Person Re-Identification Based on Significant Color With the Spatial Correspondence

2021 ◽  
Vol 12 (1) ◽  
pp. 20-36
Author(s):  
Vidhyalakshmi M. K. ◽  
Poovammal E. ◽  
Masilamani V. ◽  
Vidhyacharan Bhaskar
Keyword(s):  
Real Time ◽  
Video Surveillance ◽  
Characteristic Curve ◽  
Field Of View ◽  
Data Sets ◽  
Computationally Efficient ◽  
Criminal Offense ◽  
Spatial Correspondence ◽  
Standard Data ◽  
Identification Technique

Video surveillance has played a key role to find an individual just in case of a criminal offense. More studies were done to make the surveillance process autonomous. In this, the person re-identification technique helps to identify people. The surveillance cameras are normally mounted at a height above the head of a person. With such a position of camera, it is difficult to identify the person. Therefore, video surveillance is an application in real time. The images of the same individual may vary appreciably based on different camera field of view. Color content in an image remains an important cue to identify a person. Under the assumption that the clothing color remains unchanged over the period of surveillance, a method based on significant colors with its spatial correspondence in image is proposed. The method is applied on standard data sets like GRID, PRID450s and VIPER. The results are plotted as cumulative matching characteristic curve and compared with other methods. The approach is both computationally efficient and delivers better performance.

User Power Behavior Similarity Clustering Based on Unsupervised Extreme Learning Machine Algorithm

2020 ◽  
Vol 13 (5) ◽  
pp. 641-649
Author(s):  
Yuancheng Li ◽  
Yaqi Cui ◽  
Xiaolong Zhang
Keyword(s):  
Extreme Learning Machine ◽  
Clustering Algorithm ◽  
Characteristic Curve ◽  
Clustering Algorithms ◽  
Data Sets ◽  
Residential Areas ◽  
Processing Power ◽  
Learning Machine ◽  
Advanced Metering ◽  
Matlab Programming

Background: Advanced Metering Infrastructure (AMI) for the smart grid is growing rapidly which results in the exponential growth of data collected and transmitted in the device. By clustering this data, it can give the electricity company a better understanding of the personalized and differentiated needs of the user. Objective: The existing clustering algorithms for processing data generally have some problems, such as insufficient data utilization, high computational complexity and low accuracy of behavior recognition. Methods: In order to improve the clustering accuracy, this paper proposes a new clustering method based on the electrical behavior of the user. Starting with the analysis of user load characteristics, the user electricity data samples were constructed. The daily load characteristic curve was extracted through improved extreme learning machine clustering algorithm and effective index criteria. Moreover, clustering analysis was carried out for different users from industrial areas, commercial areas and residential areas. The improved extreme learning machine algorithm, also called Unsupervised Extreme Learning Machine (US-ELM), is an extension and improvement of the original Extreme Learning Machine (ELM), which realizes the unsupervised clustering task on the basis of the original ELM. Results: Four different data sets have been experimented and compared with other commonly used clustering algorithms by MATLAB programming. The experimental results show that the US-ELM algorithm has higher accuracy in processing power data. Conclusion: The unsupervised ELM algorithm can greatly reduce the time consumption and improve the effectiveness of clustering.

Safety Monitoring and Warning System for Subway Construction Workers using Wearable Technology

2019 ◽  
Vol 13 ◽  
Author(s):  
Jun-hua Chen ◽  
Da-hu Wang ◽  
Cun-yuan Sun
Keyword(s):  
Real Time ◽  
Video Surveillance ◽  
Early Warning ◽  
Warning System ◽  
Safety Monitoring ◽  
Wearable Technology ◽  
Construction Workers ◽  
Great Success ◽  
Subway Construction ◽  
Monitoring And Early Warning

Objective: This study focused on the application of wearable technology in the safety monitoring and early warning for subway construction workers. Methods: With the help of real-time video surveillance and RFID positioning which was applied in the construction has realized the real-time monitoring and early warning of on-site construction to a certain extent, but there are still some problems. Real-time video surveillance technology relies on monitoring equipment, while the location of the equipment is fixed, so it is difficult to meet the full coverage of the construction site. However, wearable technologies can solve this problem, they have outstanding performance in collecting workers’ information, especially physiological state data and positioning data. Meanwhile, wearable technology has no impact on work and is not subject to the inference of dynamic environment. Results and conclusion: The first time the system applied to subway construction was a great success. During the construction of the station, the number of occurrences of safety warnings was 43 times, but the number of occurrences of safety accidents was 0, which showed that the safety monitoring and early warning system played a significant role and worked out perfectly.

Real-time Approximation of Photometric Polygonal Lights

2020 ◽  
Vol 3 (1) ◽  
pp. 1-18
Author(s):  
Christian Luksch ◽  
Lukas Prost ◽  
Michael Wimmer
Keyword(s):  
Real Time ◽  
Specular Reflection ◽  
Near Field ◽  
Measurement Data ◽  
Data Sets ◽  
Photometric Measurement ◽  
Integration Technique ◽  
Time Approximation ◽  
Light Emitter ◽  
Selection Of

We present a real-time rendering technique for photometric polygonal lights. Our method uses a numerical integration technique based on a triangulation to calculate noise-free diffuse shading. We include a dynamic point in the triangulation that provides a continuous near-field illumination resembling the shape of the light emitter and its characteristics. We evaluate the accuracy of our approach with a diverse selection of photometric measurement data sets in a comprehensive benchmark framework. Furthermore, we provide an extension for specular reflection on surfaces with arbitrary roughness that facilitates the use of existing real-time shading techniques. Our technique is easy to integrate into real-time rendering systems and extends the range of possible applications with photometric area lights.

scholarly journals 3D B-Rep meshing for real-time data-based geometric parametric analysis

2021 ◽  
Vol 8 (1) ◽  
Author(s):  
Tristan Maquart ◽  
Thomas Elguedj ◽  
Anthony Gravouil ◽  
Michel Rochette
Keyword(s):  
Real Time ◽  
Surface Topology ◽  
Time Data ◽  
Standard Data ◽  
Input Surface ◽  
Cad Models ◽  
Pants Decomposition ◽  
Parametric Studies ◽  
Geometry Feature ◽  
Complicated Geometry

AbstractThis paper presents an effective framework to automatically construct 3D quadrilateral meshes of complicated geometry and arbitrary topology adapted for parametric studies. The input is a triangulation of the solid 3D model’s boundary provided from B-Rep CAD models or scanned geometry. The triangulated mesh is decomposed into a set of cuboids in two steps: pants decomposition and cuboid decomposition. This workflow includes an integration of a geometry-feature-aware pants-to-cuboids decomposition algorithm. This set of cuboids perfectly replicates the input surface topology. Using aligned global parameterization, patches are re-positioned on the surface in a way to achieve low overall distortion, and alignment to principal curvature directions and sharp features. Based on the cuboid decomposition and global parameterization, a 3D quadrilateral mesh is extracted. For different parametric instances with the same topology but different geometries, the MEG-IsoQuad method allows to have the same representation: isotopological meshes holding the same connectivity where each point on a mesh has an analogous one into all other meshes. Faithful 3D numerical charts of parametric geometries are then built using standard data-based techniques. Geometries are then evaluated in real-time. The efficiency and the robustness of the proposed approach are illustrated through a few parametric examples.

Accelerating real‐time object detection in high‐resolution video surveillance

2021 ◽  
Author(s):  
Yuefeng Wang ◽  
Kuang Mao ◽  
Tong Chen ◽  
Yanglong Yin ◽  
Shuibing He ◽  
...  
Keyword(s):  
High Resolution ◽  
Object Detection ◽  
Real Time ◽  
Video Surveillance

Unmanned aerial vehicle reconnaissance image recognition based on convolutional neural network

2021 ◽  
pp. 1-10
Author(s):  
Lipeng Si ◽  
Baolong Liu ◽  
Yanfang Fu
Keyword(s):  
Neural Network ◽  
Real Time ◽  
Image Recognition ◽  
Error Detection ◽  
Convolution Neural Network ◽  
Data Sets ◽  
Small Target ◽  
Network Algorithm ◽  
Neural Network Algorithm ◽  
Real Time Tracking

The important strategic position of military UAVs and the wide application of civil UAVs in many fields, they all mark the arrival of the era of unmanned aerial vehicles. At present, in the field of image research, recognition and real-time tracking of specific objects in images has been a technology that many scholars continue to study in depth and need to be further tackled. Image recognition and real-time tracking technology has been widely used in UAV aerial photography. Through the analysis of convolution neural network algorithm and the comparison of image recognition technology, the convolution neural network algorithm is improved to improve the image recognition effect. In this paper, a target detection technique based on improved Faster R-CNN is proposed. The algorithm model is implemented and the classification accuracy is improved through Faster R-CNN network optimization. Aiming at the problem of small target error detection and scale difference in aerial data sets, this paper designs the network structure of RPN and the optimization scheme of related algorithms. The structure of Faster R-CNN is adjusted by improving the embedding of CNN and OHEM algorithm, the accuracy of small target and multitarget detection is improved as a whole. The experimental results show that: compared with LENET-5, the recognition accuracy of the proposed algorithm is significantly improved. And with the increase of the number of samples, the accuracy of this algorithm is 98.9%.

RSNet: Rail semantic segmentation network for extracting aerial railroad images

2021 ◽  
pp. 1-18
Author(s):  
R.S. Rampriya ◽  
Sabarinathan ◽  
R. Suganya
Keyword(s):  
Real Time ◽  
Visual Processing ◽  
Feature Fusion ◽  
Semantic Segmentation ◽  
Vital Role ◽  
Obstacle Detection ◽  
Aerial Images ◽  
Computationally Efficient ◽  
Fusion Algorithm ◽  
Uav Images

In the near future, combo of UAV (Unmanned Aerial Vehicle) and computer vision will play a vital role in monitoring the condition of the railroad periodically to ensure passenger safety. The most significant module involved in railroad visual processing is obstacle detection, in which caution is obstacle fallen near track gage inside or outside. This leads to the importance of detecting and segment the railroad as three key regions, such as gage inside, rails, and background. Traditional railroad segmentation methods depend on either manual feature selection or expensive dedicated devices such as Lidar, which is typically less reliable in railroad semantic segmentation. Also, cameras mounted on moving vehicles like a drone can produce high-resolution images, so segmenting precise pixel information from those aerial images has been challenging due to the railroad surroundings chaos. RSNet is a multi-level feature fusion algorithm for segmenting railroad aerial images captured by UAV and proposes an attention-based efficient convolutional encoder for feature extraction, which is robust and computationally efficient and modified residual decoder for segmentation which considers only essential features and produces less overhead with higher performance even in real-time railroad drone imagery. The network is trained and tested on a railroad scenic view segmentation dataset (RSSD), which we have built from real-time UAV images and achieves 0.973 dice coefficient and 0.94 jaccard on test data that exhibits better results compared to the existing approaches like a residual unit and residual squeeze net.

scholarly journals MUREN: a robust and multi-reference approach of RNA-seq transcript normalization

2021 ◽  
Vol 22 (1) ◽  
Author(s):  
Yance Feng ◽  
Lei M. Li
Keyword(s):  
Biological Significance ◽  
Housekeeping Genes ◽  
R Package ◽  
Data Sets ◽  
Statistical Regression ◽  
Rna Seq ◽  
Least Trimmed Squares ◽  
Standard Data ◽  
Wide Range ◽  
Multiple References

Abstract Background Normalization of RNA-seq data aims at identifying biological expression differentiation between samples by removing the effects of unwanted confounding factors. Explicitly or implicitly, the justification of normalization requires a set of housekeeping genes. However, the existence of housekeeping genes common for a very large collection of samples, especially under a wide range of conditions, is questionable. Results We propose to carry out pairwise normalization with respect to multiple references, selected from representative samples. Then the pairwise intermediates are integrated based on a linear model that adjusts the reference effects. Motivated by the notion of housekeeping genes and their statistical counterparts, we adopt the robust least trimmed squares regression in pairwise normalization. The proposed method (MUREN) is compared with other existing tools on some standard data sets. The goodness of normalization emphasizes on preserving possible asymmetric differentiation, whose biological significance is exemplified by a single cell data of cell cycle. MUREN is implemented as an R package. The code under license GPL-3 is available on the github platform: github.com/hippo-yf/MUREN and on the conda platform: anaconda.org/hippo-yf/r-muren. Conclusions MUREN performs the RNA-seq normalization using a two-step statistical regression induced from a general principle. We propose that the densities of pairwise differentiations are used to evaluate the goodness of normalization. MUREN adjusts the mode of differentiation toward zero while preserving the skewness due to biological asymmetric differentiation. Moreover, by robustly integrating pre-normalized counts with respect to multiple references, MUREN is immune to individual outlier samples.

Facial Nerve EMG: Low-Tech Monitoring with a Stopwatch

2021 ◽  
Author(s):  
Julian Prell ◽  
Christian Scheller ◽  
Sebastian Simmermacher ◽  
Christian Strauss ◽  
Stefan Rampp
Keyword(s):  
Facial Nerve ◽  
Real Time ◽  
Random Order ◽  
Automated Analysis ◽  
Data Sets ◽  
Time Analysis ◽  
Frequency Pattern ◽  
Real Time Analysis ◽  
Offline Analysis ◽  
Automated Algorithms

Abstract Objective The quantity of A-trains, a high-frequency pattern of free-running facial nerve electromyography, is correlated with the risk for postoperative high-grade facial nerve paresis. This correlation has been confirmed by automated analysis with dedicated algorithms and by visual offline analysis but not by audiovisual real-time analysis. Methods An investigator was presented with 29 complete data sets measured during actual surgeries in real time and without breaks in a random order. Data were presented either strictly via loudspeaker (audio) or simultaneously by loudspeaker and computer screen (audiovisual). Visible and/or audible A-train activity was then quantified by the investigator with the computerized equivalent of a stopwatch. The same data were also analyzed with quantification of A-trains by automated algorithms. Results Automated (auto) traintime (TT), known to be a small, yet highly representative fraction of overall A-train activity, ranged from 0.01 to 10.86 s (median: 0.58 s). In contrast, audio-TT ranged from 0 to 1,357.44 s (median: 29.69 s), and audiovisual-TT ranged from 0 to 786.57 s (median: 46.19 s). All three modalities were correlated to each other in a highly significant way. Likewise, all three modalities correlated significantly with the extent of postoperative facial paresis. As a rule of thumb, patients with visible/audible A-train activity < 1 minute presented with a more favorable clinical outcome than patients with > 1 minute of A-train activity. Conclusion Detection and even quantification of A-trains is technically possible not only with intraoperative automated real-time calculation or postoperative visual offline analysis, but also with very basic monitoring equipment and real-time good quality audiovisual analysis. However, the investigator found audiovisual real-time-analysis to be very demanding; thus tools for automated quantification can be very helpful in this respect.

Sign in / Sign up

Export Citation Format

Share Document