Clustering Segmentation Research Articles

“Using behavioral insights to inform the COVID-19 vaccine response in Kosovo[1]: Population perceptions and interventions”

ObjectiveUnderstand population perceptions in Kosovo[1] regarding COVID-19 vaccination to inform the pandemic response. MethodsFive rounds of a cross-sectional survey in representative samples of adults during 2020–2021. Analysis includes descriptive statistics, hierarchical cluster analysis, segmentation and logistic regressions. ResultsSelf-reported intention to vaccinate increased after the introduction of COVID-19 vaccines in Kosovo.[1] In less than one year, vaccination intentions increased from 36% to 66% of those unvaccinated. Predictors for vaccine intentions included gender, age, trust in health authorities. Segmentation analysis identified population segments that had high vaccine nintentions but low uptake, informing messages and campaign initiatives designed to translate intentions into behavior. ConclusionIdentifying people's perceptions and behavior is essential to support evidence-based policy making, especially during outbreak response. InnovationBI is an innovative focus of research in Kosovo [1] where little BI data had been collected prior, and provided a unique understanding of population views, attitudes and behaviors related to COVID-19. These findings were not only essential for an evidence-based pandemic response but also laid the foundation for future broad application of BI to inform interventions that seek to enable, support and promote health-related behaviurs in Kosovo[1]

Estimation of Weight and Body Measurement Model for Pigs Based on Back Point Cloud Data.

Pig farming is a crucial sector in global animal husbandry. The weight and body dimension data of pigs reflect their growth and development status, serving as vital metrics for assessing their progress. Presently, pig weight and body dimensions are predominantly measured manually, which poses challenges such as difficulties in herding, stress responses in pigs, and the control of zoonotic diseases. To address these issues, this study proposes a non-contact weight estimation and body measurement model based on point cloud data from pig backs. A depth camera was installed above a weighbridge to acquire 3D point cloud data from 258 Yorkshire-Landrace crossbred sows. We selected 200 Yorkshire-Landrace sows as the research subjects and applied point cloud filtering and denoising techniques to their three-dimensional point cloud data. Subsequently, a K-means clustering segmentation algorithm was employed to extract the point cloud corresponding to the pigs' backs. A convolutional neural network with a multi-head attention was established for pig weight prediction and added RGB information as an additional feature. During the data processing process, we also measured the back body size information of the pigs. During the model evaluation, 58 Yorkshire-Landrace sows were specifically selected for experimental assessment. Compared to manual measurements, the weight estimation exhibited an average absolute error of 11.552 kg, average relative error of 4.812%, and root mean square error of 11.181 kg. Specifically, for the MACNN, incorporating RGB information as an additional feature resulted in a decrease of 2.469 kg in the RMSE, a decrease of 0.8% in the MAPE, and a decrease of 1.032 kg in the MAE. Measurements of shoulder width, abdominal width, and hip width yielded corresponding average relative errors of 3.144%, 3.798%, and 3.820%. In conclusion, a convolutional neural network with a multi-head attention was established for pig weight prediction, and incorporating RGB information as an additional feature method demonstrated accuracy and reliability for weight estimation and body dimension measurement.

Shape changes of a single hairy particle with mobile ligands at a liquid-liquid interface

We investigate rearrangements of a single hairy particle at a liquid-liquid interface using coarse-grained molecular dynamics simulations. We consider the particles with the same (symmetrical interactions) and different (asymmetrical interactions) affinity to the liquids. We show how ligand mobility affects the behavior of the hairy particle at the liquid-liquid interface. We found that such a hairy particle can take various shapes at the interface. For example, a Janus-like snowman consisting of a segment cluster and a bare part of the core, Saturn-like structures, and the core with a wide “plume” on one side. A configuration of the particle at the interface is characterized by the vertical displacement distance and the orientation of the particle relative to the phase boundary. The selected descriptors are used to characterize the shape of the segment cloud. We found that the shape of a particle and its localization at the interface can be determined by tuning the interactions with the liquids.

Research on Point Cloud Structure Detection of Manhole Cover Based on Structured Light Camera

This study introduced an innovative approach for detecting structural anomalies in road manhole covers using structured light cameras. Efforts have been dedicated to enhancing data quality by commencing with the acquisition and preprocessing of point cloud data from real-world manhole cover scenes. The RANSAC algorithm is subsequently employed to extract the road plane and determine the height of the point cloud structure. In the presence of non-planar point cloud exhibiting abnormal heights, the DBSCAN algorithm is harnessed for cluster segmentation, aiding in the identification of individual objects. The method culminates with the introduction of a sector fitting detection model, adept at effectively discerning manhole cover features within the point cloud and delivering comprehensive height and structural information. Experimental findings underscore the method’s efficacy in accurately gauging the degree of subsidence in manhole cover structures, with data errors consistently maintained within an acceptable range of 8 percent. Notably, the measurement speed surpasses that of traditional methods, presenting a notably efficient and dependable technical solution for road maintenance.

Abstract 4308: Same-section spatial multiomic analyses using MICS technology for investigating the dynamics of the tumor microenvironment

Abstract The recent increase in image-based, spatially-resolved technologies enables researchers to profile the tumor microenvironment (TME) by capturing gene expression profiles within tissue sections. However, a significant limitation of these technologies is the lack of ability to resolve protein and RNA information in the same section, as well as conveniently analyze multimodal data sets. Here, we report a spatial RNA detection method, RNAsky, using Miltenyi Biotec’s MACSima™ Platform as an automated, multiomic approach. Our method integrates spatial proteomics and transcriptomics data to provide in-depth profiling with single-cell resolution on the same tissue section. We demonstrate these capabilities by characterizing key immune-oncology markers across normal and diseased tissues using our specialized MACS® iQ View analysis software. MACS iQ View provides fast segmentation and intuitive gating and clustering strategies to simultaneously assess protein and RNA data. We investigated the impact of clustering using protein, RNA, or the combination of both to evaluate the contribution of different information modalities on TME spatial dynamics. This cutting-edge approach will enable the identification of valuable parameters and new cell types, furthering the discovery and development of predictive and prognostic biomarkers. Citation Format: Dongju Park, Emily Neil, Rebecca C. Hennessey, Michael DiBuono, Hanna Lafayette, Erica Lloyd, Hsinyi Lo, Julia Femel, Alex Makrigiorgos, Shaina Lu, John Lee, Sameh Soliman, Dominic Mangiardi, Paurush Praveen, Fabian Staubach, Ryan Hindman, Thomas Rothmann, Telma Santos, Stefan Borbe, Hansueli Meyer, Tanya Wantenaar, Jinling Wang, Werner Müller, Robert Pinard, Andreas Bosio. Same-section spatial multiomic analyses using MICS technology for investigating the dynamics of the tumor microenvironment [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 4308.

UAV building point cloud contour extraction based on the feature recognition of adjacent points distribution

Building point cloud contour is widely used in urban planning and building identification. The existing method cannot work well in the extraction of UAV building point cloud contour points because the low point cloud accuracy of UAV, and effect of noise. In this paper we propose extraction method of UAV building point cloud contour based on the adjacent points distribution. Firstly, we construct an information entropy model for determining optimal neighborhood points, and then use the eigenvectors corresponding to the maximum and minimum eigenvalues obtained by PCA to construct initial projection plane Π1 and Π2. Secondly, we propose the fine-tuning model of the projection plane Π2 by using the aggregation characteristics of neighborhood points, and construct the extraction model of fold points with three parameter constraints. Finally, we propose the information entropy model of points probabilities in different quadrants on the plane Π1 to extract boundary points. Real UAV point cloud data is used to test the performance and parameters of the proposed method, experiment results show that the proposed method is superior to OPAHT, 2D line detection and region clustering segmentation methods in the performance of UAV building point cloud contour extraction. The proposed method can accurately extract the point cloud contour points of UAV buildings.

Spatial Single Cell Profiling Using Imaging Mass Cytometry: Inflammatory Versus Penetrating Crohn's Disease.

Fistula formation is a major complication in Crohn's disease [CD] and the role of the immune cell compartment remains to be elucidated. Thus, we compared the immune cell compartment of CD fistula to inflammatory CD colitis using imaging mass cytometry [IMC] and immunofluorescence. A 36-marker panel including structural, functional, and lineage markers for use in IMC was established. This panel was applied to analyse paraffin-embedded CD fistula tract [n = 11], CD colitis [n = 10], and colon samples from non-inflamed controls [n = 12]. Computational methods for cell segmentation, dimensionality reduction, and cell type clustering were used to define cell populations for cell frequency, marker distribution, and spatial neighbourhood analysis. Multiplex immunofluorescence was used for higher resolution spatial analysis. Analysis of cell frequencies in CD fistulas compared to CD colitis and control colonic samples revealed a significant increase in neutrophils, effector cytotoxic T cells, and inflammatory macrophages in CD fistula samples, whereas regulatory T cells were decreased. Neutrophils in CD fistula expressed significantly more matrix metalloproteinase 9 [MMP9], correlating with extracellular matrix remodelling. Neighbourhood analysis revealed a strong association between MMP9+ neutrophils and effector cytotoxic T cells in both CD fistulas and colitis. This study presents the first highly multiplexed single cell analysis of the immune cell compartment of CD fistulas and their spatial context. It links immune cell dynamics, particularly MMP9+ neutrophils, to extracellular matrix remodelling in CD fistulas, offering insights into the complex network of cellular interactions and potential therapeutic targets for CD complications.

Investigating the temporal dynamics of motor vehicle collision density patterns in urban road networks – A case study of New York

Introduction: Motor vehicle collisions are a leading source of mortality and injury on urban highways. From a temporal perspective, the determination of a road segment as being collision-prone over time can fluctuate dramatically, making it difficult for transportation agencies to propose traffic interventions. However, there has been limited research to identify and characterize collision-prone road segments with varying collision density patterns over time. Method: This study proposes an identification and characterization framework that profiles collision-prone roads with various collision density variations. We first employ the spatio-temporal network kernel density estimation (STNKDE) method and time-series clustering to identify road segments with different collision density patterns. Next, we characterize collision-prone road segments based on spatio-temporal information, consequences, vehicle types, and contributing factors to collisions. The proposed method is applied to two-year motor vehicle collision records for New York City. Results: Seven clusters of road segments with different collision density patterns were identified. Road segments frequently determined as collision-prone were primarily found in Lower Manhattan and the center of the Bronx borough. Furthermore, collisions near road segments that exhibit greater collision densities over time result in more fatalities and injuries, many of which are caused by both human and vehicle factors. Conclusions: Collision-prone road segments with various collision density patterns over time have distinct differences in the spatio-temporal domain and the collisions that occur on them. Practical Applications: The proposed method can help policymakers understand how collision-prone road segments change over time, and can serve as a reference for more targeted traffic treatment.

Gaussian model for closed curves

In the case of image processing or understanding, one of the common important tasks is to fit closed curves (e.g., circles, ellipses, etc.) to the underlying image. In higher-dimensional situations, the problem of modeling clusters as closed curves remains even more challenging.To deal with this problem we introduce a new probability distribution, which models complicated closed curves with the use of Fourier series. Then, a mixture of such distributions is constructed, leading to our model MCEC. It is shown that MCEC can be effectively trained in the case of closed curves in Rn.MCEC was evaluated in clustering, curve fitting, and image segmentation tasks. We compare it in particular to classical GMM, Hough transform, and Snakes algorithm. In each task, we consistently obtain or outperform the current SOTA.

Artificial neural network (ANN) modelling to estimate bubble size from macroscopic image and object features

Bubble size measurements in aerated systems such as froth flotation cells are critical for controlling gas dispersion. Commonly, bubbles are measured by obtaining representative photographs, which are then analyzed using segmentation and identification software tools. Recent developments have focused on enhancing these segmentation tools. However, the main challenges around complex bubble cluster segmentation remain unresolved, while the tools to tackle these challenges have become increasingly complex and computationally expensive. In this work, we propose an alternative solution, circumventing the need for image segmentation and bubble identification. An Artificial Neural Network (ANN) was trained to estimate the Sauter mean bubble size (D<sub>32</sub>) based on macroscopic image features obtained with simple and inexpensive image analysis. The results showed excellent prediction accuracy, with a correlation coefficient, R, over 0.998 in the testing stage, and without bias in its error distribution. This machine learning tool paves the way for robust and fast estimation of bubble size under complex bubble images, without the need of image segmentation.

A Novel Approach for Predicting Anthropogenic CO2 Emissions Using Machine Learning Based on Clustering of the CO2 Concentration

The monitoring of anthropogenic CO2 emissions, which increase the atmospheric CO2 concentration, plays the most important role in the management of emission reduction and control. With the massive increase in satellite-based observation data related to carbon emissions, a data-driven machine learning method has great prospects for predicting anthropogenic CO2 emissions. Training samples, which are used to model predictions of anthropogenic CO2 emissions through machine learning algorithms, play a key role in obtaining accurate predictions for the spatial heterogeneity of anthropogenic CO2 emissions. We propose an approach for predicting anthropogenic CO2 emissions using the training datasets derived from the clustering of the atmospheric CO2 concentration and the segmentation of emissions to resolve the issue of the spatial heterogeneity of anthropogenic CO2 emissions in machine learning modeling. We assessed machine learning algorithms based on decision trees and gradient boosting (GBDT), including LightGBM, XGBoost, and CatBoost. We used multiple parameters related to anthropogenic CO2-emitting activities as predictor variables and emission inventory data from 2019 to 2021, and we compared and verified the accuracy and effectiveness of different prediction models based on the different sampling methods of training datasets combined with machine learning algorithms. As a result, the anthropogenic CO2 emissions predicted by CatBoost modeling from the training dataset derived from the clustering analysis and segmentation method demonstrated optimal prediction accuracy and performance for revealing anthropogenic CO2 emissions. Based on a machine learning algorithm using observation data, this approach for predicting anthropogenic CO2 emissions could help us quickly obtain up-to-date information on anthropogenic CO2 emissions as one of the emission monitoring tools.

Geomorphic Imprints of Active Tectonics of the Bikaner-Nagaur Petroliferous Rift Basin and its Surroundings (Western Rajasthan, India)

Abstract Geology of sedimentary rift-basins require strong geomorphic input for a proper interpretation of active tectonics. Rift-related sedimentation took place in western Rajasthan of the Indian shield, which includes the Bikaner-Nagaur basin (BNB) and a few other adjacent basins. The sedimentation history of the BNB includes Proterozoic, Cambrian, Permo-Carboniferous and from Paleocene to the Recent. This study analyses river profiles with the best-fit curve (R2) model for the BNB and the surrounding regions. The research shows that the watershed 3 within the study area is most active tectonically, through which multiple faults and lineaments pass. Hypsometric Curves (HCs) of watersheds 1, 2 and 3 indicate that these watersheds are tectonically active. This inference is based on the concave profiles of HCs at the head, and convex profiles of HCs at the body and toe sections. Clustering of sixty segments (S) of the considered rivers based on linear-scale morphometric parameters, e.g., sinuosity index (SI), stream length gradient index (SL) and concavity (θ) enabled segment-wise comparison of river profiles with similar values. These segments were compared pair-wise, and Euclidean-based dissimilarity (dR) values were calculated between each such pair. The findings too imply that tectonic activeness exists in parts of watersheds 1, 2 and 3. The channel flow lines are controlled by faults/lineaments as per the micro-scale examination of the drainage network and faults/ lineaments analysis. Under structural control, nine major geomorphic units emerged with distinct erosional surfaces, denudational hillocks, dissected hills and inselbergs. Detailed geomorphic map with micro-scale studies revealed a slope retreat process that resulted in landforms viz., pediment, pediment slope and active flood plains.

Computational Modeling of the Segmentation of Sentence Stimuli From an Infant Word-Finding Study.

Computational models of infant word-finding typically operate over transcriptions of infant-directed speech corpora. It is now possible to test models of word segmentation on speech materials, rather than transcriptions of speech. We propose that such modeling efforts be conducted over the speech of the experimental stimuli used in studies measuring infants' capacity for learning from spoken sentences. Correspondence with infant outcomes in such experiments is an appropriate benchmark for models of infants. We demonstrate such an analysis by applying the DP-Parser model of Algayres and colleagues to auditory stimuli used in infant psycholinguistic experiments by Pelucchi and colleagues. The DP-Parser model takes speech as input, and creates multiple overlapping embeddings from each utterance. Prospective words are identified as clusters of similar embedded segments. This allows segmentation of each utterance into possible words, using a dynamic programming method that maximizes the frequency of constituent segments. We show that DP-Parse mimics American English learners' performance in extracting words from Italian sentences, favoring the segmentation of words with high syllabic transitional probability. This kind of computational analysis over actual stimuli from infant experiments may be helpful in tuning future models to match human performance.

Research on Individual Tree Canopy Segmentation of Camellia oleifera Based on a UAV-LiDAR System

In consideration of the limited accuracy of individual tree canopy segmentation algorithms due to the diverse canopy structure and complex environments in mountainous and hilly areas, this study optimized the segmentation parameters of three algorithms for individual tree canopy segmentation of Camellia oleifera in such environments by analyzing their respective parameters. Utilizing an Unmanned Aerial Vehicle-Light Detecting and Ranging (UAV-LiDAR) system, we obtained Canopy Height Models (CHM) of Camellia oleifera canopies based on Digital Elevation Models (DEM) and Digital Surface Models (DSM). Subsequently, we investigated the effects of CHM segmentation, point cloud clustering segmentation, and layer stacking fitting segmentation on Camellia oleifera canopies across different research areas. Additionally, combining ground survey data from forest lands with visual interpretation of UAV orthophoto images, we evaluated the performance of these three segmentation algorithms in terms of the F-score as an evaluation indicator for individual tree canopy segmentation accuracy. Combined with the Cloth Simulation Filter (CSF) filtering algorithm after removing the ground point cloud, our findings indicate that among different camellia densities and terrain environments, the point cloud clustering segmentation algorithm achieved the highest segmentation accuracy at 93%, followed by CHM segmentation at 88% and the layer stacking fitting segmentation method at 84%. By analyzing the data from UAV-LiDAR technology involving various land and Camellia oleifera planting types, we verified the applicability of these three segmentation algorithms for extracting camellia canopies. In conclusion, this study holds significant importance for accurately delineating camellia canopies within mountainous hilly environments while providing valuable insights for further research in related fields.

An approach for traffic pattern recognition integration of ship AIS data and port geospatial features

ABSTRACT Recognition of ship traffic patterns can provide insights into the rules of navigation, maneuvering, and collision avoidance for ships at sea. This is essential for ensuring safe navigation at sea and improving navigational efficiency. With the popularization of the Automatic Identification System (AIS), numerous studies utilized ship trajectories to identify maritime traffic patterns. However, the current research focuses on the spatiotemporal behavioral feature clustering of ship trajectory points or segments while lacking consideration for multiple factors that influence ship behavior, such as ship static and maritime geospatial features, resulting in insufficient precision in ship traffic pattern recognition. This study proposes a ship traffic pattern recognition method that considers multi-attribute trajectory similarity (STPMTS), which considers ship static feature, dynamic feature, port geospatial feature, as well as semantic relationships between these features. First, A ship trajectory reconstruction method based on grid compression was introduced to eliminate redundant data and enhance the efficiency of trajectory similarity measurements. Subsequently, to quantify the degree of similarity of ship trajectories, a trajectory similarity measurement method is proposed that combines ship static and dynamic information with port geospatial features. Furthermore, trajectory clustering with hierarchical methods was applied based on the trajectory similarity matrix for dividing trajectories into different clusters. The quality of the similarity measurement results was evaluated by quality criterion to recognize the optimal number of ship traffic patterns. Finally, the effectiveness of the proposed method was verified using actual port ship trajectory data from the Tianjin Port of China, ranging from September to November 2016. Compared with other methods, the proposed method exhibits significant advantages in identifying traffic patterns of ships entering and leaving the port in terms of geometric features, dynamic features, and adherence to navigation rules. This study could serve as an inspiration for a comprehensive exploration of maritime transportation knowledge from multiple perspectives.

Optimized self-attention based cycle-consistent generative adversarial network adopted melanoma classification from dermoscopic images.

Skin is the exposed part of the human body that constantly protected from UV rays, heat, light, dust, and other hazardous radiation. One of the most dangerous illnesses that affect people is skin cancer. A type of skin cancer called melanoma starts in the melanocytes, which regulate the colour in human skin. Reducing the fatality rate from skin cancer requires early detection and diagnosis of conditions like melanoma. In this article, a Self-attention based cycle-consistent generative adversarial network optimized with Archerfish Hunting Optimization Algorithm adopted Melanoma Classification (SACCGAN-AHOA-MC-DI) from dermoscopic images is proposed. Primarily, the input Skin dermoscopic images are gathered via the dataset of ISIC 2019. Then, the input Skin dermoscopic images is pre-processed using adjusted quick shift phase preserving dynamic range compression (AQSP-DRC) for removing noise and increase the quality of Skin dermoscopic images. These pre-processed images are fed to the piecewise fuzzy C-means clustering (PF-CMC) for ROI region segmentation. The segmented ROI region is supplied to the Hexadecimal Local Adaptive Binary Pattern (HLABP) to extract the Radiomic features, like Grayscale statistic features (standard deviation, mean, kurtosis, and skewness) together with Haralick Texture features (contrast, energy, entropy, homogeneity, and inverse different moments). The extracted features are fed to self-attention based cycle-consistent generative adversarial network (SACCGAN) which classifies the skin cancers as Melanocytic nevus, Basal cell carcinoma, Actinic Keratosis, Benign keratosis, Dermatofibroma, Vascular lesion, Squamous cell carcinoma and melanoma. In general, SACCGAN not adapt any optimization modes to define the ideal parameters to assure accurate classification of skin cancer. Hence, Archerfish Hunting Optimization Algorithm (AHOA) is considered to maximize the SACCGAN classifier, which categorizes the skin cancer accurately. The proposed method attains 23.01%, 14.96%, and 45.31% higher accuracy and 32.16%, 11.32%, and 24.56% lesser computational time evaluated to the existing methods, like melanoma prediction method for unbalanced data utilizing optimized Squeeze Net through bald eagle search optimization (CNN-BES-MC-DI), hyper-parameter optimized CNN depending on Grey wolf optimization algorithm (CNN-GWOA-MC-DI), DEANN incited skin cancer finding depending on fuzzy c-means clustering (DEANN-MC-DI). RESEARCH HIGHLIGHTS: This manuscript, self-attention based cycle-consistent. SACCGAN-AHOA-MC-DI method is implemented in Python. (SACCGAN-AHOA-MC-DI) from dermoscopic images is proposed. Adjusted quick shift phase preserving dynamic range compression (AQSP-DRC). Removing noise and increase the quality of Skin dermoscopic images.

Signal Intensity Trajectories Clustering for Liver Vasculature Segmentation and Labeling (LiVaS) on Contrast-Enhanced MR Images: A Feasibility Pilot Study.

This study aims to develop a semiautomated pipeline and user interface (LiVaS) for rapid segmentation and labeling of MRI liver vasculature and evaluate its time efficiency and accuracy against manual reference standard. Retrospective feasibility pilot study. Liver MR images from different scanners from 36 patients were included, and 4/36 patients were randomly selected for manual segmentation as referenced standard. The liver was segmented in each contrast phase and masks registered to the pre-contrast segmentation. Voxel-wise signal trajectories were clustered using the k-means algorithm. Voxel clusters that best segment the liver vessels were selected and labeled by three independent radiologists and a research scientist using LiVaS. Segmentation times were compared using a paired-sample t-test on log-transformed data. The agreement was analyzed qualitatively and quantitatively using DSC for hepatic and portal vein segmentations. The mean segmentation time among four readers was significantly shorter than manual (3.6 ± 1.4 vs. 70.0 ± 29.2min; p < 0.001), even when using a higher number of clusters to enhance accuracy. The DSC for portal and hepatic veins reached up to 0.69 and 0.70, respectively. LiVaS segmentations were overall of good quality, with variations in performance related to the presence/severity of liver disease, acquisition timing, and image quality. Our semi-automated pipeline was robust to different MRI vendors in producing segmentation and labeling of liver vasculature in agreement with expert manual annotations, with significantly higher time efficiency. LiVaS could facilitate the creation of large, annotated datasets for training and validation of neural networks for automated MRI liver vascularity segmentation. HIGHLIGHTS: Key Finding: In this pilot feasibility study, our semiautomated pipeline for segmentation of liver vascularity (LiVaS) on MR images produced segmentations with simultaneous labeling of portal and hepatic veins in good agreement with the manual reference standard but at significantly shorter times (mean LiVaS 3.6 ± 1.4 vs. mean manual 70.0 ± 29.2min; p < 0.001). Importance: LiVaS was robust in producing liver MRI vascular segmentations across images from different scanners in agreement with expert manual annotations, with significant ly higher time efficiency, and therefore potential scalability.

Segmentation of Individual Tree Points by Combining Marker-Controlled Watershed Segmentation and Spectral Clustering Optimization

Accurate identification and segmentation of individual tree points are crucial for assessing forest spatial distribution, understanding tree growth and structure, and managing forest resources. Traditional methods based on Canopy Height Models (CHM) are simple yet prone to over- and/or under-segmentation. To deal with this problem, this paper introduces a novel approach that combines marker-controlled watershed segmentation with a spectral clustering algorithm. Initially, we determined the local maxima within a series of variable windows according to the lower bound of the prediction interval of the regression equation between tree crown radius and tree height to preliminarily segment individual trees. Subsequently, using this geometric shape analysis method, the under-segmented trees were identified. For these trees, vertical tree crown profile analysis was performed in multiple directions to detect potential treetops which were then considered as inputs for spectral clustering optimization. Our experiments across six plots showed that our method markedly surpasses traditional approaches, achieving an average Recall of 0.854, a Precision of 0.937, and an F1-score of 0.892.

A visual measurement method of grinding surface roughness based on aliasing region index and neural network

Most existing vision-based roughness measurements primarily rely on statistical information from grayscale images or intensity information from color images. However, the structural information of images has not been fully and effectively utilized. To more accurately measure the roughness of grinding surfaces, a visual measurement method of grinding surface roughness based on aliasing region index and neural network is proposed. Firstly, color images of grinding surface are obtained under red and green illumination. Secondly, aliasing regions of red and green images are extracted through fuzzy clustering segmentation and morphological processing. Then the aliasing width and the aliasing dispersion of aliasing region can be calculated as indices for roughness measurement. Thirdly, the relationship model between aliasing region index and grinding surface roughness is constructed using the back propagation (BP) neural network. The results demonstrate that the aliasing dispersion index has a better correlation with grinding surface roughness than the aliasing width index. The method based on the aliasing dispersion index and BP neural network is feasible and accurate for grinding surface roughness measurement.

A Robotic Arm Visual Grasp Detection Algorithm Combining 2D Images and 3D Point Clouds

Robot grasping detection methods are categorized into two-dimension (2D) and three-dimension (3D) approaches. In 2D grasp detection, gripper pose prediction occurs directly on RedGreenBlue (RGB) images, limiting grasp direction. Conversely, 3D grasp detection predicts gripper pose using 3D point clouds, allowing greater grasp flexibility. However, the data volume of 3D point clouds hampers real-time detection. To address this, this paper proposes a novel grasping detection algorithm that combines 2D images and 3D point clouds. Initially, a Single Shot MultiBox Detector (SSD) network generates 2D prediction boxes on RGB images. Through enhancements to the prior box scaling, the accuracy of bounding the target object is improved. Subsequently, 2D boxes are transformed into 3D frustums, and extraneous data points are removed. By utilizing Random Sampling Consistent Segmentation (RANSAC) and Euclidean Clustering Segmentation Algorithm (ECSA), the target point clouds are isolated, and subsequently, the spatial pose of the target is represented using an Oriented Bounding Box (OBB). Processed point clouds enter an enhanced PointNet Grasp Pose Detection (PointNetGPD) algorithm, In contrast to the original approach involving extensive random sampling of grasp candidates, the enhanced PointNetGPD method enables the selective sampling of grasp candidates by incorporating pose constraints between the target and gripper. Following this, the generated grasp candidates are subjected to evaluation through a scoring process conducted by an assessment network. Ultimately, the robotic arm is guided to perform the grasp associated with the highest score. In the experimental phase, the proposed algorithm demonstrated a high success rate in capturing multiple targets, along with a reduced grasping time. These results underscore the algorithm's superior grasping quality and enhanced real-time performance when compared to similar algorithms.