Журнал «Современная Наука»

Russian (CIS)English (United Kingdom)
 MOSCOW +7(495)-142-86-81

COMPARATIVE ANALYSIS AND EVALUATION OF NEURAL NETWORK QUANTIZATION METHODS OF COMPACT MOBILENET ARCHITECTURES FOR DEPLOYMENT ON MOBILE COMPUTING DEVICES FOR VISUAL MONITORING OF DRIVERS' SLEEP IN VARIOUS LIGHTING CONDITIONS

Afanasyev A. G.  (Postgraduate student, BMSTU)

This paper presents a comparative analysis of the impact of quantization methods on the efficiency, performance, and accuracy of neural network models based on the compact MobileNetV1, MobileNetV2, and MobileNetV3 architectures. This analysis is based on the implementation of a driver sleep/wake binary classification task using a Samsung Galaxy A50 smartphone with the front camera. A comparative analysis of detection quality under different lighting conditions (daytime and nighttime) is conducted, and tradeoffs between accuracy, inference speed, and resource requirements are considered.

Keywords:MobileNet, smartphone, falling asleep, drowsiness, driver, visual monitoring, neural network, quantization
 

Read the full article …

Citation link:
Afanasyev A. G. COMPARATIVE ANALYSIS AND EVALUATION OF NEURAL NETWORK QUANTIZATION METHODS OF COMPACT MOBILENET ARCHITECTURES FOR DEPLOYMENT ON MOBILE COMPUTING DEVICES FOR VISUAL MONITORING OF DRIVERS' SLEEP IN VARIOUS LIGHTING CONDITIONS // Современная наука: актуальные проблемы теории и практики. Серия: Естественные и Технические Науки. -2026. -№02/2. -С. 43-48 DOI 10.37882/2223-2966.2026.02-2.05
LEGAL INFORMATION:
Reproduction of materials is permitted only for non-commercial purposes with reference to the original publication. Protected by the laws of the Russian Federation. Any violations of the law are prosecuted.
© ООО "Научные технологии"

Terms Publications

Issues of the journals (Archive)

ECONOMICS AND LAW
HUMANITIES
NATURAL AND TECHNICAL SCIENCES
COGNITION

Серия - Natural and Technical Sciences