The National Science Foundation has selected Vishal Patel, an assistant professor in the Department of Electrical and Computer Engineering, for its Early CAREER Award, which recognizes early stage scholars with high levels of promise and excellence.

Using the $500,000, five-year award, Patel and his team aim to develop data-driven learning-based approaches for restoration and understanding of images degraded by atmospheric turbulence. Patel says that much of the money will go towards supporting his PhD students. The title of the project is “Seeing Through Atmospheric Turbulence: Image Restoration and Understanding using Deep Convolutional Neural Networks.”

“Atmospheric turbulence can significantly degrade the quality of images acquired by long-range imaging systems by causing spatially and temporally random fluctuations in the index of refraction of the atmosphere,” Patel said. “Variations in the refractive index cause the captured images to be geometrically distorted and blurry. These distortions reduce effective resolution and exploitable information, leading to a loss of situational awareness and a general decrease in usefulness.”

The algorithms that will be developed in this project will significantly enhance the quality of images and videos collected by long-range visible and infrared imagining systems. Patel believes that these new methods will also be useful in many applications of remote sensing, including long-range surveillance, optical communications, astronomy, road traffic monitoring, underwater imaging, and autonomous navigation.

Patel’s project also includes an extensive community outreach plan focused on helping underrepresented minority students from disadvantaged public middle and high schools in Baltimore learn about imaging and image processing. His “Computational Imaging” program will offer students a series of workshops, seminars, and lab visits spaced throughout the year, with the goal of disseminating imaging research from his group and the broader research community into classrooms.

“My goal as an educator is to develop a comprehensive education plan that builds upon and complements my research plan on machine learning-based image restoration in a cohesive fashion,” Patel said. “Image restoration creates an ideal platform for introducing the future generations of electrical engineers and computer scientists to the fascinating world of computational imaging, machine learning, computer vision and signal processing research using real world applications in hardware and software that use images and videos.”

Patel joined Johns Hopkins in 2018 from Rutgers University.