Recognition of Driver Turn Behavior Based on Video Analysis

2012 ◽  
Vol 433-440 ◽  
pp. 6230-6234 ◽  
Author(s):  
Yu Feng Lu ◽  
Chun Ling Li
Keyword(s):  
Standard Deviation ◽  
Video Analysis ◽  
Particle Filtering ◽  
Detection System ◽  
Driving Behavior ◽  
Recognition Algorithm ◽  
Driver Distraction ◽  
New Method ◽  
Video Images ◽  
State Detection

In view of the shortcoming of driver state detection system which may take turn behavior as driver distraction, a new method to recognize turn behavior was proposed based on video images analysis. The driver hands position in different driving behavior were analyzed and we found that the position of driver’s hands changed more violent when in turning than in other driving behavior. So we may use standard deviation of driver hands position to recognize driver turn behavior. In order to improve the hands locating speed the Particle Filtering was used to track the driver hands. And experiments resulted that the recognition algorithm can identify the driver's turn behavior.

Research on Enhancement Technology of Conveyor Belt Fault Images Based on Fruit Fly Optimization Algorithm

2014 ◽  
Vol 8 (1) ◽  
pp. 685-689
Author(s):  
Chunqing Ye ◽  
Changyun Miao ◽  
Xianguo Li ◽  
Yanli Yang
Keyword(s):  
Optimization Algorithm ◽  
Detection System ◽  
Fruit Fly ◽  
Conveyor Belt ◽  
Recognition Algorithm ◽  
Fruit Fly Optimization Algorithm ◽  
Fruit Fly Optimization ◽  
Wire Ropes ◽  
Fault Recognition ◽  
Steel Cord

In this research, we studied the fault recognition algorithm of steel cord conveyor belt, and obtained the wire ropes image by adopting the detection system of steel cord conveyor belt, so that the fault recognition algorithm of steel cord conveyor belt was proposed based on Fruit fly optimization algorithm. As we know that the fruit fly optimization algorithm is used for fault detection of the processing steel cord conveyor belt image and for obtaining the fault image. In the MATLAB environment, the algorithm process was designed and verified in terms of the effectiveness and accuracy. The experimental results show that with fast speed and high accuracy in detecting the fault image of steel cord conveyor belt rapidly and accurately, and in classifying scratch from fracture the proposed algorithm is suitable for the fault recognition of steel cord conveyor belt automatically.

SafeDriving: An Effective Abnormal Driving Behavior Detection System based on EMG Signals

2021 ◽  
pp. 1-1
Author(s):  
Yuanzhao Fan ◽  
Fei Gu ◽  
Jin Wang ◽  
Jianping Wang ◽  
Kejie Lu ◽  
...  
Keyword(s):  
Detection System ◽  
Driving Behavior ◽  
Behavior Detection

scholarly journals Alterações dimensionais em pastas de óxido de zinco/eugenol: um novo método para sua mensuração

2021 ◽  
pp. 39-55
Author(s):  
Léo Werner Süffert ◽  
Ennio Pessôa
Keyword(s):  
Standard Deviation ◽  
Personal Information ◽  
High Variability ◽  
Point Of View ◽  
New Method ◽  
Extensive Review ◽  
Review Of The Literature ◽  
Dimensional Changes ◽  
Impression Materials ◽  
Clinical Point

After an extensive review of the literature, regarding zinc.oxide/eugenol impression pastes, we selected 20 of the most representatives as our references. Trough personal information of several of the investigators it was discovered that dimensional changes of theese materials is one of the most difficult properties to be measured. A new method was developed to measure dimensional changes ot 4 (four) of the most widely used zinc.oxide/eugenol impression materials in Brazil. The results, presented through several graphs and tables showed that dimensional changes varied from 0,003%, values which may probably be considered negligible from a clinical point of view. We noticed, however, high values for standard deviation and variance which indicate the high variability within the experiments. Those values were not found when we used the same method with mercaptan and silicone impression materials, in which the measurement of dimensional changes was highly reproducible. One hypothesis (which we intend to investigate in a later research) is that, during storage, a sedimentation could occur, of the components of greater density! Consequently ther might result a change in composition, independent of the method used to establishe the proportion of the two pastes, be it by wheight or measurement of lenght, which could be the cause of variability of the composition of each mixture!

scholarly journals Detection of Breaking Waves in Single Wave Gauge Records of Surface Elevation Fluctuations

2019 ◽  
Vol 36 (9) ◽  
pp. 1863-1879 ◽  
Author(s):  
Dan Liberzon ◽  
Alexandru Vreme ◽  
Sagi Knobler ◽  
Itamar Bentwich
Keyword(s):  
Water Waves ◽  
Cost Effective ◽  
Breaking Waves ◽  
Recognition Algorithm ◽  
New Method ◽  
Surface Elevation ◽  
Detection Accuracy ◽  
Single Wave ◽  
Accurate Detection ◽  
Forced Waves

We report the development of a new method for accurate detection of breaking water waves that addresses the need for an accurate and cost-effective method that is independent of human decisions. The new detection method, which enables the detection of breakers using only surface elevation fluctuation measurements from a single wave gauge, supports the development of a new method for research relating to water waves and wind–wave interactions. According to the proposed method, detection is based on the use of the phase-time method to identify breaking-associated patterns in the instantaneous frequency variations of surface elevation fluctuations. A wavelet-based pattern recognition algorithm is devised to detect such patterns and provide accurate detection of breakers in the examined records. Validation and performance tests, conducted using both laboratory and open-sea data, including mechanically generated and wind-forced waves, are reported as well. These tests allow us to derive a set of parameters that assure high detection accuracy rates. The method is shown to be capable to achieve a positive detection rate exceeding 90%.

A method of estimating intracranial decompensation in man

1976 ◽  
Vol 45 (2) ◽  
pp. 155-158 ◽  
Author(s):  
Jerzy Szewczykowski ◽  
Pawel Dytko ◽  
Adam Kunicki ◽  
Jolanta Korsak-Sliwka ◽  
Stanislaw Sliwka ◽  
...  
Keyword(s):  
Standard Deviation ◽  
Intracranial Pressure ◽  
Characteristic Feature ◽  
Computer System ◽  
Monitoring System ◽  
New Method ◽  
Icp Monitoring ◽  
Content Type ◽  
On Line ◽  
Fine Print

✓ A new method of estimating intracranial decompensation in man is described. An on-line computer system is connected to an intracranial pressure (ICP) monitoring system to compute regression plots of mean ICP vs standard deviation; standard deviation is used as a measure of ICP instability. Two zones with distinctly different slopes are a characteristic feature of these plots. It is thought that the changes of slope signify intracranial decompensation.

Classification of Driver Distraction: A Comprehensive Analysis of Feature Generation, Machine Learning, and Input Measures

2019 ◽  
Vol 62 (6) ◽  
pp. 1019-1035 ◽  
Author(s):  
Anthony D. McDonald ◽  
Thomas K. Ferris ◽  
Tyler A. Wiener
Keyword(s):  
Machine Learning ◽  
Driving Behavior ◽  
Driver Distraction ◽  
Machine Learning Algorithms ◽  
Physiological Data ◽  
Learning Approaches ◽  
Feature Generation ◽  
Driver Performance ◽  
Ensemble Machine Learning ◽  
Vehicle Information

Objective The objective of this study was to analyze a set of driver performance and physiological data using advanced machine learning approaches, including feature generation, to determine the best-performing algorithms for detecting driver distraction and predicting the source of distraction. Background Distracted driving is a causal factor in many vehicle crashes, often resulting in injuries and deaths. As mobile devices and in-vehicle information systems become more prevalent, the ability to detect and mitigate driver distraction becomes more important. Method This study trained 21 algorithms to identify when drivers were distracted by secondary cognitive and texting tasks. The algorithms included physiological and driving behavioral input processed with a comprehensive feature generation package, Time Series Feature Extraction based on Scalable Hypothesis tests. Results Results showed that a Random Forest algorithm, trained using only driving behavior measures and excluding driver physiological data, was the highest-performing algorithm for accurately classifying driver distraction. The most important input measures identified were lane offset, speed, and steering, whereas the most important feature types were standard deviation, quantiles, and nonlinear transforms. Conclusion This work suggests that distraction detection algorithms may be improved by considering ensemble machine learning algorithms that are trained with driving behavior measures and nonstandard features. In addition, the study presents several new indicators of distraction derived from speed and steering measures. Application Future development of distraction mitigation systems should focus on driver behavior–based algorithms that use complex feature generation techniques.

The Development and Design of a Laser Detection System for Tire Thickness Measurement

2013 ◽  
Vol 418 ◽  
pp. 128-131
Author(s):  
King Sun Lee
Keyword(s):  
Standard Deviation ◽  
Real Time ◽  
Detection System ◽  
Measurement Data ◽  
Thickness Measurement ◽  
Inspection System ◽  
Laser Sensors ◽  
Repeatability Error ◽  
On Line ◽  
Measuring Machine

This system is a self-developed real-time thickness inspection system including high-precision laser sensors and a mobile platform for on-line detection of tire rubber skin. The measurement data is used to calculate the standard deviation and process capability indices, and to evaluate measurement capacity. The system is a real-time measurement system in which the obtained measuring data compare with the standard value and show any errors. A technician can adjust the process parameters precisely on-line to improve product quality. The standard deviation of repeatability of the system for height is within +/- 0.0081 mm. The repeatability error of the horizontal sliding rail is within 0.0145mm, while the measurement error between this system and a coordinated measuring machine is within 0.028mm.

Sign in / Sign up

Export Citation Format

Share Document