Table 1 - uploaded by Sachit Rajbhandari
Content may be subject to copyright.
![Descriptions of landslides classes sourced from [28].](profile/Sachit-Rajbhandari/publication/321354869/figure/tbl1/AS:668811900489739@1536468591681/Descriptions-of-landslides-classes-sourced-from-28.png)
Source publication
![Table 1 . Descriptions of landslides classes sourced from [28].](profile/Sachit-Rajbhandari/publication/321354869/figure/tbl1/AS:668811900489739@1536468591681/Descriptions-of-landslides-classes-sourced-from-28_Q320.jpg)
In Geographic Object-based Image Analysis (GEOBIA), identification of image objects is normally achieved using rule-based classification techniques supported by appropriate domain knowledge. However, GEOBIA currently lacks a systematic method to formalise the domain knowledge required for image object identification. Ontology provides a representat...
Contexts in source publication
Context 1
... this work, five different types of landslides are identified based on their spectral, contextual and morphometric characteristics. The landslide types and their description with feature criteria are sourced from [28] and provided in Table 1. In this table, the first column is landslide class; the second column is a description in the natural language; and the third column is the translation of these descriptions into measurable thresholds. ...
Context 2
... addition, an identification of key terminology used in our domain of interest is a vital step. Table 1 defines different types of landslides. From these definitions, firstly we identify different terms for landslide types. ...
Context 3
... develop such rules, expert domain knowledge must be extracted from domain experts or the literature. In our case, we have extracted these rules from the definition of the landslide classes (Table 1). The definition of a debris slide states that they are found in thickly covered soil of moderate slope and low length. ...
Citations
... The authors developed a spatial-relational ontology ( Figure 4) which was used to semantically interpret these spatial objects. There followed field validation of the interpretation and the final classification (see Rajbhandari et al., 2017;Argyridis and Argialas, 2019). Methods combining spatial and thematic semantics have only recently started to be developed to address complex geospatial features . ...
... Random Forest (RF) is an ensemble ML method that has been well-developed for classification, regression, and other tasks [51]. This method has some advantages, including great robustness, data adaptability, and low overfitting [52]. ...
... Characteristic itself is defined as: an abstraction of a property of an object or of a set of objects. More individually, an ontology is a representation of some preexisting domain of reality (Rajbhandari et al. 2017) which: ...
Modeling natural hazards in 3D space constitute a significant step for managing and planning our living environment. The creation of accurate maps is needed to document the impact of natural hazards such as landslides. Loss of life, natural resources or property transform landslide phenomenon to a natural disaster. In landslide analysis different factors can be incorporated and studied such as landslides occurrence and occurrence, their distribution, mechanisms, pattern of failures. The development of reliable maps is also crucial for determining landslide susceptibility and risk. Late years, the emerging geospatial technologies are able to produce different types of 2D and 3D data. Unmanned Aerial Vehicle (UAV) or Unmanned Aerial Systems (UAS), support the acquisition of ultra-high detailed geospatial data in the 3D environment. Those systems are flexible in data acquisition, with a high temporal frequency, while it is limited for site specific mapping purposes. The exploitation of 3D point-clouds has been proven tremendously efficient for analyzing data in the field of geoscience. Point cloud advantages of documenting in 3D space, data of hazardous sites at low cost and effective performance identifies them as leading primitives for site-specific 3D landslide modelling. Given the gaps between the computer vision capabilities and their applications in landslide assessment in site-specific scale, the thesis aims at developing a general framework of predefined workflows in an object-based programming environment for detection and characterization of landslide phenomena from ultra-high-resolution UAV-derived data. The proposed framework is built up in four distinct research phases: (a) on-site data collection, (b) data preprocessing, (c) OBIA (segmentation and classification), and (d) evaluation. These phases result in various novel component-wise solutions, which particular focus on the optimization phase of OBIA for landslide assessment. Different flight acquisition configurations were tested by varying the number of images, image overlap, flight height and focal length for selecting the optimal workflow for imagery collection always considering the site specifications (topography, landslide mechanism). Each configuration was processed independently with dedicated photogrammetric software following the same template for subjective evaluation. Structure from Motion (SfM) photogrammetry has been used to provide dense 3D point clouds describing surface morphology of landslide environments. An object-based classification approach of the photogrammetric point cloud products into homogeneous and spatially connected elements has been executed. The proposed methodology has been developed based on Object-Based Image Analysis (OBIA) and fusion of multivariate data resulted from photogrammetric processing in order to take full advantage of its productivity. The focus of modeling applications was particularly on landslide with rotational and translational mechanisms of failure. A critical comparative study was conducted to analyze the influence of topographic information, scale segmentation and evaluate the object-based classification of landslide ontologies with three state-of-the-art Machine Learning classifiers, KNN, DT and RF with the inclusion of spectral, spatial, and contextual characteristics. Results highlight higher performances for landslide mapping with RF when DSM information was integrated. Thus, RF presented higher predictive performance when the model was fitted and applied to a different study area. For the ML classification of landslide zones, 60% of the reference segments have been used for training and 40% for validation of the models. The proposed thesis illustrates the effectiveness and efficiency of UAV platforms to acquire accurate photogrammetric datasets from complex surface topographies and provide an efficient and transferable object-based framework to characterize the failure site based on semantic classification of the landslide elements. The outcome can be useful for prioritizing efforts to moderate the adverse consequences of landslides and provide future mitigation strategies following landslide ontologies. UAV-based landslide modelling on the investigated sites provided an amount of undiscovered knowledge for landslide elements. Complementary to the developed workflow the accomplished real-world application, this work has shown the great potential of coupling UAV photogrammetry with object-based methods for assessing the landslide features in different hierarchical scales and provide a detailed automatic classification. In the future, the developed methodology will be further extended and tested on diverse landslide mechanisms, also including engineering geological attributes collected during the field reconnaissance.
... Finally, the transition matrices were transformed into percentages of land-use cover, allowing the creating of Chord-Diagrams. The Chord-Diagram is a graphical method which allows the visual representation of the inter-relationships between data in a transition matrices (Komárek et al., 2018;Malavasi et al., 2018;Rajbhandari et al., 2017). To show the land-use cover transitions, the arrows represent the direction of the variation, while the width the percentage of the change. ...
Today’s Mediterranean landscapes result from the interaction between ecosystems and anthropogenic activities. This study aims to assess how the land-use changes between the mid-nineteenth and end of the twentieth century influenced the temporal continuity of the ecosystems in central Apennines (central Italy). Information was acquired from Gregorian cadastral maps, orthophotos and aerial photos (1850, 1954, 1980 and 2010), digitised and georeferenced using QGIS 3.10.1 software. Marked changes in land-use types were found. From 1850 to 1954, grasslands were widely transformed into arable lands, but in the next 60 years they changed again into new grasslands and forests. Forests underwent a slow but continuous expansion from 1850 to 2010. Only a small percentage of the forest and grassland patches (14 and 16%, respectively) have seen ecological continuity. These considerations call attention to temporal continuity of ecosystems, together with the historical dynamism of landscapes, in defining land management and nature conservation policies.
... The authors developed a spatial-relational ontology ( Figure 4) which was used to semantically interpret these spatial objects. There followed field validation of the interpretation and the final classification (see Rajbhandari et al., 2017;Argyridis and Argialas, 2019). Methods combining spatial and thematic semantics have only recently started to be developed to address complex geospatial features . ...
... In SWRL, there are two components of the rules 1) an antecedent, 2) consequent, the two components consisting of a set of atoms. The atoms can be shaped , where the OWL description is , the OWL property is , is a built-in relation, and and are two variables [83]. The consequent as well as an antecedent are written as . ...
One of the main risks to food security is plant diseases, but because of the absence of needed infrastructure and actual noise, scientists are faced with a difficult issue. Semantic segmentation of images divides images into non-overlapped regions, with specified semantic labels allocated. In this paper, The QPSO (quantum particle swarm optimization) algorithm has been used in segmentation of an original noisy image and Ontology has been used in classification the segmented image. Input noisy image segmentation is limited to a classification phase in which the object is transferred to Ontology. With 49,563 images from healthy and diseased plant leaves, 12 plant species were identified and 22 diseases, the proposed method is evaluated. The method proposed produces an accuracy of 86.22 percent for a stopped test set, showing that the strategy is appropriate. EPDO (Enhance Plant Disease Ontology) is built with the web ontology language (OWL). The segmented noisy image elements are paired with EPDO with derived features that come from QPSO. Our results show that a classification based on the suggested method is better than the state-of-the-art algorithms. The proposed method also saves time and effort for removing the noise at noise level from the input image σ=70
... Several studies have demonstrated the importance of semantics to understanding landscape ecology as just one of several articles published on the topic by Ahlqvist and Shortridge (2010) described the use of quantitative semantic similarity metrics for evaluating landscape heterogeneity. Since that time, interest in semantic web and ontology construction has led to some papers on the organization and standardization of land cover information like (Comber et al. 2015) continued work on citizen science applications and (Rajbhandari et al. 2017) work on the integration of ontological data structures for Geographic Object-Based Image Analysis. ...
ContextThe observation of the earth by humans has advanced our understanding of the physical patterns and processes that shape the landscape. Over time, the act of scientific interpretation has transformed into one mediated through machines, creating distance between the observer and the observed. Machine learning is expanding this gap and transforming how we gain knowledge about the world. Raising the question is there something to be lost by advancing machine learning at the expense of human visual interpretation?Objectives
Recognizing the usefulness of these computational algorithms for dealing with massive, heterogeneous, and dynamic ecological datasets, scientists should not abandon the important contributions of human intelligence to understanding landscape patterns, processes, and relationships.Methods
This paper presents a review of social, cultural, and political or military influences on the relationship between humans and remote sensing images of the landscape. This review highlights tensions between automated machine learning approaches and human interpretation.ResultsSupport for the use of human–machine integrated systems through the use of interactive, visual display, and the development of transparent machine learning methods is suggested.Conclusions
The human analyst should remain central in the design of landscape ecology applications when deploying machine learning algorithms. The complementary strengths of the human and machine in data processing suggest that the most informative insights regarding pattern and process can happen in the implementation of carefully designed Human in the Loop systems.
... A co-citation and cluster analysis of subjects, journals, and authors based on 1533 articles published between 2001 and 2016, has shown that ontologies are prominent in GIScience research across several disciplines (computer science, engineering, geography, geosciences, etc.) [12]. Ontologies have been employed for solving various research problems such as GEographic Object-Based Image Analysis (GEOBIA) [13][14][15], information extraction and retrieval [16,17], information integration [18,19], linked data [20][21][22] and geoprocessing workflows [23], geospatial data provenance on the web [24], interpretation of natural language descriptions [25], automatic feature recognition from point clouds [26], and sketch map interpretation [27]. ...
... In the context of foreshore identification, Argyridis and Argialas [15] designed an ontology to formalize the implicit spectral, geometric, and spatial relationships described in the interpretation criteria, and employ them during identification. Rajbhandari et al. [14] also emphasize that ontologies allow for a better modularization of the methods: Common knowledge can be reused together with features more specific to an application. They also facilitate the automation of the classification since knowledge is transferred minimizing human intervention. ...
... However, as mentioned in [56], vagueness can be of two kinds: conceptual vagueness and threshold (or sorite) vagueness. Ontologies provide a way to handle the first one while, in [14] the second is handled with machine learning approaches. ...
The present paper provides a review of two research topics that are central to geospatial semantics: information modeling and elicitation. The first topic deals with the development of ontologies at different levels of generality and formality, tailored to various needs and uses. The second topic involves a set of processes that aim to draw out latent knowledge from unstructured or semi-structured content: semantic-based extraction, enrichment, search, and analysis. These processes focus on eliciting a structured representation of information in various forms such as: semantic metadata, links to ontology concepts, a collection of topics, etc. The paper reviews the progress made over the last five years in these two very active areas of research. It discusses the problems and the challenges faced, highlights the types of semantic information formalized and extracted, as well as the methodologies and tools used, and identifies directions for future research.
... The atoms can be shape C(x), P(x, y), sameAs(x, y)differentFrom(x, y), or builtIn(r, x, . . . ), where the OWL description is , the OWL property is , is a built-in relation, and and are two variables [39]. The consequent as well as an antecedent are written as 1 ∧ 2 … ∧ . ...
... Önerilen yaklaşımın ikinci aşamasında yer yüzeyi ile yer yüzeyi olmayan nesnelerin birbirinden ayrılması işlemi [38][39][40]. Ontoloji entegrasyon süreçlerinin gerçekleştirilmesinde makine öğrenmesi ve doğal dil işleme süreçlerinin de göz önüne alınması gerekmektedir [41,42]. Makine öğrenmesi ve doğal dil işlemenin gerçekleştirilebilmesi için anlamlı bir ayrıntı düzeyinde ilgilenilen alan için kavramsal ve bağlamsal bilgileri kapsayan semantik bilgilerin de sürece dâhil edilmesi gerekmektedir. ...
Kent yönetiminde, yaşam alanlarına ait problemlerin çözümü, sağlıklı ve sürdürülebilir kentlerin oluşturulması, akıllı şehirlerin altyapısının kurulması gibi amaçlar için mekânsal bilgi içeren verilerden yararlanılmaktadır. Bu sebeple mekansal verilerin toplanması, işlenmesi, değerlendirilmesi ve bilgiye dönüştürülmesi kent yöneticilerinin hızlı ve doğru kararların verilebilmesi için önem arz etmektedir. Son yıllarda, mekansal verilerin değerlendirilmesi çalışmalarında, obje çıkarım tekniklerinin geliştirilmesi ve optimize edilmesi için farklı yöntem ve algoritmalar geliştirilmiştir. Ancak bu çalışmalarda kullanılan mekansal veriler, çoğunlukla farklı veri kaynaklarından elde edilmesi sebebiyle farklı teknik özelliklere (geometrik, radyometrik, zamansal çözünürlük, vb.) sahip veriler olduğundan, mekansal semantik kavramı özelinde heterojen bir yapı göstermektedir. Bu heterojen yapı uzman bilgisinin kavramsallaştırılması, birlikte çalışabilirlik ve yeniden kullanılabilirlik konularında problemler oluşturmaktadır. Ontoloji, uzman bilgisinin kavramsallaştırılarak semantik olarak tam açıklanmış ve birbirleri ile bağlı bir yapı sunması sebebi ile obje çıkarımı çalışmalarında heterojenlikten kaynaklanan sorunların giderilmesinde güncel araştırma konusu haline gelmiştir. Bu çalışmada, Kırklareli ili, Evrencik bölgesine ait LiDAR sistem verileri kullanılarak ontoloji destekli obje çıkarımı hedeflenmiştir. Bu amaç doğrultusunda obje tabanlı görüntü analiz yöntemi, bulanık mantık ile sınıflandırma kullanılarak obje çıkarımı yapılmış ve kavramsal sınıf tanımları, obje ve veri ilişkileri, kurallar ve aksiyomlar tanımlanarak semantik altyapı modeli kurulmuştur. Bu çalışmanın sonucunda doğruluk analizi ve görüntü objelerinin ontoloji ile entegrasyonu yapılmıştır.
