NASA ML for Multi-Sensor Fire and Smoke Detection Remote Internship-PAID

Description

The NASA ML for Multi-Sensor Fire and Smoke Detection Remote Internship is available to upper-division undergraduate and graduate students who are majoring in computer science, computer engineering, mathematics, statistics, and related fields. To be considered for this role, you should be knowledgeable about programming (python), machine learning, image classification, computer vision, image processing, and deep learning. Additionally, you must be a U.S. citizen. As an intern, you will help improve existing smoke and fire detection using computer science and machine learning. Your duties may include: We are looking for an intern with experience in machine learning and computer science to help improve existing smoke and fire detection. Ideal candidate is passionate about mitigating the effects of wildfires and experienced with building computer vision and image processing machine learning models. The specific problem will involve: building camera specific models for smoke/fire detection using anomaly detection theory (e.g, auto encoders), building ensemble models for different camera views of the same location, image difference calculation, and/or classification with multiple inputs (image + other sensor data). Exact model development will be left to the intern and mentors will provide guidance and access to training data. The model(s) will eventually by deployed using the Open Data Integration (ODIN) framework, which will involve full stack development with possible internship extension to summer and beyond. This internship is offered on a remote basis during the summer.