Hyper spectral imagery to track suspicious movements in dense foliage areas
The Indian Air Force (IAF) has commissioned the development of an aerial surveillance system for monitoring suspicious moments under dense foliage cover along the international border and some parts within the country. This hyper spectral imagery programme, a first for the country, seeks to locate and identify suspicious movements and is expected to be operational in the next two years.
What the programme would mean is that the data gathered from aerial sources would be processed within minutes, and researchers would be able to tell security forces on the ground of “unwanted” human presence in the area, their numbers and locations, among other inputs.
The IAF is spending ₹13 crore for the programme in which human resources from premier research and education institutions across the country are being put together, and has roped in a senior researcher and lead technologist of Indian origin from the National Aeronautics and Space Administration (NASA) as the chief technologist for the programme.
“For the first time, we will be having multi-sensor data fusion that will help identify an adversary. We have a mandate to change the way we fight the war. The IAF is willing to wait for a couple of years to reap the benefits,” said a senior source in the IAF’s Air Warfare Strategy Cell. “Some agencies in India tried to work on the technology, but have not succeeded,” sources added.
The IAF, which had tracked the work of Houston-based senior scientist Kumar Krishen’s work for nearly three years, finally roped him in after he quit NASA in September 2018.
“This is a unique patent application, which will help detect or identify the presence of a human below trees, dense foliage, shrubs or inside a structure, whether it is day or night. It can detect human presence from air even if there a cloud cover, dense fog or snow cover,” explained Dr. Krishen, who spent 43 years in NASA.
Methods in place
Data/images will be captured through optical and infra red sensors — to be most likely mounted on an unmanned aerial vehicle — and processed through deep learning algorithms.
Dr. Krishen said that the methods are in place while they were working on handling data and tuning up the algorithm.
“We are talking to nearly half a dozen educational institutes, including the IIT Delhi and Rourkee. We want to develop sophisticated sensors locally. I am confident of having optical sensors made in India,” said Dr. Krishen.
Source: Read Full Article