Combining QuickBird, LiDAR, and GIS topography indices to identify a single native tree species in a complex landscape using an object-based classification approach

Pham, Lien T.H.; Brabyn, Lars; Ashraf, Salman

doi:10.1016/j.jag.2016.03.015

Cited by 39 publications

(36 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Object-based image analysis (OBIA) has become increasingly popular for land cover classification over the last decade [32] and is proven to be economical and efficient at large scale through use of more effective, transparent, and repeatable analytical processes [33]. The updating approach has potential to update land cover dataset effectively [34,35], which integrate the post-classification and change detection approaches [36].…”

Section: Forest Mapping With An Updating and Object-based Image Analymentioning

confidence: 99%

Spatio–Temporal Changes of Forests in Northeast China: Insights from Landsat Images and Geospatial Analysis

Chunying

Chen

Wang

et al. 2019

Forests

View full text Add to dashboard Cite

Dramatic changes of forests have strong influence on regional and global carbon cycles, biodiversity, and ecosystem services. Understanding dynamics of forests from local to global scale is crucial for policymaking and sustainable development. In this study, we developed an updating and object-based image analysis method to map forests in Northeast China using Landsat images from 1990 to 2015. The spatio-temporal patterns of forests were quantified based on resultant maps and geospatial analysis. Results showed that the percentage of forested area occupying the entire northeast China was more than 40%, about 94% of initial forest cover remained unchanged (49.37 × 10 4 km 2 ) over the course of 25 years. A small net forest loss (1051 km 2 ) was observed during 1990-2015. High forest gain (10,315 km 2 ) and forest loss (9923 km 2 ) both occurred from 2010 to 2015. At the provincial level, Heilongjiang demonstrated the highest rate of deforestation, with a net loss of 1802 km 2 (0.89%). Forest changes along elevation, slope, and distance from settlements and roads were also investigated. Over 90% of forest changes occurred in plains and low mountain areas within the elevation of 200-1000 m and slope under 15 • . The most dramatic forest changes can be found within the distance of 2000 m from settlements and roads. The reclamation of sloping land, construction of settlements and roads, and possible smallholder clearing contributed more to forest loss, while ecological projects and related government policies play an important role on afforestation and reforestation. These results can provide useful spatial information for further research on the driving forces and consequences of forest changes, which have critical implications for scientific conservation and management of forests.

show abstract

Section: Forest Mapping With An Updating and Object-based Image Analymentioning

confidence: 99%

Spatio–Temporal Changes of Forests in Northeast China: Insights from Landsat Images and Geospatial Analysis

Chunying

Chen

Wang

et al. 2019

Forests

View full text Add to dashboard Cite

show abstract

“…Identifying and mapping this tree species have been mostly based on field data such as in Simpson, 6 which is costly and time-consuming to perform. Although the application of remote sensing has become widespread, so far, there has been only the research of Pham et al 7 applying remote sensing technology for classifying these trees. It is also important to note that in Ref.…”

Section: Comparison Of Combination Of Dimensionality Reduction and CLmentioning

confidence: 99%

“…It is also important to note that in Ref. 7, the researchers only employed a single machine-learning technique for DR [random forest (RF)] and vegetation species classification [support vector machine (SVM)].…”

Section: Comparison Of Combination Of Dimensionality Reduction and CLmentioning

confidence: 99%

Comparison of combination of dimensionality reduction and classification techniques for identifying tree species using integrated QuickBird imagery and Lidar data

Pham

Brabyn

Dang

et al. 2019

J. Appl. Rem. Sens.

Self Cite

View full text Add to dashboard Cite

Combining data of different types and from different sources for the classification of tree species has gained popularity recently, but training models on such datasets often requires more computational demands and does not always result in higher accuracy due to feature redundancy and irrelevance. Thus preprocessing data using dimensionality reduction (DR) methods can be employed to improve the classification accuracy and reduce computations. The objective of this research is to investigate and compare tree species classification performance for different classification algorithms [naive Bayes (NB), logistic regression (LR), random forest (RF), and support vector machine (SVM)], combined with various DR methods (correlation-based feature selection filter, information gain, wrapper methods, and principal component analysis). Two primary datasets are used-QuickBird and LiDAR, as well as derived topography data. When DR is used prior to classification, the NB classifier had a significant improvement in accuracy. SVM and RF had the best classification accuracy without DR. The overall accuracies (OA) of SVM and RF are 88.2% and 87.2% (kappa 0.84 and 0.83), respectively, followed closely by LR (OA: 84.8%, kappa: 0.79) and more distantly by NB (OA: 79%, kappa: 0.72). It is recommended to use SVM and RF without DR or NB with DR for tree species classification.

show abstract

“…Haywood and Stone [49] developed an automated approach that applied aerial photos and LiDAR CHM to delineate Eucalyptus forest boundaries and achieved 65% overall accuracy. Pham, et al [50] used Quickbird and LiDAR to classify forest species in a New Zealand urban environment and achieved an overall accuracy of 85%. Another OBIA classification by Dupuy, et al [51] used SPOT 5 and LiDAR surfaces to classify tropical vegetation type and gained 92% overall accuracy.…”

Section: Initial Land Cover Classificationmentioning

confidence: 99%

Mapping Net Stocked Plantation Area for Small-Scale Forests in New Zealand Using Integrated RapidEye and LiDAR Sensors

Xu¹,

Morgenroth²,

Manley³

2017

Forests

View full text Add to dashboard Cite

Abstract:In New Zealand, approximately 70% of plantation forests are large-scale (over 1000 ha) with accurate resource description. In contrast, the remaining 30% of plantation forests are small-scale (less than 1000 ha). It is forecasted that these small-scale forests will supply nearly 40% of the national wood production in the next decade. However, in-depth description of these forests, especially those under 100 ha, is very limited. This research evaluates the use of remote sensing datasets to map and estimate the net stocked plantation area for small-scale forests. We compared a factorial combination of two classification approaches (Nearest Neighbour (NN), Classification and Regression Tree (CART)) and two remote sensing datasets (RapidEye, RapidEye plus LiDAR) for their ability to accurately classify planted forest area. CART with a combination of RapidEye and LiDAR metrics outperformed the other three combinations producing the highest accuracy for mapping forest plantations (user's accuracy = 90% and producer's accuracy = 88%). This method was further examined by comparing the mapped plantations with manually digitised plantations based on aerial photography. The mapping approach overestimated the plantation area by 3%. It was also found that forest patches exceeding 10 ha achieved higher conformance with the digitised areas. Overall, the mapping approach in this research provided a proof of concept for deriving forest area and mapping boundaries using remote sensing data, and is especially relevant for small-scale forests where limited information is currently available.

show abstract

Combining QuickBird, LiDAR, and GIS topography indices to identify a single native tree species in a complex landscape using an object-based classification approach

Cited by 39 publications

References 36 publications

Spatio–Temporal Changes of Forests in Northeast China: Insights from Landsat Images and Geospatial Analysis

Spatio–Temporal Changes of Forests in Northeast China: Insights from Landsat Images and Geospatial Analysis

Comparison of combination of dimensionality reduction and classification techniques for identifying tree species using integrated QuickBird imagery and Lidar data

Mapping Net Stocked Plantation Area for Small-Scale Forests in New Zealand Using Integrated RapidEye and LiDAR Sensors

Contact Info

Product

Resources

About