Multi-scale object-based fuzzy classification for LULC mapping from optical satellite images

Do, Thi Hang; Raghavan, Venkatesh; Truong, Luan Xuan; Yonezawa, Go

doi:10.1007/s41324-019-00240-w

Cited by 8 publications

(5 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Then, green belts within residential land are extracted using infrared images and classified as other land types. (17)(18)(19)(20) The principle for extracting the spatial information about each of the LULUCF categories is to use satellite images but refer to thematic maps and other relevant statistical data. (17) The use of satellite images requires geometric and orthometric correction for the interpretation and digitalization of the images.…”

Section: Methods Of Extracting Spatial Information By Using Satellite mentioning

confidence: 99%

Analysis of Land Use, Land-use Change, and Forestry (LULUCF) and Construction of Statistics in Korea

Yu¹,

Shin²,

Ahn³

2019

Sensors and Materials

View full text Add to dashboard Cite

In this study, we analyze land use, land-use change, and forestry (LULUCF) in Korea, referring to the Intergovernmental Panel on Climate Change (IPCC) Good Practice Guidance (GPG) (2003), and propose a method for the systematic construction of statistics for LULUCF. Land-use changes across six categories, including settlement, forest, cropland, grassland, wetland, and other land types, were analyzed using digital topographic, land cover, and land zone maps from the Korea Land Information System (KLIS) using Landsat TM satellite images as primary data. The results of this study indicate that a classification system for each LULUCF category should be established to improve the statistical accuracy of this data and spatial big data applications for land-use analysis should be developed. Furthermore, a method is proposed for the systematic construction of statistics for LULUCF to determine the Korean national greenhouse gas inventory. An explanation of the recommended approach, including gathering satellite image data, LULUCF classifications, and establishing land-use change statistics, is presented. To fulfill the emission reduction targets set in the Paris Agreement, these greenhouse gas emission statistics are essential for different sectors to establish the diverse policies and action plans required to be effective. This will make the national-level monitoring of LULUCF more efficient, leading to a significant reduction in greenhouse gas emission.

show abstract

Section: Methods Of Extracting Spatial Information By Using Satellite mentioning

confidence: 99%

Analysis of Land Use, Land-use Change, and Forestry (LULUCF) and Construction of Statistics in Korea

Yu¹,

Shin²,

Ahn³

2019

Sensors and Materials

View full text Add to dashboard Cite

show abstract

“…Object-oriented approaches include two steps: image segmentation and image classification, wherein image segmentation is a crucial procedure that directly determines the advantages and disadvantages of the results [38], [39]. Do et al [40] used RapidEye and IKONOS imagery of the Lao Cai and Can Tho areas in Vietnam to study land cover classification and then achieved good segmentation performance for water body mapping according to a selected optimal threshold. Objectoriented methods are generally employed in combination with other strategies [41], [42].…”

Section: Related Workmentioning

confidence: 99%

SADA-Net: A Shape Feature Optimization and Multiscale Context Information-Based Water Body Extraction Method for High-Resolution Remote Sensing Images

Wang

Chen

et al. 2022

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

Convolutional neural networks (CNNs) have significance in remote sensing image mapping, and pixel-level representation allows refined results. Due to inconsistencies within a class and different scales of water bodies, the water body mapping has challenges, such as insufficient integrity and rough shape segmentation. To resolve these issues, we proposed an intelligent water bodies extraction method (named SADA-Net) for high-resolution remote sensing images. This method considers multi-scale information, context dependence, and shape features. The network framework integrates three critical components: shape feature optimization (SFO), atrous spatial pyramid pooling (ASPP), and dual attention (DA) modules. SADA-Net can accurately extract an extensive range of water bodies in complex scenarios. SADA-Net has certain advantages regarding small and dense water bodies extraction, as the SFO module effectively solves the defects of the unified processing of low-level features in the encoder stage of CNNs, which highlights the shape information of a water body. Two data types (red, green, and blue (RGB) bands and multi-spectral images) are employed to verify the performance of the proposed network. The best result achieved an evaluation index F1-Score of 96.14% in large-scale image segmentation, and the structural similarity index measure (SSIM) reached 94.70%. Overall, the proposed method achieves the purpose of maximizing the integrity and optimizing the shape of a water body. Additionally, the SADA-Net proposed in this study has a specific reference value for high-resolution remote sensing image water bodies mapping.

show abstract

“…No comparison to the accuracy of ML or SMAP was done. Finally, Do et al (2019) proposed an approach where automatically segmented parcels are classified by fuzzy logic with manually established rules. No direct comparison of this method's accuracy to other classification methods was made.…”

Section: Introductionmentioning

confidence: 99%

Improving pixel‐based classification of GRASS GIS with support vector machine

Nartišs,

Melniks

2023

Transactions in GIS

View full text Add to dashboard Cite

Open source GIS software GRASS releases 8.0 to 8.4 have received some long overdue improvements in imagery handling such as the ability to reuse spectral signature files of existing classifiers, machine readable output of accuracy assessment tool and a support vector machine (SVM) classifier. Practical comparison of all three pixel‐based classifiers of GRASS GIS indicated that the maximum likelihood discriminant analysis classifier is the fastest and least accurate one, followed by a sequential maximum a posteriori classifier with reasonably fast execution time and good accuracy. The newly added SVM classifier is the slowest one but provides the highest accuracy and also shows the highest improvement potential by hyperparameter optimization. To illustrate the capabilities of the core classifiers, a showcase is presented where a single Sentinel 2 image is classified to distinguish 18 crop types using all three classifiers available in GRASS.

show abstract

Multi-scale object-based fuzzy classification for LULC mapping from optical satellite images

Cited by 8 publications

References 26 publications

Analysis of Land Use, Land-use Change, and Forestry (LULUCF) and Construction of Statistics in Korea

Analysis of Land Use, Land-use Change, and Forestry (LULUCF) and Construction of Statistics in Korea

SADA-Net: A Shape Feature Optimization and Multiscale Context Information-Based Water Body Extraction Method for High-Resolution Remote Sensing Images

Improving pixel‐based classification of GRASS GIS with support vector machine

Contact Info

Product

Resources

About