Homogeneous Segments Research Articles

Non-Convex Metric Learning-Based Trajectory Clustering Algorithm

To address the issue of suboptimal clustering performance arising from the limitations of distance measurement in traditional trajectory clustering methods, this paper presents a novel trajectory clustering strategy that integrates the bag-of-words model with non-convex metric learning. Initially, the strategy extracts motion characteristic parameters from trajectory points. Subsequently, based on the minimum description length criterion, trajectories are segmented into several homogeneous segments, and statistical properties for each segment are computed. A non-convex metric learning mechanism is then introduced to enhance similarity evaluation accuracy. Furthermore, by combining a bag-of-words model with a non-convex metric learning algorithm, segmented trajectory fragments are transformed into fixed-length feature descriptors. Finally, the K-means method and the proposed non-convex metric learning algorithm are utilized to analyze the feature descriptors, and hence, the effective clustering of trajectories can be achieved. Experimental results demonstrate that the proposed method exhibits superior clustering performance compared to the state-of-the-art trajectory clustering approaches.

Object-based image analysis and machine learning for mapping cashew plantations in Ariyalur district, Tamil Nadu

An object-based image analysis (OBIA) approach provides a comprehensive method for delineating homogeneous segments based on spectral characteristics, geometry, and spatial imagery structures. The present study utilizes OBIA and machine learning (ML) techniques to map cashew plantations in Ariyalur district of Tamil Nadu, India. Sentinel-2 Multi-Spectral Instrument (MSI) imagery, acquired during the 2023 Kharif season, was employed as the primary data source due to its high spatial and spectral resolution, suitable for detailed land cover mapping. The OBIA methodology involved multiresolution segmentation using eCognition software to delineate homogeneous image objects based on spectral, spatial, and contextual characteristics. Machine learning algorithms, including Random Forest (RF), Support Vector Machine (SVM), and Decision Tree (DT), were evaluated to improve classification accuracy. The SVM demonstrated the best superior performance, achieving an overall accuracy of 92.1% and a kappa coefficient of 0.85. The results underscore the effectiveness of machine learning techniques in conjunction with object-based image analysis (OBIA) for precise cashew plantation mapping while contributing to improved land use/land cover mapping, agricultural resource management, and sustainable development within the region.

EXPRESS: Multi-Segment and Single-Segment Sales Contests: Evidence of Their Effectiveness and the Underlying Mechanisms

Sales contests are vital motivational tools, yet their effectiveness remains unclear. This research provides valuable insights by examining two contest types: 1) single-segment contests, where a salesperson competes against all other salespeople, and 2) multi-segment contests, where salespeople are divided into homogeneous segments and compete only within their respective segments. The authors compare both contests using different theoretical lenses and integrate managerial insights. For the empirical assessment, the authors carried out a field experiment with two treatment groups, each group participated in a different contest type, and a control group that did not participate in any contest. To strengthen the internal and external validity of the findings, they conducted four supplementary studies, which comprised two additional field studies and two experimental studies. Findings reveal that, on average, single-segment contests do not outperform a no-contest situation, while multi-segment contests enhance not only the average performance but also the performance across the segments. This performance improvement is mediated by a higher perceived chance of winning and motivation. Additionally, the authors compare multi-segment contests with 'duel-style' contests, where a salesperson competes in one-on-one duels within the same segment. Results indicate that the former yields superior results.

A hybrid method for delineating homogeneous forest segments from LiDAR data in Catalonia (Spain)

ABSTRACT Laser scanning data are increasingly used to segment forests into homogeneous units, called segments or stand compartments. Region growing and region merging have been traditionally employed for this purpose. Recently, alternative methods such as cellular automata (CA), self-organizing maps (SOM), and combinatorial optimization have emerged, promising more homogeneous stand delineations. However, these newer methods often require fine-tuning due to rugged stand boundaries. To address this, we propose and assess a novel hybrid approach that combines two segmentation algorithms, resulting in smoother boundaries and homogeneous stands. Our hybrid method outperforms traditional fine-tuning techniques, particularly using CA or SOM for initial segmentation and then refining it with another cellular automaton (CA’), adjusted to produce smooth boundaries. The new two-step hybrid method resulted in better segmentation results than the one-step algorithm combined with previously suggested fine-tuning methods. SOM-CA’ showed the highest degree of explained variance (R2) of the LiDAR metrics used in the segmentation. CA-CA’ reached almost the same R2 values with a larger average segment area. When the R2 of the LiDAR metrics was used as the criterion, the segments produced by the hybrid method outperformed the stand delineations of the current Spanish forest map with a clear margin.

Analyzing Rear-End Crash Counts on Ohio Interstate Freeways Using Advanced Multilevel Modeling

This study presents a new modeling approach for rear-end crash counts on Ohio’s interstate freeways based on a dataset for 2021 that contains 2745 rear-end crashes. The analysis encompasses 20 interstate freeways, comprising 1833 homogeneous segments and extending over approximately 1313 miles. These interstate freeways exhibit varying safety performances, indicating a significant degree of heterogeneity. A unique rear-end crash risk rate was devised for each interstate, capturing diverse risk profiles. Three distinct models were developed: a standard negative binomial model, an uncorrelated two-level negative binomial model, and a correlated two-level negative binomial model. The correlated two-level negative binomial model demonstrated superior fit, as evidenced by the likelihood ratio test, Akaike information criterion, and Bayesian information criterion. The correlated two-level negative binomial model exhibited enhanced forecasting precision, as measured by the Root Mean Square Error. A significant finding is that the rear-end crash risk rate significantly improves the fit of the models. The study also reveals that rear-end crashes are expected to occur more frequently in urban segments of interstate freeways with high rear-end risk rates. However, rural segments experience no such significant variations in the rear-end crash risk rate. However, an increase in the inner shoulder width is associated with a decrease in expected rear-end crashes. This research offers a valuable methodology for modeling rear-end crashes on interstate freeways, providing insights into the contributing variables that could inform targeted safety improvements.

Characterizing humpback whale behavior along the North-Norwegian coast

BackgroundStudying movement patterns of individual animals over time can give insight into how they interact with the environment and optimize foraging strategies. Humpback whales (Megaptera novaeangliae) undertake long seasonal migrations between feeding areas in polar regions and breeding grounds in tropical areas. During the last decade, several individuals have had up to a 3-month stop-over period around specific fjord-areas in Northern Norway to feed on Norwegian spring-spawning (NSS-) herring (Culpea harengus L.). Their behavioral patterns during this period are not well understood, including why some whales seemingly leave the fjords and then later return within the same season.MethodsTo investigate whale behavior during this seasonal stopover, we classified humpback whale tracks into five distinct movement modes; ranging, encamped, nomadic, roundtrip and semi-roundtrip. A behavioral change point analysis (BCPA) was used to select homogeneous segments based on persistence velocity. Then, net squared displacement (NSD) over time was modeled to differentiate movement modes. This study also manually identified longer roundtrips away from the fjords that lasted several days and examined movement modes within these.ResultsInside the fjord systems, encamped mode was most prevalent in December–January, suggesting the whales were mainly foraging on overwintering NSS-herring in this area. During the same winter seasons, half of the whales left the fjords and then returned. We hypothesize that these trips serve as ‘searching trips’ during which the whales seek better feeding opportunities outside the fjords. If better foraging conditions are not found, they return to the fjords to continue their feeding. The overall most common mode was ranging (54%), particularly seen during the start of their southwards migration and in areas outside the fjord systems, indicating that the whales mainly moved over larger distances in the offshore habitat.ConclusionsThis study serves as a baseline for future studies investigating both the searching trip theory and humpback whale behavior in general, and confirms that this method could be useful to analyze local scale movement patterns of satellite tagged whales.

Impact of urban road characteristics on vehicle speed: Insights from Brescia, Italy

In managing road infrastructures, a key benchmark is the 85th percentile of vehicle speeds (V85). While V85 can be derived from spot speed samples, these are often lacking on each urban road. Thus, prediction models become valuable tools for examining the relationship between V85 and road characteristics. Although various models exist for rural roads, the impact of roadside characteristics and markings on V85 in urban road networks has been partially investigated, and the effect of traffic calming measures remains fragmented.This study aims to address these gaps by applying a methodology that sheds light on the effects of some variables that influence V85 on urban roads. Specifically, the methodology selects and segments roads along the urban road network of the municipality of Brescia (Italy) and collects data on both road characteristics and 48,000+ spot speed information. Following data cleansing, it processes these data according to three different multiple regression models to analyse the influence of various predictors on V85. Once the best model is estimated, its performance is evaluated, and the final list of significant predictors is obtained.The results revealed that V85 increases with longer homogeneous segments, greater distance to successive intersections, bituminous conglomerate roads with more lanes, and the presence of trees, visible road markings, and posted speed limits. Conversely, V85 decreases in the presence of on-street parking and other obstacles (e.g., walls and road posts), when the density of road intersections and pedestrian crossings increases, when the left crossbar width increases and when the land use crossed is commercial or office, residential or industrial. Nevertheless, no significant effect was found for several traffic calming measures included in the model.These findings can assist road authorities in verifying road operating conditions and planning infrastructure interventions to reduce speeds, thereby creating a safer urban environment for all users.

Ordinal Clustering Based Homogeneous Road Segments in Asphalt Pavement Maintenance and Rehabilitation Optimized Decision-Making

An increase in the number of work zones on roads and highways will result in more costs from the increased frequency of maintenance and rehabilitation (M&R) equipment transfers. The number of M&R decision-making segments can affect decision-making efficiency and the number of work zones. Fixed segmentation and traditional dynamic homogeneous segmentation techniques, such as the cumulative difference approach (CDA) and K-means algorithms, cannot determine the optimal number of decision-making segments. To address this issue, this paper proposes a method to develop a multi-year asphalt pavement M&R plan that incorporates homogeneous road segmentation based on an ordinal clustering approach (OCA). The proposed method first applies an ordinal clustering approach to the survey units to identify homogeneous segments. These segments are then incorporated into a multi-year asphalt pavement M&R optimization decision-making model to determine the M&R plan. Data from 2022 covering 15.9 km of continuous asphalt pavement on the Donglv Highway in Shanxi Province were selected for analysis. These results demonstrate: 1) the OCA exhibits superior decision-making accuracy compared with the CDA; 2) compared with the M&R plans for survey units and CDA for homogeneous segments, the proposed method's M&R plan can be resolved faster and see fewer work zones, all achieved with a similar level of investment and meeting performance improvement; and 3) the M&R plan generated using the proposed method requires lower M&R investment but achieves higher performance compared with the actual M&R plan. These findings validate the effectiveness of the proposed method in producing improved M&R plans.

Engaging psychologically or behaviourally? A segmentation of the integrated resort customers

Academic studies that highlight the importance of customer engagement for sustaining organisational success have generally assumed that engaged customers are a homogenous market segment. This study provides an alternative scholarly perspective by reconsidering the multidimensionality of customer engagement and uses cognitive, affective, and behavioural elements to segment customers. A cluster analysis with data from 574 integrated resort customers in Macau, China supports the three-dimensional structure, and additional analyses show that customers can be segmented into totally-engaged, shallowly-engaged, potentially-engaged, and non-engaged groups based on their psychological investment (which consists of cognitive and affective engagements) and behavioural engagement. The findings both address the dimensional structure of customer engagement in the market segmentation context and provide some valuable theoretical and market segment-related managerial implications.

An empirical study on train choice preferences of high-speed railway passengers: the case of Nanning-Guangzhou railway and Nanning-Beihai railway

ABSTRACT This paper presents a systematic investigation of the train choice preference heterogeneity of High-Speed Railway (HSR) passengers under market segmentation to understand their train choice comprehensively. A stated preference survey was conducted for the Nanning-Guangzhou Railway and Nanning-Beihai Railway. Latent Class Analysis (LCA) was employed to identify homogeneous subgroups and segment the passenger market of each line into three segments: private travelers with a long total duration (PTLTD), business travelers (BT), and private travelers with a short total duration (PTSTD). Mixed Logit (ML) models were constructed for each subgroup sample to assess passengers' preferences in train choice. The results show that each class exhibited unique characteristics and preferences, and train fare and running time, departure date and time, and train frequency were statistically significant factors affecting train choice. This study can furnish theoretical and decisional support for HSR operators to design train operating schemes and flexible fare systems.

A new approach to dust source mapping using visual interpretation and object-oriented segmentation of satellite imagery

The emission of dust particles, mainly from arid and semi-arid lands, as a result of climate change and human activities, is known to be a global issue. Identifying dust emission sources is the first key step in dealing with the hazardous consequences of this rising phenomenon. This study is an attempt to address one of the major challenges in mapping dust emission sources. Accordingly, an innovative approach based on visual interpretation of multi-temporal MODIS-Terra/Aqua imagery and object-oriented image segmentation techniques has been developed and implemented in the study areas of the Tigris and Euphrates basin and eastern Iran. This approach takes advantage of land surface characteristics (i.e., dust drivers), including geomorphology, soil, land use/cover, and land surface radiation, to attribute dust emission hotspots to their corresponding areas using multi-source remote sensing data. The results show that the multi-resolution segmentation algorithm with optimized parameters can identify homogeneous segments corresponding to dust emission sources in the study areas with an average spatial agreement of ∼92% compared to the reference areas. Our findings emphasize the robustness and generalizability of the proposed approach, and its capabilities can be used in a complementary way with visual interpretation of satellite images to map dust sources with high spatial accuracy.

Graph convolutional LSTM algorithm for real-time crash prediction on mountainous freeways

Accurate real-time traffic crash prediction is crucial for proactive traffic safety management. Currently, the majority of real-time models forecast crashes every 5 min to support different intelligent transportation systems. However, these intervals might be too short for practical use in manually implementing proactive traffic safety measures such as deploying traffic law enforcement and emergency rescue resources. Therefore, this study develops hourly crash prediction models to provide network operators with sufficient time to take measures in advance. A section of a mountainous freeway in Guizhou province is divided into homogeneous segments, with crash data, traffic operations data, and meteorological data being collected hourly. As the result is an imbalanced dataset of crash and non-crash instances, the training dataset is resampled using synthetic minority over-sampling technique (SMOTE) to address the issue. To fully capture the complex spatiotemporal relationships in the data and achieve high crash prediction accuracy, a graph convolutional network-long short-term memory (GCN-LSTM) model is constructed for the first time, combining a graph convolutional network (GCN) and long short-term memory (LSTM) neural network. For comparison purposes, LSTM, Extreme Gradient Boosting (XGBoost), and logistic regression (LR) models are developed. The results show that the GCN-LSTM model outperforms other models in hourly traffic crash prediction, and the optimal prediction performance is achieved with the crash-to-non-crash ratio of 1:4. The GCN-LSTM method is found to effectively capture the complex spatiotemporal relationships in prediction data and to handle imbalanced traffic crash data.

InAs–InP Superlattice Nanowires with Tunable Phonon Frequencies

AbstractThe control of heat conduction through the manipulation of phonons in solids is of fundamental interest and can be exploited in applications for thermoelectric conversion. In this context, the advent of novel semiconductor nanomaterials with high surface‐to‐volume ratio, e.g. nanowires, offer exciting perspectives, leading to significant leaps forwarding the efficiency of solid‐state thermoelectric converters after decades of stagnation. Beyond the high aspect ratio, the nanowire geometry offers unprecedented possibilities of materials combination and crystal phase engineering not achievable with 2D counterparts. In this work, the growth of long (up to 100 repetitions) wurtzite InAs/InP superlattice nanowires with homogeneous segment thicknesses is reported, with control down to the single digit of nanometer. By means of Raman scattering experiments, clear modifications of the phonon dispersion in superlattice nanowires are found, where both InAs‐like and InP‐like modes are present. The experimentally measured modes are well reproduced by density functional perturbation theory calculations. Remarkably, it is found that the phonon frequencies can be tuned by the superlattice periodicity, opening exciting perspectives for phonon engineering and thermoelectric applications.

Evaluation of models for estimating free-flow speed on two-lane rural highways in Brazil

The objective of this article was to evaluate different models found in the literature to estimate the free-flow speed (FFS) on Brazilian two-lane rural highways. The models available in three editions of the Highway Capacity Manual (HCM) and two adaptations of the manual were evaluated to reproduce the behavior found on Brazilian highways. This evaluation was carried out based on traffic data collected in nine homogeneous segments of a Brazilian highway (MG-050) in the state of Minas Gerais. The FFS results estimated with the analytical methods were compared with the FFS data observed directly in the field. It was identified that there is no correlation between the FFS values produced by the analytical methods and the values found in the field. However, two models, one HCM2000 and another HCM2022, adapted to Brazilian traffic conditions, have similar average FFS values to those observed in the field.

Perceptions from Member-Consumers of a University Community for Sustainable and Healthy Eating: Evidence from Greece

Healthy eating is a vital component of a sustainable daily life, especially after the COVID-19 pandemic. In this study, we investigated the perceptions and attitudes of the members of a Greek university community, especially the young members, on: (i) health itself, (ii) healthy eating, and (iii) healthy eating campaigns today. The research was conducted electronically in the last three months of 2023 via a questionnaire survey and 1046 member-consumers participated. Statistical analysis, including descriptive and cluster analysis to group respondents into homogenous segments, was performed by employing a one-way ANOVA. The highlights of the results indicate that consumers perceive physical and mental health and energy efficiency as health preconditions, while they perceive a balanced and sustainable diet as the main parameter of healthy eating. They are influenced mostly by doctors and health providers and use information mostly based on common sense and their upbringing for their healthy eating choices. The cluster analysis revealed two distinctive groups of consumers categorized by this study as the “approachables” (54.5%), affected by today’s sustainable knowledge and concepts on healthy eating, and the “conservatives” (45.5%), who are reluctant to accepted these as such.

Machine Learning-Assisted Speech Analysis for Early Detection of Parkinson's Disease: A Study on Speaker Diarization and Classification Techniques.

Parkinson's disease (PD) is a neurodegenerative disorder characterized by a range of motor and non-motor symptoms. One of the notable non-motor symptoms of PD is the presence of vocal disorders, attributed to the underlying pathophysiological changes in the neural control of the laryngeal and vocal tract musculature. From this perspective, the integration of machine learning (ML) techniques in the analysis of speech signals has significantly contributed to the detection and diagnosis of PD. Particularly, MEL Frequency Cepstral Coefficients (MFCCs) and Gammatone Frequency Cepstral Coefficients (GTCCs) are both feature extraction techniques commonly used in the field of speech and audio signal processing that could exhibit great potential for vocal disorder identification. This study presents a novel approach to the early detection of PD through ML applied to speech analysis, leveraging both MFCCs and GTCCs. The recordings contained in the Mobile Device Voice Recordings at King's College London (MDVR-KCL) dataset were used. These recordings were collected from healthy individuals and PD patients while they read a passage and during a spontaneous conversation on the phone. Particularly, the speech data regarding the spontaneous dialogue task were processed through speaker diarization, a technique that partitions an audio stream into homogeneous segments according to speaker identity. The ML applied to MFCCS and GTCCs allowed us to classify PD patients with a test accuracy of 92.3%. This research further demonstrates the potential to employ mobile phones as a non-invasive, cost-effective tool for the early detection of PD, significantly improving patient prognosis and quality of life.

Redefining green consumerism: a diminutive approach to market segmentation for sustainability.

The purpose of this study is to accelerate green consumerism efforts by assisting green marketers in identifying the homogeneous and significant eco-friendly customer segments emerging in India. The study determines the antecedents driving customers to purchase green products, which can be leveraged while targeting the studied customer group and designing promotional strategies for these microgreen segments. Non-probability criterion-based sampling technique was used in collecting the data across Pan India through various online platforms like LinkedIn, Twitter, and Facebook. The exploratory factor analysis, followed by the cluster and discriminant analysis, is conducted for inferential results. The results reveal eight major factors influencing green consumer behavior, out of which green habit, green culture awareness and attitude, interpersonal influence, and green purchase intention/behavior emerged as the most significant factors. The study establishes two important clusters of green consumers, that is, "Green Dads"-generation Y males and "Green Janes"-generation Z females, with social media marketing and subjective norms as the most influential factors in discriminating between these two clusters. This research magnifies the importance of profiling customers based on demographics, psychographics, behavioral variables, and external marketing cues. The varied combinations of factors concerning ecological behavior imply the focus on micro variables by marketers, hence promulgating the projected granularity of green market segmentation and consequential consumer behavior.

Conditional spatial biased intuitionistic clustering technique for brain MRI image segmentation.

In clinical research, it is crucial to segment the magnetic resonance (MR) brain image for studying the internal tissues of the brain. To address this challenge in a sustainable manner, a novel approach has been proposed leveraging the power of unsupervised clustering while integrating conditional spatial properties of the image into intuitionistic clustering technique for segmenting MRI images of brain scans. In the proposed technique, an Intuitionistic-based clustering approach incorporates a nuanced understanding of uncertainty inherent in the image data. The measure of uncertainty is achieved through calculation of hesitation degree. The approach introduces a conditional spatial function alongside the intuitionistic membership matrix, enabling the consideration of spatial relationships within the image. Furthermore, by calculating weighted intuitionistic membership matrix, the algorithm gains the ability to adapt its smoothing behavior based on the local context. The main advantages are enhanced robustness with homogenous segments, lower sensitivity to noise, intensity inhomogeneity and accommodation of degree of hesitation or uncertainty that may exist in the real-world datasets. A comparative analysis of synthetic and real datasets of MR brain images proves the efficiency of the suggested approach over different algorithms. The paper investigates how the suggested research methodology performs in medical industry under different circumstances including both qualitative and quantitative parameters such as segmentation accuracy, similarity index, true positive ratio, false positive ratio. The experimental outcomes demonstrate that the suggested algorithm outperforms in retaining image details and achieving segmentation accuracy.

Understanding Italian consumers' perceptions of tomato agricultural innovation: Exploring the nexus between sustainability, health and consumer beliefs

The growing emphasis on sustainability and health concerns has underscored the necessity for low-pesticide and safe foods, leading to increased adoption of organic production process and Integrated Pest Management (IPM). From this concept emerged the “Zero Residue" certification, while the “Nickel-free" claim serves health-related purposes. These attributes are related to consumer beliefs and, along with intrinsic quality cues such as colour, intervene in consumer decision-making process and result from technological and varietal agricultural innovation. They were evaluated alongside price to examine consumer preferences, using tomatoes as a case study. Conjoint Analysis and Ward's clustering method helped segregate Italian consumers into homogenous market segments based on their stated preferences. Organic production was the most significant attribute, followed by price, the “Nickel-free” claim, and colour. Thus, consumer demand increasingly favours environmentally sustainable and health-focused attributes. The “Zero Residue" concept offers an approach to reduce synthetic fertilizers and pesticides, appealing to environmental conscious consumers. Attributes like “Nickel-free" claim have emerged, which is important for consumers seeking innovative health features. Businesses should diversify products and employ clear labelling to inform consumers about eco-friendly practices. They should also offer competitive prices for innovative tomato products, thus implementing price segmentation may be necessary for diverse consumer segments. Policymakers can engage consumers through educational campaigns, enforce product labelling regulations highlighting health and environmental benefits, and support certification programs.

Forest stand delineation using airborne LiDAR and hyperspectral data

Forest stands, crucial for inventory, planning, and management, traditionally rely on time-consuming visual analysis by forest managers. To enhance efficiency, there is a growing need for automated methods that take into account essential forest attributes. In response, we propose a novel approach utilizing airborne Light Detection and Ranging (LiDAR) and hyperspectral data for automated forest stand delineation. Our approach initiates with over-segmentation of the Canopy Height Model (CHM), followed by attribute calculation for each segment using both CHM and hyperspectral data. Two rules are applied to merge homogeneous segments and eliminate others based on calculated attributes. The effectiveness of our method was validated using three types of reference forest stands with two indices: the explained variance (R2) and Intersection over Union (IoU). Results from our study demonstrated notable accuracy, with a R2 of 97.35% and 97.86% for mean tree height and mean diameter at breast height (DBH), respectively. The R2 for mean canopy height is 81.80%, outperforming manual delineation by 7.31% and multi-scale segmentation results by 2.13%. Furthermore, our approach achieved high IoU values, which indicates a strong spatial agreement with manually delineated forest stands and leading to fewer manual adjustments when applied directly to forest management. In conclusion, our forest stand delineation method enhances both internal consistency and spatial accuracy. This method contributes to improving practical performance and forest management efficiency.