Convolutional Neural Network-Based Disaster Assessment Using Unmanned Aerial Vehicles

Bocanegra, Maria Gonzalez; Haddad, Rami J.

doi:10.1109/southeastcon45413.2021.9401928

Cited by 5 publications

(1 citation statement)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…After hurricane Harvey, structural damage assessment using DJI Phantom 4 pro is performed after capturing the images at height of 80 m and at an overlap of 80 % Yeom et al (2019). Damage assessment of roads with the captured images from UAV using convolution neural networks is discussed in (Bocanegra and Haddad (2021)). In this case, CNN is used on the images of roads captured from DJI Matrice 600 Pro after the natural disasters, and AlexNet is reported to be the best network with an average value of 74.07 %.…”

Section: Introductionmentioning

confidence: 99%

Cyclone Preparedness, Rescue Operations and Damage Assessment using UAVs

Majumder¹,

Jana²,

Menon³

et al. 2022

Preprint

View full text Add to dashboard Cite

UAV's capability to access remote and inaccessible areas within a quick time can be utilized for effective cyclone management. This paper presents the possible application of UAVs at different stages of cyclone mitigation. The overall system architecture necessary for preparedness, rescue operation, resource allocation, and damage assessment using UAVs during cyclones is described. Although general commercial UAVs are reported to be used in cyclone operations, UAV systems should be planned specifically for cyclone operations to improve efficiency. Here, the specification required for effective and safe UAV operations in the post-cyclone scenario is presented. Mission planning required for various rescue, relief, and damage assessment missions related to cyclone management is discussed. A case study of deploying UAV in Amphan cyclone operation in West Bengal is also presented. This paper can help disaster management authorities to develop UAV systems specifically to cater to cyclone operations.

show abstract

Section: Introductionmentioning

confidence: 99%