Machine vision systems for detection of fast-moving objects in low visibility conditions

L.M. Tovstenko, М.A. Kosovets, O.A. Tovstenko

Abstract


Machine vision technology for the detection of fast moving objects in low visibility conditions requires a more careful approach to the optical system, image processing tools, cameras based on a silicon and indium gallium arsenide focal plane matrix in the infrared range, which take into account the magnitude and variation of atmospheric brightness, ground illumination and evaluation of sensitivity and cameras in the terahertz range. Since the main focus is on aerial objects that have low visibility, especially in the dark, the use of infrared cameras has become the standard. The movement of the object, the shortcomings of the optical system create additional difficulties when processing the effects of interference, thermal noise of cameras, respectively, the volume and time of calculations increases, which play a key role in real-time systems for detecting and tracking moving targets. In order to take into account the spectral characteristics of cameras and the influence of external factors, neural networks with deep learning are applied with the maximum use of image processing packages.

Prombles in programming 2024; 2-3: 207-214


Keywords


bandwidth; terahertz range; hardware accelerators; spectrum; convolutional neural network; distributed information processing; tensor processing; embedded system tests; deep learning

References


Kosovets. Preliminary Tests Terahertz 3D Imaging Radar, №2, 2017. pp. 54-61.

M.Коsovets, L. Tovstenko. Smart-Handled THz 3D Radar with Cloud Intelligent technology and artificial intelligence algorithms by level of abstraction.

Коsovets M., Tovstenko L. The Practical Aspect of using the Artificial intellectual Technology for Building a multidimensional Function CFAR for SMART-HANDLED LPI Radar.

M. Kosovets, L. Tovstenko. Deep Learning of Convolution Neural Networks IN IoT.

Коsovets M., Tovstenko L. Specific Features of the use of Artificial Intelligence in the Development of the Architecture of Intelligent Fault Tolerant Radar Systems.

Коsovets M., Tovstenko L. The Problem of Development the Architecture of modern cognitive.

Коsovets M., Tovstenko L. Neural Network Component of modern information System on Mobile Platforms: Cognitive Radar System.Cybernetics and Computer Technologies 2023.

Коsovets M., Tovstenko L. Artificial Intelligence in Cloud Based Mobile Radar Computing. 2023. №1.

Van A. Hodgkin U.S. Army Research and Development Command Night Vision and electronic Sensors Directorate Fort Belvoir, Virginia. (2011)

E. Bender, "Present Image Intensifier Tube Structures", Chapter 5 in Electro-Optical Imaging: System Performance and Modeling, Lucien (2002).

X. C. Liu et al., "CCD/CMOS hybrid FPA for low light level imaging", Proc. SPIE 5881, 58810C (2005).

B. M. Onat et al., "Ultra-low dark current In GaAs technology for focal plane arrays for lowlight level visible-shortwave infrared imaging", Proc. SPIE 6542, 65420L (2007).

M. MacDougal et al., "Short-wavelength infrared imaging using low dark current InGaAs detector arrays and vertical-cavity surface-emitting laser illuminators". Opt. Eng. 50(6), 061011 (2011).

C. Leinert et al., "The 1997 reference of diffuse night sky brightness", Astron. Astrophys. Suppl. Ser. 127(1), 14–30 (1998).

L. Broadfoot and K. R. Kendall, "The airglow spectrum, 3100–10,000 A", J. Geophys. Res. 73(1), 426–428 (1968).

C. Leinert et al., "The 1997 reference of diffuse night sky brightness", Astron. Astrophys. Suppl. Ser. 127(1), 14–30 (1998).

F. a. o. Chollet, "Keras" GitHub. 2015. Radar System.


Refbacks

  • There are currently no refbacks.