada into classes involved in landslide risk management. Finally, the approach is compared to machine learning integrating recent featuring strategies for rock slope classification with limited training data (which is usually the case). This alternative to machine learning semantic segmentation approaches reduces substantially the model size and complexity and provides an adaptable framework for tailored decision-making systems leveraging rock slope semantics.

• A novel method for lung ventilation estimation using supervoxels and deformable image registration. • The method tracks intensity changes in supervoxels extracted from exhale phase 4DCT. • A correlation between the estimated ventilation maps with XeMRI ventilation images is calculated. • The results suggest that the presented technique may be advantageous for CT-based ventilation estimation, when compared with other methods. • Our method performs favorably independently of the applied image registration approach. Patch-based approaches have received substantial attention over the recent years in medical imaging. One of their potential applications may be to provide more anatomically consistent ventilation maps estimated on dynamic lung CT. An assessment of regional lung function may act as a guide for radiotherapy, ensuring a more accurate treatment plan. This in turn, could spare well-functioning parts of the lungs. We present a novel method for lung ventilation estimation from dynamic lung CT imaging, combining a supervoxel-based image representation with deformations estimated during deformable image registration, performed between peak breathing phases. For this we propose a method that tracks changes of the intensity of previously extracted supervoxels. For the evaluation of the method we calculate correlation of the estimated ventilation maps with static ventilation images acquired from hyperpolarized Xenon129 MRI. We also investigate the influence of different image registration methods used to estimate deformations between the peak breathing phases in the dynamic CT imaging. We show that our method performs favorably to other ventilation estimation methods commonly used in the field, independently of the image registration method applied to dynamic CT. Due to the patch-based approach of our method, it may be physiologically more consistent with lung anatomy than previous methods relying on voxel-wise relationships. In our method the ventilation is estimated for supervoxels, which tend to group spatially close voxels with similar intensity values. The proposed method was evaluated on a dataset consisting of three lung cancer patients undergoing radiotherapy treatment, and this resulted in a correlation of 0.485 with XeMRI ventilation images, compared with 0.393 for the intensity-based approach, 0.231 for the Jacobian-based method and 0.386 for the Hounsfield units averaging method, on average. Within the limitation of the small number of cases analyzed, results suggest that the presented technique may be advantageous for CT-based ventilation estimation. The results showing higher values of correlation of the proposed method demonstrate the potential of our method to more accurately mimic the lung physiology. [ABSTRACT FROM AUTHOR]

In the last decade, supervoxels have become a useful mid-level representation of volumetric medical images such as MRIs and CT scans. Several methods were suggested to produce uniform supervoxels, yet little has been done to generate content-sensitive over-segmentations. This is particularly beneficial to 3D medical image analysis, where sizes of anatomical structures vary largely. In this paper, we propose AdaSLIC as an adaptive supervoxel generation technique that applies to volumetric medical images. In small structures, it generates tiny supervoxels to capture the details of the image. Meanwhile, it partitions large structures into bigger supervoxels, hence leading to a sparse description. The proposed technique is an extension of the Simple Linear Iterative Clustering (SLIC) algorithm. Rather than using a regular sampling to initiate supervoxel centers, a content-sensitive initialization is performed using a Poisson-disk sampling algorithm (PDS). It relies on a map of distances to the main image contours. The size of each supervoxel depends on the distance of its center to the closest image contour. We compare our algorithm to the SLIC algorithm as well as to an extension of the DBSCAN algorithm (Density-Based Spatial Clustering of Applications with Noise). Two datasets are used for this purpose: knee MRIs and cardiovascular magnetic resonance (CMR) images. We use different metrics to assess the quality of the generated over-segmentations. Experimental results show that our algorithm achieves comparable or better boundary adherence than the state of the art algorithms while producing compact and adaptive supervoxels. [ABSTRACT FROM AUTHOR]

Vegetation classification is a fundamental task in several applications such as forest management, remote sensing-based crop monitoring, and mitigation of plant diseases, digital prototyping of plants, and plant phenotyping, among others. We propose a novel supervoxel-based methodology to accurately detect vegetation from small-footprint full-waveform airborne laser scanning data in urban and mountainous scenes. Mathematically, the full-waveform decomposition and fitting model based on multiple kernels is presented to generate high-density 3D point clouds and their relevant attributes. The homogeneous supervoxels are then generated by using an enhanced probability density clustering (PDC) algorithm. For each supervoxel, we employ latent Dirichlet allocation to obtain supervoxels features through generalisation of geometric and full-waveform features of point clouds. The Support Vector Machine (SVM) and ensemble classifier random forest (RF) are used to classify these supervoxels into vegetation and non-vegetation. Our experiments on urban and mountainous scenes demonstrate that our approach achieves an overall accuracy of 98.27% and 96.47% respectively by RF classifier and achieves an overall accuracy of 98.16% and 97.67% on the same data sets by SVM classifier. By integrating full-waveform information and more meaningful generalised features, our method outperforms state-of-the-art methods at preserving a trade-off between missing alarm rate and false alarm rate. [ABSTRACT FROM PUBLISHER]

We propose combining a supervoxel-based image representation with the concept of graph cuts as an efficient optimization technique for three-dimensional (3-D) deformable image registration. Due to the pixels/voxels- wise graph construction, the use of graph cuts in this context has been mainly limited to two-dimensional (2- D) applications. However, our work overcomes some of the previous limitations by posing the problem on a graph created by adjacent supervoxels, where the number of nodes in the graph is reduced from the number of voxels to the number of supervoxels. We demonstrate how a supervoxel image representation combined with graph cuts-based optimization can be applied to 3-D data. We further show that the application of a relaxed graph representation of the image, followed by guided image filtering over the estimated deformation field, allows us to model “sliding motion.” Applying this method to lung image registration results in highly accurate image registration and anatomically plausible estimations of the deformations. Evaluation of our method on a publicly available computed tomography lung image dataset leads to the observation that our approach compares very favorably with state of the art methods in continuous and discrete image registration, achieving target registration error of 1.16 mm on average per landmark. [ABSTRACT FROM AUTHOR]

This article presents a general method for estimating pairwise image correspondences, which is a fundamental problem in image analysis. The method consists of over-segmenting a pair of images into supervoxels. A forest classifier is then trained on one of the images, the source, by using supervoxel indices as voxel-wise class labels. Applying the forest on the other image, the target, yields a supervoxel labelling, which is then regularised using majority voting within the boundaries of the target's supervoxels. This yields semi-dense correspondences in a fully automatic, unsupervised, efficient and robust manner. The advantage of our approach is that no prior information or manual annotations are required, making it suitable as a general initialisation component for various medical imaging tasks that require coarse correspondences, such as atlas/patch-based segmentation, registration, and atlas construction. We demonstrate the effectiveness of our approach in two different applications: a) initialisation of longitudinal registration on spine CT data of 96 patients, and b) atlas-based image segmentation using 150 abdominal CT images. Comparison to state-of-the-art methods demonstrate the potential of supervoxel classification forests for estimating image correspondences. [ABSTRACT FROM AUTHOR]

Purpose : Toward an efficient clinical management of hepatocellular carcinoma (HCC), we propose a classification framework dedicated to tumor necrosis rate estimation from dynamic contrast-enhanced CT scans. Based on machine learning, it requires weak interaction efforts to segment healthy, active and necrotic liver tissues. Methods : Our contributions are two-fold. First, we apply random forest (RF) on supervoxels using multi-phase supervoxel-based features that discriminate tissues based on their dynamic in response to contrast agent injection. Second, we extend this technique in a hierarchical multi-scale fashion to deal with multiple spatial extents and appearance heterogeneity. It translates in an adaptive data sampling scheme combining RF and hierarchical multi-scale tree resulting from recursive supervoxel decomposition. By concatenating multi-phase features across the hierarchical multi-scale tree to describe leaf supervoxels, we enable RF to automatically infer the most informative scales without defining any explicit rules on how to combine them. Results : Assessment on clinical data confirms the benefits of multi-phase information embedded in a multi-scale supervoxel representation for HCC tumor segmentation. Conclusion : Dedicated but not limited only to HCC management, both contributions reach further steps toward more accurate multi-label tissue classification. [ABSTRACT FROM AUTHOR]

Alzheimer’s disease (AD) represents the prevalent type of dementia in the elderly, and is characterized by the presence of neurofibrillary tangles and amyloid plaques that eventually leads to the loss of neurons, resulting in atrophy in specific brain areas. Although the process of degeneration can be visualized through various modalities of medical imaging and has proved to be a valuable biomarker, the accurate diagnosis of Alzheimer’s disease remains a challenge, especially in its early stages. In this paper, we propose a novel classification method for Alzheimer’s disease/cognitive normal discrimination in structural magnetic resonance images (MRI), based on the extension of the concept of histons to volumetric images. The proposed method exploits the relationship between grey matter, white matter and cerebrospinal fluid degeneration by means of a segmentation using supervoxels. The calculated histons are then processed for a reduction in dimensionality using principal components analysis (PCA) and the resulting vector is used to train an support vector machine (SVM) classifier. Experimental results using the OASIS-1 database have proven to be a significant improvement compared to a baseline classification made using the pipeline provided by Clinica software.

Over the past 20 years, the field of medical image registration has significantly advanced from multi-modal image fusion to highly non-linear, deformable image registration for a wide range of medical applications and imaging modalities, involving the compensation and analysis of physiological organ motion or of tissue changes due to growth or disease patterns. While the original focus of image registration has predominantly been on correcting for rigid-body motion of brain image volumes acquired at different scanning sessions, often with different modalities, the advent of dedicated longitudinal and cross-sectional brain studies soon necessitated the development of more sophisticated methods that are able to detect and measure local structural or functional changes, or group differences. Moving outside of the brain, cine imaging and dynamic imaging required the development of deformable image registration to directly measure or compensate for local tissue motion. Since then, deformable image registration has become a general enabling technology. In this work we will present our own contributions to the state-of-the-art in deformable multi-modal fusion and complex motion modelling, and then discuss remaining challenges and provide future perspectives to the field. [ABSTRACT FROM AUTHOR]

Supervoxel segmentation has strong potential to be incorporated into early video analysis as superpixel segmentation has in image analysis. However, there are many plausible supervoxel methods and little understanding as to when and where each is most appropriate. Indeed, we are not aware of a single comparative study on supervoxel segmentation. To that end, we study seven supervoxel algorithms, including both off-line and streaming methods, in the context of what we consider to be a good supervoxel: namely, spatiotemporal uniformity, object/region boundary detection, region compression and parsimony. For the evaluation we propose a comprehensive suite of seven quality metrics to measure these desirable supervoxel characteristics. In addition, we evaluate the methods in a supervoxel classification task as a proxy for subsequent high-level uses of the supervoxels in video analysis. We use six existing benchmark video datasets with a variety of content-types and dense human annotations. Our findings have led us to conclusive evidence that the hierarchical graph-based (GBH), segmentation by weighted aggregation (SWA) and temporal superpixels (TSP) methods are the top-performers among the seven methods. They all perform well in terms of segmentation accuracy, but vary in regard to the other desiderata: GBH captures object boundaries best; SWA has the best potential for region compression; and TSP achieves the best undersegmentation error. [ABSTRACT FROM AUTHOR]

Efficient and accurate segmentation of cellular structures in microscopic data is an essential task in medical imaging. Many state-of-the-art approaches to image segmentation use structured models whose parameters must be carefully chosen for optimal performance. A popular choice is to learn them using a large-margin framework and more specifically structured support vector machines (SSVM). Although SSVMs are appealing, they suffer from certain limitations. First, they are restricted in practice to linear kernels because the more powerful nonlinear kernels cause the learning to become prohibitively expensive. Second, they require iteratively finding the most violated constraints, which is often intractable for the loopy graphical models used in image segmentation. This requires approximation that can lead to reduced quality of learning. In this paper, we propose three novel techniques to overcome these limitations. We first introduce a method to “kernelize” the features so that a linear SSVM framework can leverage the power of nonlinear kernels without incurring much additional computational cost. Moreover, we employ a working set of constraints to increase the reliability of approximate subgradient methods and introduce a new way to select a suitable step size at each iteration. We demonstrate the strength of our approach on both 2-D and 3-D electron microscopic (EM) image data and show consistent performance improvement over state-of-the-art approaches. [ABSTRACT FROM AUTHOR]

This paper considers the problem of image segmentation using the random walker algorithm. In the case of 3D images, the method uses an extreme amount of memory and time resources. These are required in order to represent the corresponding enormous image graph and to solve the resulting sparse linear system. Having in mind these limitations, this paper proposes techniques for the optimization of the random walker approach. The optimization is obtained by processing supervoxels representing homogeneous image regions rather than single voxels. A fast and efficient method for supervoxel determination is introduced. A method for the creation of an image adjacency graph from an irregular grid of supervoxels is also proposed. The results of applying the introduced approach to segmentation of 3D CT data sets are presented and compared with the results of the original random walker approach and other state-of-the-art methods. The accuracy and the computational overhead is regarded in the comparison. The analysis of results shows that the modified method can be successfully applied for the segmentation of volumetric images and provides results in a reasonable time without a significant loss in the image segmentation accuracy. It also outperforms the state-of-the-art methods considered in the comparison. [ABSTRACT FROM AUTHOR]

• Anomaly detection-inspired few-shot medical image segmentation model. • Supervoxel-based self-supervised learning to exploit 3D image nature. • Evaluated on representative datasets for organ and cardiac segmentation. • Increased robustness to over-segmentation by only modeling the foreground. [Display omitted] Recent work has shown that label-efficient few-shot learning through self-supervision can achieve promising medical image segmentation results. However, few-shot segmentation models typically rely on prototype representations of the semantic classes, resulting in a loss of local information that can degrade performance. This is particularly problematic for the typically large and highly heterogeneous background class in medical image segmentation problems. Previous works have attempted to address this issue by learning additional prototypes for each class, but since the prototypes are based on a limited number of slices, we argue that this ad-hoc solution is insufficient to capture the background properties. Motivated by this, and the observation that the foreground class (e.g., one organ) is relatively homogeneous, we propose a novel anomaly detection-inspired approach to few-shot medical image segmentation in which we refrain from modeling the background explicitly. Instead, we rely solely on a single foreground prototype to compute anomaly scores for all query pixels. The segmentation is then performed by thresholding these anomaly scores using a learned threshold. Assisted by a novel self-supervision task that exploits the 3D structure of medical images through supervoxels, our proposed anomaly detection-inspired few-shot medical image segmentation model outperforms previous state-of-the-art approaches on two representative MRI datasets for the tasks of abdominal organ segmentation and cardiac segmentation. [ABSTRACT FROM AUTHOR]

Most existing superpixel algorithms ignore the spatial structure and regularity properties, which result in undesirable sizes and location relationships for the subsequent processing. In this paper, we introduce a new method to generate the regularity preserved superpixels. Starting from the lattice seeds, our method relocates them to the pixel with locally maximal edge magnitudes and treats them as the superpixel junctions. Then, the shortest path algorithm is employed to find the local optimal boundary connecting each adjacent junction pair. Thanks to the local constraints, our method obtains homogeneous superpixels with adjacency in lowly textured and uniform regions and simultaneously preserves the boundary adherence in the high contrast contents. Our method preserves the regularity property without significantly sacrificing the segmentation accuracy. Moreover, we extend this regular constraint for generating the supervoxels. Our method obtains the regular supervoxels, which preserves the structural relation on both spatial and temporal spaces of the video. Quantitative and qualitative experimental results on benchmark datasets demonstrate that our simple but effective method outperforms the existing regular superpixel methods. [ABSTRACT FROM PUBLISHER]

We propose an information processing pipeline for segmenting parts of the bowel in abdominal magnetic resonance images that are affected with Crohn's disease. Given a magnetic resonance imaging test volume, it is first oversegmented into supervoxels and each supervoxel is analyzed to detect presence of Crohn's disease using random forest (RF) classifiers. The supervoxels identified as containing diseased tissues define the volume of interest (VOI). All voxels within the VOI are further investigated to segment the diseased region. Probability maps are generated for each voxel using a second set of RF classifiers which give the probabilities of each voxel being diseased, normal or background. The negative log-likelihood of these maps are used as penalty costs in a graph cut segmentation framework. Low level features like intensity statistics, texture anisotropy and curvature asymmetry, and high level context features are used at different stages. Smoothness constraints are imposed based on semantic information (importance of each feature to the classification task) derived from the second set of learned RF classifiers. Experimental results show that our method achieves high segmentation accuracy with Dice metric values of 0.90 \pm 0.04 and Hausdorff distance of 7.3 \pm 0.8 mm. Semantic information and context features are an integral part of our method and are robust to different levels of added noise. [ABSTRACT FROM AUTHOR]

It is becoming increasingly clear that mitochondria play an important role in neural function. Recent studies show mitochondrial morphology to be crucial to cellular physiology and synaptic function and a link between mitochondrial defects and neuro-degenerative diseases is strongly suspected. Electron microscopy (EM), with its very high resolution in all three directions, is one of the key tools to look more closely into these issues but the huge amounts of data it produces make automated analysis necessary. State-of-the-art computer vision algorithms designed to operate on natural 2-D images tend to perform poorly when applied to EM data for a number of reasons. First, the sheer size of a typical EM volume renders most modern segmentation schemes intractable. Furthermore, most approaches ignore important shape cues, relying only on local statistics that easily become confused when confronted with noise and textures inherent in the data. Finally, the conventional assumption that strong image gradients always correspond to object boundaries is violated by the clutter of distracting membranes. In this work, we propose an automated graph partitioning scheme that addresses these issues. It reduces the computational complexity by operating on supervoxels instead of voxels, incorporates shape features capable of describing the 3-D shape of the target objects, and learns to recognize the distinctive appearance of true boundaries. Our experiments demonstrate that our approach is able to segment mitochondria at a performance level close to that of a human annotator, and outperforms a state-of-the-art 3-D segmentation technique. [ABSTRACT FROM PUBLISHER]

Voxelisation methods are extensively employed for efficiently processing large point clouds. However, it is possible to lose geometric information and extract inaccurate features through these voxelisation methods. A novel, flexibly shaped 'supervoxel' algorithm, called boundary‐enhanced supervoxel segmentation, for sparse and complex outdoor light detection and ranging (LiDAR) data is proposed. The algorithm consists of two key components: (i) detecting boundaries by analysing consecutive points and (ii) clustering the points by first excluding the boundary points. The generated supervoxels include spatial and geometric properties and maintain the shape of the object's boundary. The proposed algorithm is tested using sparse LiDAR data obtained from outdoor urban environments. [ABSTRACT FROM AUTHOR]

Reality capture technologies, also known as close-range sensing, have been increasingly popular within the field of engineering geology and particularly rock slope management. Such technologies provide accurate and high-resolution n-dimensional spatial representations of our physical world, known as 3D point clouds, that are mainly used for visualization and monitoring purposes. To extract knowledge from point clouds and inform decision-making within rock slope management systems, semantic injection through automated processes is necessary. In this paper, we propose a model that utilizes a segmentation procedure which delivers segments ready to classify and be retained or rejected according to complementary knowledge-based filter criteria. First, we provide relevant voxel-based features based on the local dimensionality, orientation, and topology and partition them in an assembly of homogenous segments. Subsequently, we build a decision tree that utilizes geometrical, topological, and contextual information and enables the classification of a multi-hazard railway rock slope section in British Columbia, Canada into classes involved in landslide risk management. Finally, the approach is compared to machine learning integrating recent featuring strategies for rock slope classification with limited training data (which is usually the case). This alternative to machine learning semantic segmentation approaches reduces substantially the model size and complexity and provides an adaptable framework for tailored decision-making systems leveraging rock slope semantics. [ABSTRACT FROM AUTHOR]