I made already a model with a python language (semantic segmentation) to detect the oil spill in the ocean and it is working with no errors and it gives a result but the accuracy is not very good and it gives a false alarm sometimes (it means even if there is no oil spill and there is a land or sun illumination the model mask these objects as an oil spill)
So, the task as follow:
1- Improve the detection accuracy
2- Make the model able to select many images at the same time not one by one, and result must be divided into two folders (true oil spill - no oil spill)
3- Calculate the accuracy by using the F1 method (precision and recall) and represent it on a graph
7 freelancer đang chào giá trung bình $37 cho công việc này
Hi there, I have read through your project description. I will be willing to help you complete the project. I will be looking forward to hear from you. Please contact me on PM for details.