Determining the most relevant frequency bands in motion identification by accelerometer sensors

Öztekin, Abdulkerim; Ertuğrul, Ömer Faruk; Aldemir, Erdoğan; Acar, Emrullah

doi:10.1007/s11042-022-12099-5

Cited by 2 publications

(2 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…(3) Remove trending items Since the accelerometer acquires data in the form of a vehicle, vibrations, bumps, and magnetic field interference from vehicle electrical appliances during vehicle driving it will cause the integrated speed and displacement signals to deviate from the baseline [27], and the amplitude of the deviation from the baseline will change at any time with the increase in the detection distance. This phenomenon is a trend term.…”

Section: (10𝜋)mentioning

confidence: 99%

Detection and Analysis of Pavement-Section Based on Laser Displacement Sensor

Han,

Xiong,

Liu

2023

Sensors

View full text Add to dashboard Cite

The section detection of the pavement is the data basis for measuring the road smoothness, rutting, lateral slope, and structural depth. The detection of the Pavement-Section includes longitudinal-section inspection and cross-section inspection. In this paper, based on multiple laser displacement sensors, fused accelerometers and attitude sensors, and using vehicle-mounted high-speed detection, we design a sensor-fused pavement section data acquisition method, establish the relevant mathematical model, and realize the automatic acquisition of pavement longitudinal and transverse sections. The acceleration sensor is filtered to improve the accuracy of data acquisition, and the error of the detection system is calculated and analyzed. Through the actual measurement, the vehicle-mounted high-speed pavement profile detection method adopted in this paper can not only accurately detect the profile of the pavement profile, but also improve the detection efficiency, providing a cost-effective detection mode for road surface detection.

show abstract

Section: (10𝜋)mentioning

confidence: 99%

Detection and Analysis of Pavement-Section Based on Laser Displacement Sensor

Han,

Xiong,

Liu

2023

Sensors

View full text Add to dashboard Cite

show abstract

“…Machine learning [ 6 – 8 ] can not only be applied to the preprocessing stage of monitoring data but also, more importantly, it is necessary to establish a learning model and actively learn the information contained in the data to evaluate the safety of the structural state and reflect the health state in time. In the face of a large amount of information generated all the time, the traditional manual method is simply unable to make efficient and rapid judgment [ 9 , 10 ]. Machine learning first appeared in the last century.…”

Section: Introductionmentioning

confidence: 99%

Analysis of Human Exercise Health Monitoring Data of Smart Bracelet Based on Machine Learning

2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

The smart bracelet has become a hot-selling commodity, according to a daily consumption survey. Based on people’s interest and concern for their health, the smart bracelet, as a design and application for achieving healthy weight loss monitoring, is quickly becoming a popular new favorite. This bracelet detects fat using the near-infrared diffuse reflection principle, with the goal of assisting people in controlling and maintaining a healthy weight. A large amount of data has been accumulated in all walks of life due to the development of the Internet network and data storage technology. As a result, the emergence of machine learning plays a critical role in the data analysis of human sports health monitoring of smart bracelets. Based on machine learning, this paper investigates the data analysis of human sports health monitoring smart bracelets. When the population index reaches 50 in the analysis of health monitoring data, the average accuracy of data mining is 86.8 percent, the average accuracy of the association rule algorithm is 85.9 percent, the average accuracy of the collaborative filtering algorithm is 84.3 percent, and the average accuracy of the machine learning algorithm is 90.1 percent in this paper. Among the four algorithms, the method presented in this paper is clearly the most effective, stable, and accurate. The system’s stability and accuracy have been greatly improved by the addition of GPS-assisted and hand-up misjudgment algorithms. Because the smart bracelet is inexpensive, easy to wear, and consistent with consumer psychology, it is becoming increasingly popular to use it to monitor the human body’s sports health.

show abstract

Determining the most relevant frequency bands in motion identification by accelerometer sensors

Cited by 2 publications

References 34 publications

Detection and Analysis of Pavement-Section Based on Laser Displacement Sensor

Detection and Analysis of Pavement-Section Based on Laser Displacement Sensor

Analysis of Human Exercise Health Monitoring Data of Smart Bracelet Based on Machine Learning

Contact Info

Product

Resources

About