scholarly journals Night-Time Vehicle Sensing in Far Infrared Image with Deep Learning

2016 ◽  
Vol 2016 ◽  
pp. 1-8 ◽  
Author(s):  
Hai Wang ◽  
Yingfeng Cai ◽  
Xiaobo Chen ◽  
Long Chen
Keyword(s):  
Detection Rate ◽  
False Positive Rate ◽  
Infrared Image ◽  
Far Infrared ◽  
Night Vision ◽  
Infrared Sensors ◽  
Night Time ◽  
Detection Rates ◽  
Positive Rate ◽  
And Performance

The use of night vision systems in vehicles is becoming increasingly common. Several approaches using infrared sensors have been proposed in the literature to detect vehicles in far infrared (FIR) images. However, these systems still have low vehicle detection rates and performance could be improved. This paper presents a novel method to detect vehicles using a far infrared automotive sensor. Firstly, vehicle candidates are generated using a constant threshold from the infrared frame. Contours are then generated by using a local adaptive threshold based on maximum distance, which decreases the number of processing regions for classification and reduces the false positive rate. Finally, vehicle candidates are verified using a deep belief network (DBN) based classifier. The detection rate is 93.9% which is achieved on a database of 5000 images and video streams. This result is approximately a 2.5% improvement on previously reported methods and the false detection rate is also the lowest among them.

scholarly journals Night-Time Vehicle Detection Algorithm Based on Visual Saliency and Deep Learning

2016 ◽  
Vol 2016 ◽  
pp. 1-7 ◽  
Author(s):  
Yingfeng Cai ◽  
Xiaoqiang Sun ◽  
Hai Wang ◽  
Long Chen ◽  
Haobin Jiang
Keyword(s):  
Deep Learning ◽  
Detection Rate ◽  
Processing Time ◽  
Infrared Image ◽  
Visual Saliency ◽  
Vehicle Detection ◽  
Far Infrared ◽  
Detection Algorithm ◽  
Night Vision ◽  
Night Time

Night vision systems get more and more attention in the field of automotive active safety field. In this area, a number of researchers have proposed far-infrared sensor based night-time vehicle detection algorithm. However, existing algorithms have low performance in some indicators such as the detection rate and processing time. To solve this problem, we propose a far-infrared image vehicle detection algorithm based on visual saliency and deep learning. Firstly, most of the nonvehicle pixels will be removed with visual saliency computation. Then, vehicle candidate will be generated by using prior information such as camera parameters and vehicle size. Finally, classifier trained with deep belief networks will be applied to verify the candidates generated in last step. The proposed algorithm is tested in around 6000 images and achieves detection rate of 92.3% and processing time of 25 Hz which is better than existing methods.

The efficacy of combining several risk factors as a screening test

2005 ◽  
Vol 12 (4) ◽  
pp. 197-201 ◽  
Author(s):  
Nicholas J Wald ◽  
Joan K Morris ◽  
Simon Rish
Keyword(s):  
Risk Factors ◽  
Risk Factor ◽  
False Positive ◽  
Detection Rate ◽  
False Positive Rate ◽  
Individual Risk ◽  
Detection Rates ◽  
Screening Performance ◽  
Positive Rate ◽  
The Individual

Objective: To determine the quantitative effect on overall screening performance (detection rate for a given false-positive rate) of using several moderately strong, independent risk factors in combination as screening markers. Setting: Theoretical statistical analysis. Methods: For the purposes of this analysis, it was assumed that all risk factors were independent, had Gaussian distributions with the same standard deviation in affected and unaffected individuals and had the same screening performance. We determined the overall screening performance associated with using an increasing number of risk factors together, with each risk factor having a detection rate of 10%, 15% or 20% for a 5% false-positive rate. The overall screening performance was estimated as the detection rate for a 5% false-positive rate. Results: Combining the risk factors increased the screening performance, but the gain in detection at a constant false-positive rate was relatively modest and diminished with the addition of each risk factor. Combining three risk factors, each with a 15% detection rate for a 5% false-positive rate, yields a 28% detection rate. Combining five risk factors increases the detection rate to 39%. If the individual risk factors have a detection rate of 10% for a 5% false-positive rate, it would require combining about 15 such risk factors to achieve a comparable overall detection rate (41%). Conclusion: It is intuitively thought that combining moderately strong risk factors can substantially improve screening performance. For example, most cardiovascular risk factors that may be used in screening for ischaemic heart disease events, such as serum cholesterol and blood pressure, have a relatively modest screening performance (about 15% detection rate for a 5% false-positive rate). It would require the combination of about 15 or 20 such risk factors to achieve detection rates of about 80% for a 5% false-positive rate. This is impractical, given the risk factors so far discovered, because there are too few risk factors and their associations with disease are too weak.

scholarly journals Starting the polypill: the use of a single age cut-off in males and females

2016 ◽  
Vol 24 (1) ◽  
pp. 50-53 ◽  
Author(s):  
Nicholas J Wald ◽  
Johannes M Luteijn ◽  
Joan K Morris
Keyword(s):  
Myocardial Infarction ◽  
False Positive ◽  
Detection Rate ◽  
False Positive Rate ◽  
Rate Sensitivity ◽  
Detection Rates ◽  
Preventive Medication ◽  
Positive Rate ◽  
Males And Females ◽  
First Myocardial Infarction

Objective Age screening and preventive medication for future myocardial infarction and stroke has been previously described. We aimed to ascertain whether different age cut-offs are needed for males and females. Methods We determined five parameters for each sex according to age cut-off: detection rate (sensitivity), false-positive rate, proportion of the population eligible for treatment with a polypill, proportion who benefit from taking a polypill (simvastatin 20 mg, losartan 25 mg, hydrochlorothiazide 12.5 mg, amlodipine 2.5 mg), and among these, years of life gained without a first myocardial infarction or stroke. Results Approximately one-third benefit, regardless of the age cut-off. For males and females combined, using ages 40 and 80, the detection rates are 98% and 52%, false-positive rates are 51% and 7%, population percentages eligible for treatment are 52% and 7%, and years of life gained without a first myocardial infarction or stroke are 8.4 and 3.6. Using age 50, detection rates are 93% (males) 98% (females), false-positive rates 37% (males) 40% (females), percentage of the population eligible for treatment 38% (males) 41% (females), percentage who benefit 35% (males) 33% (females), and years of life gained without an event 8.5 (males) 7.0 (females). At a given age cut-off, the sex differences are relatively small. Conclusion A single age cut-off can be used for both sexes.

Predicting Fetal Chromosome Anomalies in the First Trimester Using Pregnancy Associated Plasma Protein-A: A Comparison of Statistical Methods

1993 ◽  
Vol 32 (02) ◽  
pp. 175-179 ◽  
Author(s):  
B. Brambati ◽  
T. Chard ◽  
J. G. Grudzinskas ◽  
M. C. M. Macintosh
Keyword(s):  
Logistic Regression ◽  
General Population ◽  
Likelihood Ratio ◽  
False Positive ◽  
False Positive Rate ◽  
Ratio Method ◽  
Detection Rates ◽  
Gaussian Distributions ◽  
Positive Rate ◽  
Likelihood Ratio Method

Abstract:The analysis of the clinical efficiency of a biochemical parameter in the prediction of chromosome anomalies is described, using a database of 475 cases including 30 abnormalities. A comparison was made of two different approaches to the statistical analysis: the use of Gaussian frequency distributions and likelihood ratios, and logistic regression. Both methods computed that for a 5% false-positive rate approximately 60% of anomalies are detected on the basis of maternal age and serum PAPP-A. The logistic regression analysis is appropriate where the outcome variable (chromosome anomaly) is binary and the detection rates refer to the original data only. The likelihood ratio method is used to predict the outcome in the general population. The latter method depends on the data or some transformation of the data fitting a known frequency distribution (Gaussian in this case). The precision of the predicted detection rates is limited by the small sample of abnormals (30 cases). Varying the means and standard deviations (to the limits of their 95% confidence intervals) of the fitted log Gaussian distributions resulted in a detection rate varying between 42% and 79% for a 5% false-positive rate. Thus, although the likelihood ratio method is potentially the better method in determining the usefulness of a test in the general population, larger numbers of abnormal cases are required to stabilise the means and standard deviations of the fitted log Gaussian distributions.

Association Rule-Mining-Based Intrusion Detection System With Entropy-Based Feature Selection

2020 ◽  
pp. 1-24
Author(s):  
Devaraju Sellappan ◽  
Ramakrishnan Srinivasan
Keyword(s):  
Intrusion Detection ◽  
Association Rule ◽  
Intrusion Detection System ◽  
Association Rule Mining ◽  
Detection System ◽  
False Positive Rate ◽  
Rule Mining ◽  
Detection Rates ◽  
Mining Algorithm ◽  
Positive Rate

Intrusion detection system (IDSs) are important to industries and organizations to solve the problems of networks, and various classifiers are used to classify the activity as malicious or normal. Today, the security has become a decisive part of any industrial and organizational information system. This chapter demonstrates an association rule-mining algorithm for detecting various network intrusions. The KDD dataset is used for experimentation. There are three input features classified as basic features, content features, and traffic features. There are several attacks are present in the dataset which are classified into Denial of Service (DoS), Probe, Remote to Local (R2L), and User to Root (U2R). The proposed method gives significant improvement in the detection rates compared with other methods. Association rule mining algorithm is proposed to evaluate the KDD dataset and dynamic data to improve the efficiency, reduce the false positive rate (FPR) and provides less time for processing.

Impact of Including or Removing Nuchal Translucency Measurement on the Detection and False-Positive Rates of First-Trimester Down Syndrome Screening

2015 ◽  
Vol 40 (3) ◽  
pp. 214-218 ◽  
Author(s):  
Emmanuel Spaggiari ◽  
Isabelle Czerkiewicz ◽  
Corinne Sault ◽  
Sophie Dreux ◽  
Armelle Galland ◽  
...  
Keyword(s):  
Down Syndrome ◽  
False Positive ◽  
False Positive Rate ◽  
Method Comparison ◽  
First Trimester ◽  
Nuchal Translucency ◽  
Routine Practice ◽  
Detection Rates ◽  
Significant Difference ◽  
Positive Rate

Introduction: First-trimester Down syndrome (DS) screening combining maternal age, serum markers (pregnancy-associated plasma protein-A and beta-human chorionic gonadotropin) and nuchal translucency (NT) gives an 85% detection rate for a 5% false-positive rate. These results largely depend on quality assessment of biochemical markers and of NT. In routine practice, despite an ultrasound quality control organization, NT images can be considered inadequate. The aim of the study was to evaluate the consequences for risk calculation when NT measurement is not taken into account. Material and Method: Comparison of detection and false-positive rates of first-trimester DS screening (PerkinElmer, Turku, Finland), with and without NT, based on a retrospective study of 117,126 patients including 274 trisomy 21-affected fetuses. NT was measured by more than 3,000 certified sonographers. Results: There was no significant difference in detection rates between the two strategies including or excluding NT measurement (86.7 vs. 81.8%). However, there was a significant difference in the false-positive rates (2.23 vs. 9.97%, p < 0.001). Discussion: Sonographers should be aware that removing NT from combined first-trimester screening would result in a 5-fold increase in false-positive rate to maintain the expected detection rates. This should be an incentive for maintaining quality in NT measurement.

scholarly journals Hybrid Botnet Detection Based on Host and Network Analysis

2020 ◽  
Vol 2020 ◽  
pp. 1-16
Author(s):  
Suzan Almutairi ◽  
Saoucene Mahfoudh ◽  
Sultan Almutairi ◽  
Jalal S. Alowibdi
Keyword(s):  
Cyber Security ◽  
False Positive Rate ◽  
General Technique ◽  
Botnet Detection ◽  
Security Issues ◽  
Positive Rate ◽  
Communication Traffic ◽  
And Performance ◽  
High Level ◽  
And Control

Botnet is one of the most dangerous cyber-security issues. The botnet infects unprotected machines and keeps track of the communication with the command and control server to send and receive malicious commands. The attacker uses botnet to initiate dangerous attacks such as DDoS, fishing, data stealing, and spamming. The size of the botnet is usually very large, and millions of infected hosts may belong to it. In this paper, we addressed the problem of botnet detection based on network’s flows records and activities in the host. Thus, we propose a general technique capable of detecting new botnets in early phase. Our technique is implemented in both sides: host side and network side. The botnet communication traffic we are interested in includes HTTP, P2P, IRC, and DNS using IP fluxing. HANABot algorithm is proposed to preprocess and extract features to distinguish the botnet behavior from the legitimate behavior. We evaluate our solution using a collection of real datasets (malicious and legitimate). Our experiment shows a high level of accuracy and a low false positive rate. Furthermore, a comparison between some existing approaches was given, focusing on specific features and performance. The proposed technique outperforms some of the presented approaches in terms of accurately detecting botnet flow records within Netflow traces.

scholarly journals Real-Time Surgical Problem Detection and Instrument Tracking in Cataract Surgery

2020 ◽  
Vol 9 (12) ◽  
pp. 3896
Author(s):  
Shoji Morita ◽  
Hitoshi Tabuchi ◽  
Hiroki Masumoto ◽  
Hirotaka Tanabe ◽  
Naotake Kamiura
Keyword(s):  
Cataract Surgery ◽  
Real Time ◽  
False Positive Rate ◽  
Surgical Techniques ◽  
Surgical Skill ◽  
Skill Levels ◽  
Detection Rates ◽  
Surgical Methods ◽  
Instrument Tracking ◽  
Positive Rate

Surgical skill levels of young ophthalmologists tend to be instinctively judged by ophthalmologists in practice, and hence a stable evaluation is not always made for a single ophthalmologist. Although it has been said that standardizing skill levels presents difficulty as surgical methods vary greatly, approaches based on machine learning seem to be promising for this objective. In this study, we propose a method for displaying the information necessary to quantify the surgical techniques of cataract surgery in real-time. The proposed method consists of two steps. First, we use InceptionV3, an image classification network, to extract important surgical phases and to detect surgical problems. Next, one of the segmentation networks, scSE-FC-DenseNet, is used to detect the cornea and the tip of the surgical instrument and the incisional site in the continuous curvilinear capsulorrhexis, a particularly important phase in cataract surgery. The first and second steps are evaluated in terms of the area under curve (i.e., AUC) of the figure of the true positive rate versus (1—false positive rate) and the intersection over union (i.e., IoU) obtained by the ground truth and prediction associated with the region of interest. As a result, in the first step, the network was able to detect surgical problems with an AUC of 0.97. In the second step, the detection rate of the cornea was 99.7% when the IoU was 0.8 or more, and the detection rates of the tips of the forceps and the incisional site were 86.9% and 94.9% when the IoU was 0.1 or more, respectively. It was thus expected that the proposed method is one of the basic techniques to achieve the standardization of surgical skill levels.

scholarly journals Observability of dusty debris discs around M-stars

2020 ◽  
Vol 499 (3) ◽  
pp. 3932-3942
Author(s):  
Patricia Luppe ◽  
Alexander V Krivov ◽  
Mark Booth ◽  
Jean-François Lestrade
Keyword(s):  
Detection Rate ◽  
Second Generation ◽  
Far Infrared ◽  
Detection Rates ◽  
Cool Star ◽  
Solar Type ◽  
M Stars ◽  
Cent Detection

ABSTRACT Debris discs are second-generation dusty discs formed by collisions of planetesimals. Many debris discs have been found and resolved around hot and solar-type stars. However, only a handful have been discovered around M-stars, and the reasons for their paucity remain unclear. Here, we check whether the sensitivity and wavelength coverage of present-day telescopes are simply unfavourable for detection of these discs or if they are truly rare. We approach this question by looking at the Herschel/DEBRIS survey that has searched for debris discs including M-type stars. Assuming that these cool-star discs are ‘similar’ to those of the hotter stars in some sense (i.e. in terms of dust location, temperature, fractional luminosity, or mass), we check whether this survey should have found them. With our procedure we can reproduce the $2.1^{+4.5}_{-1.7}$ per cent detection rate of M-star debris discs of the DEBRIS survey, which implies that these discs can indeed be similar to discs around hotter stars and just avoid detection. We then apply this procedure to IRAM NIKA-2 and ALMA bands 3, 6, and 7 to predict possible detection rates and give recommendations for future observations. We do not favour observing with IRAM, since it leads to detection rates lower than for the DEBRIS survey, with 0.6–4.5 per cent for a 15 min observation. ALMA observations, with detection rates 0.9–7.2 per cent, do not offer a significant improvement either, and so we conclude that more sensitive far-infrared and single dish sub-millimetre telescopes are necessary to discover the missing population of M-star debris discs.

scholarly journals SARS-CoV-2 Neutralizing Antibody Levels Post COVID-19 Vaccination Based on ELISA Method—A Small Real-World Sample Exploration

Vaccines ◽  
2021 ◽  
Vol 9 (10) ◽  
pp. 1139
Author(s):  
Xiaoguang Li ◽  
Chao Liang ◽  
Xiumei Xiao
Keyword(s):  
Neutralizing Antibodies ◽  
Detection Rate ◽  
Neutralizing Antibody ◽  
Population Characteristics ◽  
Detection Rates ◽  
Positive Detection Rate ◽  
Significant Difference ◽  
Positive Rate ◽  
Antibody Levels

This study investigated the detection of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) neutralizing antibodies following inoculation with the coronavirus disease (COVID-19) vaccine. From June to July 2021, 127 participants who had completed COVID-19 vaccination (inactivated SARS-CoV-2 vaccine, 64; CoronaVac, 61; CanSino, 2) were recruited and tested using SARS-CoV-2 neutralizing antibody kits. The positive detection rate (inhibition of neutralizing antibodies ≥ 30%) was calculated and stratified according to population characteristics and inoculation time. The positive rate of neutralizing antibody was 47.22% (17/36) in men and 53.85% (49/91) in women, and 54.55% (24/44) in BMI ≥ 24 and 50.60% (42/83) in BMI < 24. Age was stratified as 20–29, 30–39, 40–49, and ≥50; positive detection rates of SARS-CoV-2 neutralizing antibodies were observed in 60.00% (24/40), 50.00% (21/42), 48.39% (15/31), and 42.86% (6/14), respectively, but with no significant difference (x2 = 1.724, p = 0.632). Among 127 vaccinated participants, 66 (51.97%) were positive. The positive detection rate was 63.93% (39/61) with CoronaVac and 42.19% (27/64) with the inactivated SARS-CoV-2 vaccine (significance x2 = 5.927, p = 0.015). Multivariate analysis revealed a significant difference in vaccination times, with average vaccination weeks in the positive and negative groups of 11.57 ± 6.48 and 17.87 ± 9.17, respectively (t= −4.501, p < 0.001). The positive neutralizing antibody rate was 100.00%, 60.00%, 58.33%, 55.56%, 43.14%, 28.57%, and 0.00% at 2–4, 5–8, 9–12, 13–16,17–20, 21–24, and >24 weeks, respectively (x2 = 18.030, p = 0.006). Neutralizing antibodies were detected after COVID-19 inoculation, with differences relating to inoculation timing. This study provides a reference for vaccine evaluation and follow-up immunization strengthening.

Sign in / Sign up

Export Citation Format

Share Document