scholarly journals Semi-Supervised Classification and Landscape Metrics for Mapping and Spatial Pattern Change Analysis of Tropical Forest Types in Thua Thien Hue Province, Vietnam

Forests ◽  
2019 ◽  
Vol 10 (8) ◽  
pp. 673
Author(s):  
Tuong ◽  
Tani ◽  
Wang ◽  
Thang
Keyword(s):  
Spatial Pattern ◽  
Landscape Metrics ◽  
Remotely Sensed ◽  
Natural Forest ◽  
Least Square ◽  
Future Research ◽  
Forest Transition ◽  
Forest Types ◽  
Remotely Sensed Data ◽  
Forest Classification

Research Highlights: In this study, we classified natural forest into four forest types using time-series multi-source remotely sensed data through a proposed semi-supervised model developed and validated for mapping forest types and assessing forest transition in Vietnam. Background and Objectives: Data on current forest state and changes detection are always essential for forest management and planning. There is, therefore, a need for improved tools to classify and evaluate forest dynamics more accurately and effectively. Our objective is to develop such tools using a semi-supervised model and landscape metrics to classify and map changes in natural forest types by using multi-source remotely sensed data. Materials and Methods: A combination of Landsat data with PALSAR and PALSAR-2 was used for forest classification through the proposed semi-supervised model. This model turned a kernel least square into a self-learning algorithm, trained by a small number of samples with given labels, and then used this classifier to assign labels to the unlabeled data. The overall accuracy, kappa, user’s accuracy, and producer’s accuracy were used to evaluate the classification accuracy by comparing the classified image with the results of ground truth interpretation. Based on the classified images, forest transition was evaluated using certain landscape metrics at the class and landscape levels. Results: The multi-source data approach achieved improved discrimination of forest types compared to only using single data (optical or radar data). Good classification accuracies were obtained, with kappas of 0.81, 0.76, and 0.74 for the years 2007, 2010, and 2016, respectively. The analysis of landscape metrics indicated that there were different behaviors in the four forest types, as well as provided much information about the trends in spatial pattern changes. Conclusions: This study highlights the utilization of a semi-supervised model in forest classification, and the analysis of forest transition using landscape metrics. However, future research should include a comparison of different models to estimate the improvement of the proposed model. Another important study that should be conducted is to test the proposed method on larger areas.

scholarly journals Combining remotely sensed data using aggregation algorithms

1998 ◽  
Vol 2 (2/3) ◽  
pp. 149-158 ◽  
Author(s):  
W. J. Shuttleworth
Keyword(s):  
Surface Energy ◽  
Land Surface ◽  
Remotely Sensed ◽  
Surface Radiation ◽  
Control Surface ◽  
Future Research ◽  
Practical Implementation ◽  
Remotely Sensed Data ◽  
Theoretical Concepts ◽  
Average Value

Abstract. This paper describes a strategic approach for providing documentation of the surface energy exchange for heterogeneous land surfaces via the simultaneous, four-dimensional assimilation of several streams of remotely sensed data into a coupled land surface-atmosphere model. The basic concepts and underlying theory behind this proposed approach are presented with the intent that this will guide, facilitate, and stimulate future research focused on its practical implementation when appropriate data from the Earth Observing System (EOS) become available. The theoretical concepts that underlie the approach are derived from relationships between the values of parameters which control surface exchanges at pixel (or patch) scale and the area-average value of equivalent parameters applicable at larger, grid scale. A three-step implementation method is proposed which involves (a) estimating grid-average surface radiation fluxes from appropriate remotely sensed data; (b) absorbing these radiation flux estimates into a four-dimensional data assimilation model in which grid-average values of vegetation-related parameters are calculated from pertinent remotely sensed data using the equations that link pixel and grid scales; and (c) improving the resulting estimate of the surface energy balance-again using scale-linking equations by estimating the effect of soil-moisture availability, perhaps assuming that cloud-free pixels are an unbiased subsample of all the pixels in the grid square.

Landscape fragmentation

2010 ◽  
Author(s):  
Ned Horning ◽  
Julie A. Robinson ◽  
Eleanor J. Sterling ◽  
Woody Turner ◽  
Sacha Spector
Keyword(s):  
Forest Fragmentation ◽  
Landscape Metrics ◽  
Forest Cover ◽  
Remotely Sensed ◽  
Clear Understanding ◽  
Remotely Sensed Data ◽  
Deforestation Rates ◽  
Habitat Conversion ◽  
Conservation Organizations ◽  
Total Core Area

The country of Vietnam has long been recognized as an important region for biodiversity (Sterling et al. 2006). High-profile discoveries in the 1990s of many species new to science including large ones such as the Saola (Pseudoryx nghetinhensis), an 85 kg basal member of the cattle subfamily Bovinae and the first new genus of large land-dwelling mammal described since the okapi (Okapia johnstoni) in 1901, have focused the attention of national and international conservation organizations on Vietnam and surrounding countries in mainland Southeast Asia (Hurley et al. in prep.). Conservation action for these endemic, endangered species relies on a clear understanding of trends in habitat conversion. To track deforestation rates through time in Vietnam, Meyfroidt and Lambin (2008) combined remotely sensed data with landscape metrics such as number of patches, mean patch size, mean proximity index, and total core area index. They tested their analyses across a variety of land cover studies including those using Advanced Very High Resolution Radiometer (AVHRR), Landsat, SPOT, and MODIS data sources. They found that forest cover decreased nationally from the 1980s to the 1990s and then showed an increase between 1990 and 2000, due to plantation forests as well as natural forest regeneration. However, the effects of this forest transition on fragmentation metrics noted above differed across the country. For instance, in some places, such as central Vietnam where forest cover is relatively large and well connected, reforestation led to a decrease in forest fragmentation and secondary forests recovered rapidly. However in others, such as areas in the north where forest fragmentation dates back centuries and forests have therefore long been isolated, reforestation did not seem to have an impact on continued fragmentation and habitat loss. In this chapter we detail the importance of fragmentation and landscape metrics to ecology and conservation, outlining when and where remotely sensed data can help in these analyses. We then discuss a subset of fragmentation metrics and point to some challenges in processing fragmentation data. We provide examples of composition and connectivity metrics illuminated with examples from the remote sensing literature.

scholarly journals Application of random forest classification and remotely sensed data in geological mapping on the Jebel Meloussi area (Tunisia)

2021 ◽  
Vol 14 (21) ◽  
Author(s):  
Gáspár Albert ◽  
Seif Ammar
Keyword(s):  
Random Forest ◽  
Remotely Sensed ◽  
Geological Mapping ◽  
Vegetation Coverage ◽  
Remotely Sensed Data ◽  
Random Forest Classification ◽  
Radar Images ◽  
Forest Classification ◽  
Geological Map ◽  
Sentinel 2

Abstract Remotely sensed data such as satellite photos and radar images can be used to produce geological maps on arid regions, where the vegetation coverage does not have a significant effect. In central Tunisia, the Jebel Meloussi area has unique geological features and characteristic morphology (i.e. flat areas with dune fields in contrast with hills of folded and eroded stratigraphic sequences), which makes it an ideal area for testing new methods of automatic terrain classification. For this, data from the Sentinel 2 satellite sensor and the SRTM-based MERIT DEM (digital elevation model) were used in the present study. Using R scripts and the random forest classification method, modelling was performed on four lithological variables—derived from the different bands of the Sentinel 2 images—and two morphometric parameters for the area of the 1:50,000 geological map sheet no. 103. The four lithological variables were chosen to highlight the iron-bearing minerals since the spectral parameters of the Sentinel 2 sensors are especially useful for this purpose. The training areas of the classification were selected on the geological map. The results of the modelling identified Eocene and Cretaceous evaporite-bearing sedimentary series (such as the Jebs and the Bouhedma Formations) with the highest producer accuracy (> 60% of the predicted pixels match with the map). The pyritic argillites of the Sidi Khalif Formation were also recognized with the same accuracy, and the Quaternary sebhkas and dunes were also well predicted. The study concludes that the classification-based geological map is useful for field geologist prior to field surveys.

scholarly journals Road Extraction from Remotely Sensed Data: A Review

2021 ◽  
Author(s):  
Mohd Jawed Khan ◽  
Pankaj Pratap Singh
Keyword(s):  
Deep Learning ◽  
Traffic Management ◽  
Remotely Sensed ◽  
Future Research ◽  
Learning Approaches ◽  
Road Extraction ◽  
Remotely Sensed Data ◽  
Remotely Sensed Imagery ◽  
Urban Rural ◽  
Future Research Directions

Up-to-date road networks are crucial and challenging in computer vision tasks. Road extraction is yet important for vehicle navigation, urban-rural planning, disaster relief, traffic management, road monitoring and others. Road network maps facilitate a great number of applications in our everyday life. Therefore, a systematic review of deep learning approaches applied to remotely sensed imagery for road extraction is conducted in this paper. Four main types of deep learning approaches, namely, the GANs model, deconvolutional networks, FCNs, and patch-based CNNs models are presented in this paper. We also compare these various deep learning models applied to remotely sensed imagery to show their performances in extracting road parts from high-resolution remote sensed imagery. Later future research directions and research gaps are described.

A forest classification based on management and treatment

1972 ◽  
Vol 30 ◽  
Author(s):  
M. Van Miegroet
Keyword(s):  
Forest Policy ◽  
Forest Stand ◽  
Natural Forest ◽  
Forest Types ◽  
Human Interference ◽  
Morphological Aspects ◽  
Silvicultural Treatment ◽  
Forest Classification ◽  
Marginal Forest

The  following forest types are used as basic units for classification:    1. The natural forest    2. The semi-natural forest    3. The intermediary forest    4. The artificial forest    5. The naturalized forest    6. The marginal forest forms    The proposed classification of forests tries to establish a systematic  order, based partly on morphological aspects of the forest stand, but  principally on the degree and the characteristics of human interference as  expressed by use, treatment and the aims of management. Its application  belongs essentially to the domain of forest policy. The number of types it  covers is not to be considered limitative: practical use of the  classification will give the opportunity to find out its weaker points and  eventually lead to necessary modifications.     It can be used simultaneously with other classifications based on  floristic, ecological, phytosoeiological and phytogeographical  characteristics. Neither is it intended as a substitute for usual stand  description, needed for planning management and silvicultural treatment,  because it does not take into consideration the particularities of the local  situation.

scholarly journals Remotely Sensed Tree Characterization in Urban Areas: A Review

2021 ◽  
Vol 13 (23) ◽  
pp. 4889
Author(s):  
Luisa Velasquez-Camacho ◽  
Adrián Cardil ◽  
Midhun Mohan ◽  
Maddi Etxegarai ◽  
Gabriel Anzaldi ◽  
...  
Keyword(s):  
Urban Areas ◽  
Urban Forestry ◽  
Remotely Sensed ◽  
Critical Discussion ◽  
Urban Trees ◽  
Future Research ◽  
Remotely Sensed Data ◽  
Individual Tree ◽  
Processing Methods ◽  
Urban Tree

Urban trees and forests provide multiple ecosystem services (ES), including temperature regulation, carbon sequestration, and biodiversity. Interest in ES has increased amongst policymakers, scientists, and citizens given the extent and growth of urbanized areas globally. However, the methods and techniques used to properly assess biodiversity and ES provided by vegetation in urban environments, at large scales, are insufficient. Individual tree identification and characterization are some of the most critical issues used to evaluate urban biodiversity and ES, given the complex spatial distribution of vegetation in urban areas and the scarcity or complete lack of systematized urban tree inventories at large scales, e.g., at the regional or national levels. This often limits our knowledge on their contributions toward shaping biodiversity and ES in urban areas worldwide. This paper provides an analysis of the state-of-the-art studies and was carried out based on a systematic review of 48 scientific papers published during the last five years (2016–2020), related to urban tree and greenery characterization, remote sensing techniques for tree identification, processing methods, and data analysis to classify and segment trees. In particular, we focused on urban tree and forest characterization using remotely sensed data and identified frontiers in scientific knowledge that may be expanded with new developments in the near future. We found advantages and limitations associated with both data sources and processing methods, from which we drew recommendations for further development of tree inventory and characterization in urban forestry science. Finally, a critical discussion on the current state of the methods, as well as on the challenges and directions for future research, is presented.

scholarly journals A Remote Sensing and GIS-Based Model of Habitat as a Predictor of Biodiversity

1994 ◽  
Vol 18 ◽  
pp. 3-16
Author(s):  
Diane Debinski ◽  
Kelly Kindscher
Keyword(s):  
Remote Sensing ◽  
Species Diversity ◽  
Species Distribution ◽  
Animal Species ◽  
Habitat Type ◽  
Distribution Patterns ◽  
Remotely Sensed ◽  
Forest Types ◽  
Remotely Sensed Data ◽  
Habitat Specificity

Conservation biologists need better methods for predicting species diversity. This research investigated some new methods to analyze biodiversity patterns through the use of Geographic Information Systems and remote sensing technologies. We tested the correlation between remotely sensed habitat types and species distributions. The goal was not to do away with ground-based fieldwork, but rather to optimize and focus fieldwork by using GIS and remotely sensed data as tools for making the work more accurate and specific. Our research was conducted at a fine (30 x30 m) landscape scale using on-the ground locations of birds, butterflies, and plants in the northwest portion of the Greater Yellowstone Ecosystem. Three remotely sensed forest types (distinguished by species density and coverage) and six remotely sensed meadow types (ranging from xeric to hydric) were surveyed and coverage data were collected for grasses, shrubs, forbs and trees. Presence/absence data were collected for birds and butterflies. The objectives of this research were: 1) to determine the extent of the correlation between spectral reflectance patterns and plant or animal species distribution patterns, and 2) to test the spatial correspondence of species diversity "hotspots" among taxonomic groups. Field surveys in 1993 and 1994 validated the vegetation density, cover, and moisture gradients expected from satellite data interpretation. Both tree species composition and diameter at breast height were significant in discriminating among forest types. Twenty-two species of grasses and forbs were significant in distinguishing among meadow types. However, a smaller percentage of the animal species was significantly correlated with one habitat type. In order to find a strong correlation between species distribution patterns and remotely sensed data, a species must be moderately common and show some habitat specificity. Hotspots of species diversity coincided for shrubs, grasses, forbs, birds, and butterflies and were found in mesic meadows.

Sign in / Sign up

Export Citation Format

Share Document