Rapid Earthquake Assessment from Satellite Imagery Using RPN and Yolo v3

Author(s):  
Sanjeeb Prasad Panday ◽  
Saurav Lal Karn ◽  
Basanta Joshi ◽  
Aman Shakya ◽  
Rom Kant Pandey
2020 ◽  
Vol 2020 (8) ◽  
pp. 114-1-114-7
Author(s):  
Bryan Blakeslee ◽  
Andreas Savakis

Change detection in image pairs has traditionally been a binary process, reporting either “Change” or “No Change.” In this paper, we present LambdaNet, a novel deep architecture for performing pixel-level directional change detection based on a four class classification scheme. LambdaNet successfully incorporates the notion of “directional change” and identifies differences between two images as “Additive Change” when a new object appears, “Subtractive Change” when an object is removed, “Exchange” when different objects are present in the same location, and “No Change.” To obtain pixel annotated change maps for training, we generated directional change class labels for the Change Detection 2014 dataset. Our tests illustrate that LambdaNet would be suitable for situations where the type of change is unstructured, such as change detection scenarios in satellite imagery.


Author(s):  
SiMing Liang ◽  
FengYang Qi ◽  
YiFan Ding ◽  
Rui Cao ◽  
Qiang Yang ◽  
...  

Author(s):  
Vincentius P. Siregar ◽  
Sam Wouthuyzen ◽  
Andriani Sunuddin ◽  
Ari Anggoro ◽  
Ade Ayu Mustika

Shallow marine waters comprise diverse benthic types forming habitats for reef fish community, which important for the livelihood of coastal and small island inhabitants. Satellite imagery provide synoptic map of benthic habitat and further utilized to estimate reef fish stock. The objective of this research was to estimate reef fish stock in complex coral reef of Pulau Pari, by utilizing high resolution satellite imagery of the WorldView-2 in combination with field data such as visual census of reef fish. Field survey was conducted between May-August 2013 with 160 sampling points representing four sites (north, south, west, and east). The image was analy-zed and grouped into five classes of benthic habitats i.e., live coral (LC), dead coral (DC), sand (Sa), seagrass (Sg), and mix (Mx) (combination seagrass+coral and seagrass+sand). The overall accuracy of benthic habitat map was 78%. Field survey revealed that the highest live coral cover (58%) was found at the north site with fish density 3.69 and 1.50 ind/m2at 3 and 10 m depth, respectively. Meanwhile, the lowest live coral cover (18%) was found at the south site with fish density 2.79 and 2.18  ind/m2 at 3 and 10 m depth, respectively. Interpolation on fish density data in each habitat class resulted in standing stock reef fish estimation:  LC (5,340,698 ind), DC (56,254,356 ind), Sa (13,370,154 ind), Sg (1,776,195 ind) and Mx (14,557,680 ind). Keywords: mapping, satellite imagery, benthic habitat, reef fish, stock estimation


Sign in / Sign up

Export Citation Format

Share Document