Explicit area-based accuracy assessment for mangrove tree crown delineation using Geographic Object-Based Image Analysis (GEOBIA)

Muhammad Kamal; Kasper Johansen

doi:10.1117/12.2278179

5 October 2017 Explicit area-based accuracy assessment for mangrove tree crown delineation using Geographic Object-Based Image Analysis (GEOBIA)

Muhammad Kamal, Kasper Johansen

Author Affiliations +

Proceedings Volume 10428, Earth Resources and Environmental Remote Sensing/GIS Applications VIII; 104280I (2017) https://doi.org/10.1117/12.2278179
Event: SPIE Remote Sensing, 2017, Warsaw, Poland

Abstract

Effective mangrove management requires spatially explicit information of mangrove tree crown map as a basis for ecosystem diversity study and health assessment. Accuracy assessment is an integral part of any mapping activities to measure the effectiveness of the classification approach. In geographic object-based image analysis (GEOBIA) the assessment of the geometric accuracy (shape, symmetry and location) of the created image objects from image segmentation is required. In this study we used an explicit area-based accuracy assessment to measure the degree of similarity between the results of the classification and reference data from different aspects, including overall quality (OQ), user’s accuracy (UA), producer’s accuracy (PA) and overall accuracy (OA). We developed a rule set to delineate the mangrove tree crown using WorldView-2 pan-sharpened image. The reference map was obtained by visual delineation of the mangrove tree crowns boundaries form a very high-spatial resolution aerial photograph (7.5cm pixel size). Ten random points with a 10 m radius circular buffer were created to calculate the area-based accuracy assessment. The resulting circular polygons were used to clip both the classified image objects and reference map for area comparisons. In this case, the area-based accuracy assessment resulted 64% and 68% for the OQ and OA, respectively. The overall quality of the calculation results shows the class-related area accuracy; which is the area of correctly classified as tree crowns was 64% out of the total area of tree crowns. On the other hand, the overall accuracy of 68% was calculated as the percentage of all correctly classified classes (tree crowns and canopy gaps) in comparison to the total class area (an entire image). Overall, the area-based accuracy assessment was simple to implement and easy to interpret. It also shows explicitly the omission and commission error variations of object boundary delineation with colour coded polygons.

Citation Download Citation

Muhammad Kamal and Kasper Johansen "Explicit area-based accuracy assessment for mangrove tree crown delineation using Geographic Object-Based Image Analysis (GEOBIA)", Proc. SPIE 10428, Earth Resources and Environmental Remote Sensing/GIS Applications VIII, 104280I (5 October 2017); https://doi.org/10.1117/12.2278179

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available