scholarly journals Face Mask Wearing Detection Algorithm Based on Improved YOLO-v4

Sensors ◽  
2021 ◽  
Vol 21 (9) ◽  
pp. 3263
Author(s):  
Jimin Yu ◽  
Wei Zhang
Keyword(s):  
Semantic Information ◽  
Face Mask ◽  
Detection Algorithm ◽  
Frame Rate ◽  
Learning Ability ◽  
Complex Environment ◽  
Data Set ◽  
Evaluation Indexes ◽  
Image Scaling ◽  
Time Performance

To solve the problems of low accuracy, low real-time performance, poor robustness and others caused by the complex environment, this paper proposes a face mask recognition and standard wear detection algorithm based on the improved YOLO-v4. Firstly, an improved CSPDarkNet53 is introduced into the trunk feature extraction network, which reduces the computing cost of the network and improves the learning ability of the model. Secondly, the adaptive image scaling algorithm can reduce computation and redundancy effectively. Thirdly, the improved PANet structure is introduced so that the network has more semantic information in the feature layer. At last, a face mask detection data set is made according to the standard wearing of masks. Based on the object detection algorithm of deep learning, a variety of evaluation indexes are compared to evaluate the effectiveness of the model. The results of the comparations show that the mAP of face mask recognition can reach 98.3% and the frame rate is high at 54.57 FPS, which are more accurate compared with the exiting algorithm.

scholarly journals Distant Measurement of Plethysmographic Signal in Various Lighting Conditions Using Configurable Frame-Rate Camera

2016 ◽  
Vol 23 (4) ◽  
pp. 579-592 ◽  
Author(s):  
Jaromir Przybyło ◽  
Eliasz Kańtoch ◽  
Mirosław Jabłoński ◽  
Piotr Augustyniak
Keyword(s):  
Heart Rate ◽  
Assisted Living ◽  
Detection Algorithm ◽  
Frame Rate ◽  
Infrared Light ◽  
Fluorescent Light ◽  
Detection Accuracy ◽  
Rate Measurement ◽  
Lighting Conditions ◽  
Heart Rate Measurement

Abstract Videoplethysmography is currently recognized as a promising noninvasive heart rate measurement method advantageous for ubiquitous monitoring of humans in natural living conditions. Although the method is considered for application in several areas including telemedicine, sports and assisted living, its dependence on lighting conditions and camera performance is still not investigated enough. In this paper we report on research of various image acquisition aspects including the lighting spectrum, frame rate and compression. In the experimental part, we recorded five video sequences in various lighting conditions (fluorescent artificial light, dim daylight, infrared light, incandescent light bulb) using a programmable frame rate camera and a pulse oximeter as the reference. For a video sequence-based heart rate measurement we implemented a pulse detection algorithm based on the power spectral density, estimated using Welch’s technique. The results showed that lighting conditions and selected video camera settings including compression and the sampling frequency influence the heart rate detection accuracy. The average heart rate error also varies from 0.35 beats per minute (bpm) for fluorescent light to 6.6 bpm for dim daylight.

FPGA and ASIC implementation of the algorithm for traffic monitoring in urban areas

2011 ◽  
Vol 59 (2) ◽  
pp. 137-140 ◽  
Author(s):  
S. Szczepański ◽  
M. Wöjcikowski ◽  
B. Pankiewicz ◽  
M. KŁosowski ◽  
R. Żaglewski
Keyword(s):  
Urban Areas ◽  
Traffic Monitoring ◽  
Detection Algorithm ◽  
Frame Rate ◽  
Integrated Sensor ◽  
The Road ◽  
Moving Vehicles ◽  
Direction Of Movement ◽  
Account Due ◽  
Selection Of

FPGA and ASIC implementation of the algorithm for traffic monitoring in urban areas This paper describes the idea and the implementation of the image detection algorithm, that can be used in integrated sensor networks for environment and traffic monitoring in urban areas. The algorithm is dedicated to the extraction of moving vehicles from real-time camera images for the evaluation of traffic parameters, such as the number of vehicles, their direction of movement and their approximate speed. The authors, apart from the careful selection of particular steps of the algorithm towards hardware implementation, also proposed novel improvements, resulting in increasing the robustness and the efficiency. A single, stationary, monochrome camera is used, simple shadow and highlight elimination is performed. The occlusions are not taken into account, due to placing the camera at a location high above the road. The algorithm is designed and implemented in pipelined hardware, therefore high frame-rate efficiency has been achieved. The algorithm has been implemented and tested in FPGA and ASIC.

scholarly journals Pedestrian detection algorithm in traffic scene based on weakly supervised hierarchical deep model

2016 ◽  
Vol 14 (1) ◽  
pp. 172988141769231 ◽  
Author(s):  
Yingfeng Cai ◽  
Youguo He ◽  
Hai Wang ◽  
Xiaoqiang Sun ◽  
Long Chen ◽  
...  
Keyword(s):  
Deep Learning ◽  
Pedestrian Detection ◽  
Recognition Rate ◽  
Detection Algorithm ◽  
Training Methods ◽  
Two Dimensional ◽  
Data Set ◽  
Deep Model ◽  
Unsupervised Training ◽  
Weakly Supervised

The emergence and development of deep learning theory in machine learning field provide new method for visual-based pedestrian recognition technology. To achieve better performance in this application, an improved weakly supervised hierarchical deep learning pedestrian recognition algorithm with two-dimensional deep belief networks is proposed. The improvements are made by taking into consideration the weaknesses of structure and training methods of existing classifiers. First, traditional one-dimensional deep belief network is expanded to two-dimensional that allows image matrix to be loaded directly to preserve more information of a sample space. Then, a determination regularization term with small weight is added to the traditional unsupervised training objective function. By this modification, original unsupervised training is transformed to weakly supervised training. Subsequently, that gives the extracted features discrimination ability. Multiple sets of comparative experiments show that the performance of the proposed algorithm is better than other deep learning algorithms in recognition rate and outperforms most of the existing state-of-the-art methods in non-occlusion pedestrian data set while performs fair in weakly and heavily occlusion data set.

An automatic phase picker for local and teleseismic events

1987 ◽  
Vol 77 (4) ◽  
pp. 1437-1445
Author(s):  
M. Baer ◽  
U. Kradolfer
Keyword(s):  
Arrival Time ◽  
Detection Algorithm ◽  
Dynamic Threshold ◽  
Data Set ◽  
Automatic Phase ◽  
Nonlinear Amplifier ◽  
First Motion ◽  
Transient Events ◽  
Teleseismic Events ◽  
Automatic Detection Algorithm

Abstract An automatic detection algorithm has been developed which is capable of time P-phase arrivals of both local and teleseismic earthquakes, but rejects noise bursts and transient events. For each signal trace, the envelope function is calculated and passed through a nonlinear amplifier. The resulting signal is then subjected to a statistical analysis to yield arrival time, first motion, and a measure of reliability to be placed on the P-arrival pick. An incorporated dynamic threshold lets the algorithm become very sensitive; thus, even weak signals are timed precisely. During an extended performance evaluation on a data set comprising 789 P phases of local events and 1857 P phases of teleseismic events picked by an analyst, the automatic picker selected 66 per cent of the local phases and 90 per cent of the teleseismic phases. The accuracy of the automatic picks was “ideal” (i.e., could not be improved by the analyst) for 60 per cent of the local events and 63 per cent of the teleseismic events.

scholarly journals State model based face mask detection

2018 ◽  
Vol 7 (2.22) ◽  
pp. 35
Author(s):  
Kavitha M ◽  
Mohamed Mansoor Roomi S ◽  
K Priya ◽  
Bavithra Devi K
Keyword(s):  
Face Mask ◽  
Search Space ◽  
Gaussian Mixture ◽  
Detection Algorithm ◽  
State Model ◽  
Face Region ◽  
Automatic Teller Machine ◽  
Foreground Region ◽  
The Face ◽  
Surveillance Mechanism

The Automatic Teller Machine plays an important role in the modern economic society. ATM centers are located in remote central which are at high risk due to the increasing crime rate and robbery.These ATM centers assist with surveillance techniques to provide protection. Even after installing the surveillance mechanism, the robbers fool the security system by hiding their face using mask/helmet. Henceforth, an automatic mask detection algorithm is required to, alert when the ATM is at risk. In this work, the Gaussian Mixture Model (GMM) is applied for foreground detection to extract the regions of interest (ROI) i.e. Human being. Face region is acquired from the foreground region through  the torso partitioning and applying Viola-Jones algorithm in this search space. Parts of the face such as Eye pair, Nose, and Mouth are extracted and a state model is developed to detect  mask.  

scholarly journals Atmospheric rivers moisture sources from a Lagrangian perspective

2016 ◽  
Vol 7 (2) ◽  
pp. 371-384 ◽  
Author(s):  
Alexandre M. Ramos ◽  
Raquel Nieto ◽  
Ricardo Tomé ◽  
Luis Gimeno ◽  
Ricardo M. Trigo ◽  
...  
Keyword(s):  
North Atlantic ◽  
Dispersion Model ◽  
Local Maximum ◽  
Detection Algorithm ◽  
Ocean Basin ◽  
Particle Dispersion ◽  
Western Coast ◽  
Moisture Uptake ◽  
Lagrangian Analysis ◽  
Data Set

Abstract. An automated atmospheric river (AR) detection algorithm is used for the North Atlantic Ocean basin, allowing the identification of the major ARs affecting western European coasts between 1979 and 2012 over the winter half-year (October to March). The entire western coast of Europe was divided into five domains, namely the Iberian Peninsula (9.75° W, 36–43.75° N), France (4.5° W, 43.75–50° N), UK (4.5° W, 50–59° N), southern Scandinavia and the Netherlands (5.25° E, 50–59° N), and northern Scandinavia (5.25° E, 59–70° N). Following the identification of the main ARs that made landfall in western Europe, a Lagrangian analysis was then applied in order to identify the main areas where the moisture uptake was anomalous and contributed to the ARs reaching each domain. The Lagrangian data set used was obtained from the FLEXPART (FLEXible PARTicle dispersion) model global simulation from 1979 to 2012 and was forced by ERA-Interim reanalysis on a 1° latitude–longitude grid. The results show that, in general, for all regions considered, the major climatological areas for the anomalous moisture uptake extend along the subtropical North Atlantic, from the Florida Peninsula (northward of 20° N) to each sink region, with the nearest coast to each sink region always appearing as a local maximum. In addition, during AR events the Atlantic subtropical source is reinforced and displaced, with a slight northward movement of the sources found when the sink region is positioned at higher latitudes. In conclusion, the results confirm not only the anomalous advection of moisture linked to ARs from subtropical ocean areas but also the existence of a tropical source, together with midlatitude anomaly sources at some locations closer to AR landfalls.

scholarly journals Detour learning ability and the effect of novel sensory cues on learning in Australian bull ants, Myrmecia midas

2021 ◽  
Author(s):  
Muzahid Islam ◽  
Sudhakar Deeti ◽  
Zakia Mahmudah ◽  
J. Frances Kamhi ◽  
Ken Cheng
Keyword(s):  
Animal Cognition ◽  
Sensory Information ◽  
Learning Ability ◽  
Visual Environment ◽  
Sensory Cues ◽  
Complex Environment ◽  
Physical Barriers ◽  
Tactile Cues ◽  
Detour Learning ◽  
Do So

ABSTRACTMany animals navigate in a structurally complex environment which requires them to detour around physical barriers that they encounter. While many studies in animal cognition suggest that they are able to adeptly avoid obstacles, it is unclear whether a new route is learned to navigate around these barriers and, if so, what sensory information may be used to do so. We investigated detour learning ability in the Australian bull ant, Myrmecia midas, which primarily uses visual landmarks to navigate. We first placed a barrier on the ants’ natural path of their foraging tree. Initially, 46% of foragers were unsuccessful in detouring the obstacle. In subsequent trips, the ants became more successful and established a new route. We observed up to eight successful foraging trips detouring around the barrier. When we subsequently changed the position of the barrier, made a new gap in the middle of the obstacle, or removed the barrier altogether, ants mostly maintained their learned motor routine, detouring with a similar path as before, suggesting that foragers were not relying on barrier cues and therefore learned a new route around the obstacle. In additional trials, when foragers encountered new olfactory or tactile cues, or the visual environment was blocked, their navigation was profoundly disrupted. These results suggest that changing sensory information, even in modalities that foragers do not usually need for navigation, drastically affects the foragers’ ability to successful navigate.Subject CategoryNeuroscience and Cognition

A method of locating the 3D centers of retroreflectors based on deep learning

2021 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
BinBin Zhang ◽  
Fumin Zhang ◽  
Xinghua Qu
Keyword(s):  
Deep Learning ◽  
Data Augmentation ◽  
Measurement Techniques ◽  
Recognition Rate ◽  
Detection Algorithm ◽  
Large Field ◽  
Small Scale ◽  
Data Set ◽  
Content Type ◽  
Ellipse Detection

Purpose Laser-based measurement techniques offer various advantages over conventional measurement techniques, such as no-destructive, no-contact, fast and long measuring distance. In cooperative laser ranging systems, it’s crucial to extract center coordinates of retroreflectors to accomplish automatic measurement. To solve this problem, this paper aims to propose a novel method. Design/methodology/approach We propose a method using Mask RCNN (Region Convolutional Neural Network), with ResNet101 (Residual Network 101) and FPN (Feature Pyramid Network) as the backbone, to localize retroreflectors, realizing automatic recognition in different backgrounds. Compared with two other deep learning algorithms, experiments show that the recognition rate of Mask RCNN is better especially for small-scale targets. Based on this, an ellipse detection algorithm is introduced to obtain the ellipses of retroreflectors from recognized target areas. The center coordinates of retroreflectors in the camera coordinate system are obtained by using a mathematics method. Findings To verify the accuracy of this method, an experiment was carried out: the distance between two retroreflectors with a known distance of 1,000.109 mm was measured, with 2.596 mm root-mean-squar error, meeting the requirements of the coarse location of retroreflectors. Research limitations/implications The research limitations/implications are as follows: (i) As the data set only has 200 pictures, although we have used some data augmentation methods such as rotating, mirroring and cropping, there is still room for improvement in the generalization ability of detection. (ii) The ellipse detection algorithm needs to work in relatively dark conditions, as the retroreflector is made of stainless steel, which easily reflects light. Originality/value The originality/value of the article lies in being able to obtain center coordinates of multiple retroreflectors automatically even in a cluttered background; being able to recognize retroreflectors with different sizes, especially for small targets; meeting the recognition requirement of multiple targets in a large field of view and obtaining 3 D centers of targets by monocular model-based vision.

Sign in / Sign up

Export Citation Format

Share Document