scholarly journals An Improved Roadside Parking Space Occupancy Detection Method Based on Magnetic Sensors and Wireless Signal Strength

Sensors ◽  
2019 ◽  
Vol 19 (10) ◽  
pp. 2348 ◽  
Author(s):  
Liangliang Lou ◽  
Jinyi Zhang ◽  
Yong Xiong ◽  
Yanliang Jin
Keyword(s):  
Detection Method ◽  
Dead Zone ◽  
Sampling Rate ◽  
Vehicle Detection ◽  
Magnetic Sensors ◽  
Experimental Results ◽  
Detection Accuracy ◽  
Parking Space ◽  
Wireless Transceivers ◽  
Occupancy Detection

Smart Parking Management Systems (SPMSs) have become a research hotspot in recent years. Many researchers are focused on vehicle detection technology for SPMS which is based on magnetic sensors. Magnetism-based wireless vehicle detectors (WVDs) integrate low-power wireless communication technology, which improves the convenience of construction and maintenance. However, the magnetic signals are not only susceptible to the adjacent vehicles, but also affected by the magnetic signal dead zone of high-chassis vehicles, resulting in a decrease in vehicle detection accuracy. In order to improve the vehicle detection accuracy of the magnetism-based WVDs, the paper introduces an RF-based vehicle detection method based on the characteristics analysis of received signal strengths (RSSs) generated by the wireless transceivers. Since wireless transceivers consume more energy than magnetic sensors, the proposed RF-based method is only activated to extract the data characteristics of RSSs to further judge the states of vehicles when the data feature of magnetic signals is not sufficient to provide accurate judgment on parking space status. The proposed method was evaluated in an actual roadside parking lot and experimental results show that when the sampling rate of magnetic sensor is 1 Hz, the vehicle detection accuracy is up to 99.62%. Moreover, compared with machine-learning-based vehicle detection method, the experimental results show that our method has achieved a good compromise between detection accuracy and power consumption.

Semi supervised inspection algorithm of automatic packaging curve based on deep learning

2021 ◽  
pp. 1-10
Author(s):  
Yong He
Keyword(s):  
Deep Learning ◽  
Supervised Learning ◽  
Optimization Algorithm ◽  
Posterior Probability ◽  
Detection Method ◽  
Detection System ◽  
Experimental Results ◽  
Detection Accuracy ◽  
Data Set ◽  
Packaging Process

The current automatic packaging process is complex, requires high professional knowledge, poor universality, and difficult to apply in multi-objective and complex background. In view of this problem, automatic packaging optimization algorithm has been widely paid attention to. However, the traditional automatic packaging detection accuracy is low, the practicability is poor. Therefore, a semi-supervised detection method of automatic packaging curve based on deep learning and semi-supervised learning is proposed. Deep learning is used to extract features and posterior probability to classify unlabeled data. KDD CUP99 data set was used to verify the accuracy of the algorithm. Experimental results show that this method can effectively improve the performance of automatic packaging curve semi-supervised detection system.

Video Vehicle Detection Method Based on Multiple Color Space Information Fusion

2012 ◽  
Vol 546-547 ◽  
pp. 721-726
Author(s):  
Hong Xiang Shao ◽  
Xiao Ming Duan
Keyword(s):  
Information Fusion ◽  
Traditional Method ◽  
Detection Method ◽  
Color Space ◽  
Vehicle Detection ◽  
Data Accuracy ◽  
Traffic Information ◽  
Detection Accuracy ◽  
Relative Independence ◽  
Hsi Color Space

A detection method which selective fuses the nine detection results of RGB, YCbCr and HSI color space according to the image color space relative independence of each component and complementarities is approached in order to improve vehicle video detection accuracy. The method fuses three different detection results in nine components by the value of H when the value of both S and I are higher and does another three detection results when the value of both S and I are smaller. Experiments show that the method compared to the traditional method using only the detection results of the brightness component improved substantial, reduced empty of the detected vehicle a large extent and increased traffic information data accuracy depending on the detection result.

scholarly journals A Novel Vehicle Detection Method Based on the Fusion of Radio Received Signal Strength and Geomagnetism

Sensors ◽  
2018 ◽  
Vol 19 (1) ◽  
pp. 58 ◽  
Author(s):  
Liangliang Lou ◽  
Jinyi Zhang ◽  
Yong Xiong ◽  
Yanliang Jin
Keyword(s):  
Error Detection ◽  
Detection Method ◽  
Signal Strength ◽  
Vehicle Detection ◽  
Received Signal Strength ◽  
Detection Accuracy ◽  
Detection Problem ◽  
Traffic Light ◽  
Radio Stations ◽  
Multiple Detection

A geomagnetic signal blind zone exists between the front and rear axle of high-chassis vehicle such as trucks and buses, which leads to multiple-detection problem in detecting those vehicles running at low speed on roads or error-detection problem in the case of the stopping position of the vehicle is not standard when waiting for the traffic light to change. In order to improve the detection accuracy of any type of vehicle running at any speed, a novel two-sensors data fusion vehicle detection method through combining received signal strength from radio stations with geomagnetism around vehicles is designed and verified in the paper. Experimental results show that the accuracy of our proposed method can reach 95.4% and traditional single magnetism-based detection method was only 83.4% in the detection of high-chassis vehicles.

scholarly journals Detecting Shilling Attacks with Automatic Features from Multiple Views

2019 ◽  
Vol 2019 ◽  
pp. 1-13 ◽  
Author(s):  
Yaojun Hao ◽  
Fuzhi Zhang ◽  
Jian Wang ◽  
Qingshan Zhao ◽  
Jianfang Cao
Keyword(s):  
Principal Component Analysis ◽  
Recommender Systems ◽  
Detection Method ◽  
Principal Component ◽  
Component Analysis ◽  
Experimental Results ◽  
Detection Methods ◽  
Detection Accuracy ◽  
Multiple Views ◽  
Fine Grained

Due to the openness of the recommender systems, the attackers are likely to inject a large number of fake profiles to bias the prediction of such systems. The traditional detection methods mainly rely on the artificial features, which are often extracted from one kind of user-generated information. In these methods, fine-grained interactions between users and items cannot be captured comprehensively, leading to the degradation of detection accuracy under various types of attacks. In this paper, we propose an ensemble detection method based on the automatic features extracted from multiple views. Firstly, to collaboratively discover the shilling profiles, the users’ behaviors are analyzed from multiple views including ratings, item popularity, and user-user graph. Secondly, based on the data preprocessed from multiple views, the stacked denoising autoencoders are used to automatically extract user features with different corruption rates. Moreover, the features extracted from multiple views are effectively combined based on principal component analysis. Finally, according to the features extracted with different corruption rates, the weak classifiers are generated and then integrated to detect attacks. The experimental results on the MovieLens, Netflix, and Amazon datasets indicate that the proposed method can effectively detect various attacks.

Fast Face Detection Based on AdaBoost and Canny Operators

2014 ◽  
Vol 716-717 ◽  
pp. 936-939
Author(s):  
Lin Zhang
Keyword(s):  
Face Detection ◽  
Detection Method ◽  
Face Image ◽  
Experimental Results ◽  
Detection Accuracy ◽  
Human Face ◽  
Adaboost Algorithm ◽  
Amount Of Information ◽  
Detection Speed

Detection speed of traditional face detection method based on AdaBoost algorithm is slow since AdaBoost asks a large number of features. Therefore, to address this shortcoming, we proposed a fast face detection method based on AdaBoost and canny operators in this paper. Firstly, we use canny operators to detect edge of face image which separates the region of the possible human face from image, and then do face detection in the separated region using Modest AdaBoost algorithm (MAB). Before using MAB to achieve face detection, utilizing canny operators to detect edge can make this algorithm effectively filter information, retain useful information, reduce the amount of information and improve detection speed. Experimental results show that the algorithm can obtain higher detection accuracy and detection speed has been significantly improved at the same time.

An anti-interference detection algorithm for parking monitoring systems

2019 ◽  
Vol 12 (4) ◽  
pp. 481-496
Author(s):  
Ani Dong ◽  
Zusheng Zhang ◽  
Jiaming Chen
Keyword(s):  
Data Fusion ◽  
Local Level ◽  
Detection Algorithm ◽  
Magnetic Sensors ◽  
Sensor Nodes ◽  
Detection Accuracy ◽  
Fusion Algorithm ◽  
Content Type ◽  
Occupancy Detection ◽  
Local Decision

Purpose Magnetic sensors have recently been proposed for parking occupancy detection. However, there has adjacent interference problem, i.e. the magnetic signal is easy to be interfered by the vehicles which are parking on adjacent spaces. The purpose of this paper is to propose a sensing algorithm to eliminate the adjacent interference. Design/methodology/approach The magnetic signals are converted to the pattern representation sequences, and the similarity is calculated using the pattern distance. The detection algorithm includes two levels: local decision and data fusion. In the local decision level, the sampled signals can be divided into three classes: vacant, occupied and uncertain. Then a collaborative decision is used to fusion the signals which belong to the uncertain class for the second level. Findings An experiment system included 60 sensor nodes that were deployed on bay parking spaces. Experiment results show that the proposed algorithm has better detection accuracy than existing algorithms. Originality/value This paper proposes a data fusion algorithm to eliminate adjacent interference. To balance the energy consumption and detection accuracy, the algorithm includes two levels: local decision and data fusion. In most of cases, the local decision can obtain the accurate detection result. Only the signals that cannot be correctly detected at the local level need data fusion operation.

scholarly journals All-in-One Framework for Detection, Unpacking, and Verification for Malware Analysis

2019 ◽  
Vol 2019 ◽  
pp. 1-16 ◽  
Author(s):  
Mi-Jung Choi ◽  
Jiwon Bang ◽  
Jongwook Kim ◽  
Hajin Kim ◽  
Yang-Sae Moon
Keyword(s):  
Detection Method ◽  
State Of The Art ◽  
Experimental Results ◽  
Detection Accuracy ◽  
Malware Analysis ◽  
Integrated Framework ◽  
Detection Techniques ◽  
Verification Algorithm ◽  
Extensive Analysis ◽  
Hybrid Detection

Packing is the most common analysis avoidance technique for hiding malware. Also, packing can make it harder for the security researcher to identify the behaviour of malware and increase the analysis time. In order to analyze the packed malware, we need to perform unpacking first to release the packing. In this paper, we focus on unpacking and its related technologies to analyze the packed malware. Through extensive analysis on previous unpacking studies, we pay attention to four important drawbacks: no phase integration, no detection combination, no real-restoration, and no unpacking verification. To resolve these four drawbacks, in this paper, we present an all-in-one structure of the unpacking system that performs packing detection, unpacking (i.e., restoration), and verification phases in an integrated framework. For this, we first greatly increase the packing detection accuracy in the detection phase by combining four existing and new packing detection techniques. We then improve the unpacking phase by using the state-of-the-art static and dynamic unpacking techniques. We also present a verification algorithm evaluating the accuracy of unpacking results. Experimental results show that the proposed all-in-one unpacking system performs all of the three phases well in an integrated framework. In particular, the proposed hybrid detection method is superior to the existing methods, and the system performs unpacking very well up to 100% of restoration accuracy for most of the files except for a few packers.

A Fast Forward Vehicle Detection Method by Multiple Vision Clues Fusion

2014 ◽  
Vol 543-547 ◽  
pp. 2647-2651
Author(s):  
Tai Qi Wu ◽  
Ye Zhang ◽  
Bin Bin Wang ◽  
Jia Heng Yu ◽  
De Wei Zhu
Keyword(s):  
Detection Method ◽  
System Development ◽  
Warning System ◽  
Vehicle Detection ◽  
Detection Methods ◽  
Vehicle Body ◽  
Detection Accuracy ◽  
Collision Warning ◽  
Moving Vehicle ◽  
Forward Collision Warning

With the development of intelligent vehicle technology, vehicle detection based on vision analysis has become an research hotspot in forward collision warning system development. Aiming to solve the existing problems in the current vehicle detection methods, for example, the detection accuracy is sensitive to the variation of illumination and object angle, we propose a forward moving vehicle detection method according to multiple vision clues fusion. Firstly, we locate the rough position using vehicle bottom shadow detection. The shadow is detected using an adaptive threshold image segmentation approach twice. Secondly, the symmetry of vehicle body and the perspective of camera field of view are both referenced to remove the inaccurate location in the first stage. The proposed method has been tested on several videos recorded in real urban conditions. Experimental results show that our method achieves 93.67% average detection accuracy in daytime, and its processing speed is more than 25fps. The proposed method has certain application prospects for improving the vision based forward collision warning system performance.

scholarly journals Vehicle Detection Using Deep Learning Technique in Tunnel Road Environments

Symmetry ◽  
2020 ◽  
Vol 12 (12) ◽  
pp. 2012
Author(s):  
JongBae Kim
Keyword(s):  
Deep Learning ◽  
Real Time ◽  
Detection Method ◽  
Vehicle Detection ◽  
Ground Truth ◽  
Noise Removal ◽  
Detection Accuracy ◽  
Vehicle Exhaust ◽  
Learning Techniques ◽  
Real Time Detection

This paper proposes a real-time detection method for a car driving ahead in real time on a tunnel road. Unlike the general road environment, the tunnel environment is irregular and has significantly lower illumination, including tunnel lighting and light reflected from driving vehicles. The environmental restrictions are large owing to pollution by vehicle exhaust gas. In the proposed method, a real-time detection method is used for vehicles in tunnel images learned in advance using deep learning techniques. To detect the vehicle region in the tunnel environment, brightness smoothing and noise removal processes are carried out. The vehicle region is learned after generating a learning image using the ground-truth method. The YOLO v2 model, with an optimal performance compared to the performances of deep learning algorithms, is applied. The training parameters are refined through experiments. The vehicle detection rate is approximately 87%, while the detection accuracy is approximately 94% for the proposed method applied to various tunnel road environments.

Sign in / Sign up

Export Citation Format

Share Document