Validating the defect detection performance advantage of group designs for software reviews: report of a replicated experiment

2002 ◽  
Author(s):  
L. Pek Wee Land ◽  
R. Jeffery ◽  
C. Sauer
Keyword(s):  
Defect Detection ◽  
Detection Performance ◽  
Software Reviews

Defect detection performance of the UCSD non-contact air-coupled ultrasonic guided wave inspection of rails prototype

2016 ◽  
Author(s):  
Stefano Mariani ◽  
Thompson V. Nguyen ◽  
Simone Sternini ◽  
Francesco Lanza di Scalea ◽  
Mahmood Fateh ◽  
...  
Keyword(s):  
Defect Detection ◽  
Detection Performance ◽  
Guided Wave ◽  
Ultrasonic Guided Wave

Predictions of defect detection performance of air-coupled ultrasonic rail inspection system

2017 ◽  
Vol 17 (3) ◽  
pp. 684-705 ◽  
Author(s):  
Stefano Mariani ◽  
Francesco Lanza di Scalea
Keyword(s):  
Field Test ◽  
Defect Detection ◽  
Guided Waves ◽  
Numerical Study ◽  
Detection Performance ◽  
Probability Of Detection ◽  
Inspection System ◽  
False Alarms ◽  
Operating Variables ◽  
Rail Inspection

A rail inspection system based on ultrasonic guided waves and non-contact (air-coupled) ultrasound transduction is under development at the University of California at San Diego. The system targets defects in the rail head that are major causes of train accidents. Because of the high acoustic impedance mismatch between air and steel, the non-contact system poses severe challenges and questions on the defect detection performance. This article presents an extensive numerical study, conducted with a local interaction simulation approach, to model the ultrasound propagation and interaction with defects in the proposed system. This model was used to predict the expected detection performance of the system in the presence of various defects of different sizes and positions, and at varying levels of signal-to-noise ratios. When possible, operating variables for the model were chosen consistently with the field test of an experimental prototype that was conducted in 2014. The defect detection performance was evaluated through the computation of receiver operating characteristic curves in terms of probability of detection versus probability of false alarms. The study indicates that despite the challenges of non-contact probing of the rail, quite satisfactory inspection performance can be expected for a variety of defect types, sizes, and positions. Beyond the specific cases examined in this article, this numerical framework can also be used in the future to examine a larger variety of field test conditions.

scholarly journals An Efficient Network for Surface Defect Detection

2020 ◽  
Vol 10 (17) ◽  
pp. 6085 ◽  
Author(s):  
Zesheng Lin ◽  
Hongxia Ye ◽  
Bin Zhan ◽  
Xiaofeng Huang
Keyword(s):  
Defect Detection ◽  
Surface Defect ◽  
Characteristic Curve ◽  
Detection Performance ◽  
Training Dataset ◽  
Multi Scale ◽  
Surface Defect Detection ◽  
Cascade Structure ◽  
Asymmetric Loss Function ◽  
Roc Score

Convolutional neural networks (CNN) have achieved promising performance in surface defect detection recently. Although many CNN-based methods have been proposed, most of them are limited by the few samples available for training, and the imbalance of positive and negative samples. Hence, their detection performance needs to be further improved. To this end, we propose a multi-scale cascade CNN called MobileNet-v2-dense to detect defects more efficiently. Specifically, the multi-scale cascade structure used in our network can help capture the weak defect semantics that may be lost in the deep network. Then, we propose a novel asymmetric loss function to further improve detection performance. Lastly, a two-stage augmentation method effectively enlarges the training dataset. Experimental results show that, compared to the state-of-the-art, the area under the receiver-operating characteristic curve (AUC-ROC) score of our method increased by 0.16.

scholarly journals Defect detection in guided wave signals using nonlinear autoregressive exogenous method

2021 ◽  
pp. 147592172110186
Author(s):  
Kangwei Wang ◽  
Jie Zhang ◽  
Yi Shen ◽  
Benjamin Karkera ◽  
Anthony J Croxford ◽  
...  
Keyword(s):  
Experimental Data ◽  
Defect Detection ◽  
Detection Performance ◽  
Guided Wave ◽  
Localized Corrosion ◽  
Measured Signal ◽  
Single Pair ◽  
Using Data ◽  
Detection And Localization ◽  
Nonlinear Autoregressive

To perform long-term structural health monitoring, a method based on a nonlinear autoregressive exogenous network is used to learn the features present in signals acquired from a pristine structure. When a subsequent measured signal is input to the trained nonlinear autoregressive exogenous network, the output is a prediction of the equivalent signal from a pristine structure. The residual when the pristine predicted signal is subtracted from the measured signal is used for defect detection and localization. A methodology of how to train, test and assess a nonlinear autoregressive exogenous network for guided wave signals is introduced and applied to experimental data obtained over a period of 8 years from a sparse array of guided wave sensors deployed on a steel storage tank. A separate nonlinear autoregressive exogenous model is trained for each sensor pair in the array using data captured in 2012. The method is first tested using data from a single pair of sensors. Defect signals are synthesized by superposing simulated responses from defects onto later experimental signals obtained from the real structure. The test results for the nonlinear autoregressive exogenous method show better detection performance than those from the optimal baseline selection method, in terms of receiver operating characteristic curves. The detection performance of the nonlinear autoregressive exogenous method is further assessed on signals from the whole sensor array, again with simulated defect responses superposed. It is shown that good detection and localization performance can be achieved by combining the nonlinear autoregressive exogenous residual signals from different sensor pairs. The nonlinear autoregressive exogenous method is tested on experimental data acquired at intervals over the following 7 years as the condition of the tank naturally degrades. Indications from localized corrosion are observed. Finally, an artificial localized anomaly is added to the tank and is visible at the correct location in the image formed using the nonlinear autoregressive exogenous method.

Case Study of Software Reviews

2011 ◽  
pp. 253-267
Author(s):  
Yuk Kuen Wong
Keyword(s):  
Defect Detection ◽  
Research Method ◽  
Case Study Research ◽  
Structured Interview ◽  
Detection Task ◽  
Software Review ◽  
Test Cases ◽  
Knowledge And Skills ◽  
Software Reviews

The aim of this chapter is to identify the key software review inputs that significantly affect review performance in practice. A case study research method is employed. Five companies are volunteers participating in the study. In-depth semi-structured interview approach is used for data collection. From the results, the typical issues for conducting software review include 1) selecting right reviewers to perform a defect detection task, 2) the limitation of time and resources for organizing and conducting software review, and 3) no standard and specific guideline to measure an effective review for different types of software artefacts (i.e., requirement, design, code, and test cases). Thus the result shows that the experience (i.e., knowledge and skills) of reviewers is the most significant input influencing software review performance.

Sign in / Sign up

Export Citation Format

Share Document