SAMARA, February 1 Scientists at Samara University have developed a high-speed neuron capable of analyzing hyperspectral data from drones or space satellites, the university’s press service reported.
«
"Scientists of Samara University named after. Korolev developed and tested a high-speed neural network capable of analyzing an incoming video stream in real time and almost instantly recognizing and finding specified objects and images in this video stream. Along with analyzing images from a regular video camera, development can quickly <…> also analyze data obtained using hyperspectrometers — devices that see reality in a multi-channel spectral display and make it possible to detect objects invisible to conventional means of observation,” says a statement from Samara University.
< /span>
The optical neural network was created on the basis of an analog photonic computing system, which has a number of advantages: complete immunity from electromagnetic interference, low power consumption and the possibility of parallel data processing. Scientists created a demonstration sample using standard laboratory optical-mechanical components, various modulators and video cameras, said Roman Skidanov, a professor at the Department of Technical Cybernetics at Samara University.
«Recognition reliability during the first experiments on the demonstration sample was 93.75%. In 2024, it is planned to assemble and test an experimental sample of the system in a fairly compact case the size of a small computer system unit. Accuracy and the recognition reliability of the experimental sample should increase due to the selection of components with improved characteristics,” Skidanov noted.
Hyperspectral imaging or hyperspectral remote sensing of the Earth, which is carried out using drones or a space satellite, can effectively detect greenhouse gases, conduct geological exploration of hard-to-reach areas, and more accurately monitor the occurrence of forest fires.
< br />
Свежие комментарии