Post natural disaster object recognition system using python-based scale invariant feature transform (SIFT) algorithm

• Published in: IOP conference series. Earth and environmental science Vol. 1355; no. 1; pp. 12040 - 12045
• Main Authors: Zakaria, A, Fauzan, M F, Wahidin, S A N, Adam, A A, Mukhlis, B
• Format: Journal Article
• Language: English
• Published: Bristol IOP Publishing 01.06.2024
• Subjects: Algorithms; Buildings; Damage detection; Data analysis; Feature recognition; Image processing; Invariants; Natural disasters; Object recognition; Pattern recognition; Programming languages; Python; Search and rescue; SIFT
• ISSN: 1755-1307; 1755-1315; 1755-1315
• DOI: 10.1088/1755-1315/1355/1/012040

The natural disaster in Palu became a significant challenge in identifying damaged building objects. The large number of damaged buildings made it difficult for the government to identify the level of damage to buildings and the feasibility of the construction process after natural disasters. This research is an object recognition system after natural disasters using the Scale Invariant Feature Transform (SIFT) algorithm implemented in Python programming language. This research aims to assist the government or National Search and Rescue Agency team in identifying the level of damage to buildings after natural disasters. The methods used in this research include literature study, image capture, processing, testing, and data analysis. The SIFT algorithm was selected to see the accuracy of comparisons of images of buildings before and after natural disasters. The test results show that the SIFT method can identify the average accuracy level of damage to buildings before and after natural disasters, reaching 98.24%.