Online detection system for electric bike in elevator or corridors based on multi-scale fusion

2021 ◽  
Author(s):  
Yibin Lin ◽  
Xiangrong Chen ◽  
Weizhao Zhong ◽  
Zhixin Pan
Keyword(s):  
Detection System ◽  
Online Detection ◽  
Multi Scale ◽  
Electric Bike

scholarly journals Multi-Scale Feature Pyramid Network: A Heavily Occluded Pedestrian Detection Network Based on ResNet

Sensors ◽  
2021 ◽  
Vol 21 (5) ◽  
pp. 1820
Author(s):  
Xiaotao Shao ◽  
Qing Wang ◽  
Wei Yang ◽  
Yun Chen ◽  
Yi Xie ◽  
...  
Keyword(s):  
Semantic Information ◽  
Detection System ◽  
Pedestrian Detection ◽  
Detection Accuracy ◽  
The Public ◽  
Scale Feature ◽  
Detection Algorithms ◽  
Multi Scale ◽  
Art Works ◽  
Feature Pyramid

The existing pedestrian detection algorithms cannot effectively extract features of heavily occluded targets which results in lower detection accuracy. To solve the heavy occlusion in crowds, we propose a multi-scale feature pyramid network based on ResNet (MFPN) to enhance the features of occluded targets and improve the detection accuracy. MFPN includes two modules, namely double feature pyramid network (FPN) integrated with ResNet (DFR) and repulsion loss of minimum (RLM). We propose the double FPN which improves the architecture to further enhance the semantic information and contours of occluded pedestrians, and provide a new way for feature extraction of occluded targets. The features extracted by our network can be more separated and clearer, especially those heavily occluded pedestrians. Repulsion loss is introduced to improve the loss function which can keep predicted boxes away from the ground truths of the unrelated targets. Experiments carried out on the public CrowdHuman dataset, we obtain 90.96% AP which yields the best performance, 5.16% AP gains compared to the FPN-ResNet50 baseline. Compared with the state-of-the-art works, the performance of the pedestrian detection system has been boosted with our method.

Mathematical model establishment for cable eccentricity online detection system based on X-ray

2017 ◽  
Author(s):  
Lili Wang ◽  
Chuansheng Wang ◽  
Jingyuan Cheng ◽  
Tao Zhang ◽  
Huihui Wang
Keyword(s):  
Mathematical Model ◽  
Detection System ◽  
Online Detection ◽  
X Ray

An Online Detection System for Atomic Beam

2012 ◽  
Vol 39 (5) ◽  
pp. 0508004
Author(s):  
王晓佳 Wang Xiaojia ◽  
焱颖 Feng Yanying ◽  
薛洪波 Xue Hongbo
Keyword(s):  
Atomic Beam ◽  
Detection System ◽  
Online Detection

Powerful conveyer belt real-time online detection system based on x-ray

2009 ◽  
Author(s):  
Feng Rong ◽  
Chang-yun Miao ◽  
Wei Meng
Keyword(s):  
Real Time ◽  
Detection System ◽  
Conveyer Belt ◽  
Online Detection ◽  
X Ray

The Impact Mechanism of Fractal Noise on PN Code Detection System

2014 ◽  
Vol 13 (02) ◽  
pp. 1450017 ◽  
Author(s):  
Gang Xiong ◽  
Shu-Ning Zhang ◽  
Hui-Chang Zhao
Keyword(s):  
White Noise ◽  
Detection System ◽  
Random Phase ◽  
Average Power ◽  
Fractal Model ◽  
Sea Clutter ◽  
Impact Mechanism ◽  
Multi Scale ◽  
Fractal Noise ◽  
The Impact

Taking the pseudo-random phase modulated CW radar for example, this paper studies the impact mechanism of a class of non-stationary fractal noise on PN code detection system, especially signal mixing and matching filter. The cross correlation function, power spectrum function and average power of pseudo-random signal and fractal noise are deduced, compared with the impact of white noise on the pseudo code detection system. We analyze the impact mechanism of three kinds of sea clutter model, namely fractal Brownian model (FBM), the multifractal (MF) model and the non-stationary random fractal model (e.g., infinitely divisible cascades, IDC), on the pseudo-random code detection system, and demonstrate the reason why the multi-scale filtering method in wavelet domain and the MF methods fail to eliminate the effect of sea clutter. Based on the natural sea clutter data, we simulate and analyze the influence of white noise and fractal noise comparatively on detection system, which indicates that the effect of fractal noise cannot be inhibited effectively by the traditional correlation detection and MF analysis, and finally we put forward possible solutions.

Design of the scanning mode coated glass color difference online detection system

2007 ◽  
Author(s):  
Weihong Bi ◽  
Yu Zhang ◽  
Dajiang Wang ◽  
Baojun Zhang ◽  
Guangwei Fu
Keyword(s):  
Detection System ◽  
Color Difference ◽  
Online Detection ◽  
Coated Glass ◽  
Glass Color

scholarly journals A Noble Approach of Real Time Intrusion Detection System (NART-IDS)

2019 ◽  
pp. 10-22
Author(s):  
Deepak Kumar Yadav ◽  
Akhilesh Bansiya
Keyword(s):  
Intrusion Detection ◽  
Real Time ◽  
Intrusion Detection System ◽  
Detection System ◽  
Critical Systems ◽  
Online Detection ◽  
Report Generation ◽  
Improve Efficiency ◽  
Network Intrusion ◽  
System Overhead

Malicious users use different techniques such as cracking passwords, text traffic, sniffing unencrypted or light, etc. System overhead and compromise critical systems. Therefore, there must be some sort of security for the organization's private resources from the Internet and from the inside. Therefore, the intrusion detection system (IDS) could be the best solution. It complements the firewall to improve the security holes. An intrusion detection system includes a management console and sensors. The management console holds all the responsibility of functionality of IDS comprises with its initialization, packet capturing, and report generation, whereas the sensors used to monitor hosts or networks in real time. There may be different categories of Intrusion Detection System. IDS can be designed in the concept of Signature analysis as well as anomaly behavior analysis. Therefore IDS used to capture the behavior of suspected packets. These functions are in host mode and called as Host Intrusion Detection System (HIDS) and in Network mode called as Network Intrusion Detection System (NIDS). The entitled dissertation work is carried out to obtain the best analysis performance through signature based detection system. It is efficient for host as well as network system .here basically Transmission Control Packets (TCP) and User Datagram Packets (UDP) considered to analysis for finding different attacks like Probe,DoS,R2l and U2R. This system is being found functionally efficient and also provide layer wise attacks details. Here different agent modules used to perform desired isolated responsibility like Mobile Agent (MA) to activate different IDS chest at different hosts, Tenet Agent (TA) for signature rule, Analysis Agent (AA) etc. The proposed system can greatly improve efficiency from offline detection to real-time online detection. Since the proposed system derives features from packet headers. Many attacks were experimented in this system. Experiments were performed to demonstrate the excellent effectiveness and efficiency of the proposed system. The proposed system can greatly improve efficiency from offline detection to real-time online detection. Since the proposed system derives features from packet headers. The entitled system can be further enhanced to capture more type of attacks at the levels of multiple layers and also may stop attacks as well.

Sign in / Sign up

Export Citation Format

Share Document