Build a data set for fire detection
To train a deep learning model that detects fires, we need a data set with bounding boxes. We label the images and create .csv
files with the box location (xmin, ymin, xmax, ymax) and a class(fire) in the image. For management and quick training, we need to create .record
files supported by tensorflow through .csv
files and images.
*It's a good to upload the images we collected, but it can cause copyright issues because it's important for implementing deep learning models. The current purpose of this project is to experience machine learning and team collaboration rather than commercialization, so it may be a good idea not to upload the images to avoid other problems.
-
Started branch
1-build-a-data-set-for-fire-detection
-
Please register or login to post a comment