Department of Computer Engineering, Sandip Institute of Technology and Research Centre, Nashik, Maharashtra, India
Abstract
For building safety design and resilience assessments, accurate fire load data is essential. Traditional fire load estimating techniques, including fire load surveys, take a lot of time, are laborious, and are prone to mistakes. This research suggests a vision-based way to automatically detect indoor fire load using deep learning-based instance segmentation as a first step in solving this issue. First, several categories of indoor components are determined by the materials they are made of. The development of an interior scene image collection with instance annotations follows. Finally, a model for pixel-level fire load detection is created using CNN). Results demonstrate that our model is capable of segmenting the dataset with a promising accuracy of training model.
Chaudhari, P. A., Karad, P., Hyalij, T., Chandroth, S., & Nawale, K. (2023). A Survey on Indoor Fire Load Recognition Using Image Dataset. Int. J. of Aquatic Science, 14(1), 90-95.
MLA
Pallavi A. Chaudhari; Prajakta Karad; Tanuja Hyalij; Shradha Chandroth; Kajal Nawale. "A Survey on Indoor Fire Load Recognition Using Image Dataset". Int. J. of Aquatic Science, 14, 1, 2023, 90-95.
HARVARD
Chaudhari, P. A., Karad, P., Hyalij, T., Chandroth, S., Nawale, K. (2023). 'A Survey on Indoor Fire Load Recognition Using Image Dataset', Int. J. of Aquatic Science, 14(1), pp. 90-95.
VANCOUVER
Chaudhari, P. A., Karad, P., Hyalij, T., Chandroth, S., Nawale, K. A Survey on Indoor Fire Load Recognition Using Image Dataset. Int. J. of Aquatic Science, 2023; 14(1): 90-95.
)