Accuracy Of Defects Research Articles

Surface defect detection method for discarded mechanical parts under heavy rust coverage

With a significant number of mechanical products approaching the retirement phase, the batch recycling of discarded mechanical parts necessitates a preliminary assessment of their surface condition. However, the presence of surface rust poses a challenge to defect identification. Therefore, this paper proposes a method for detecting heavily rusted surface defects based on an improved YOLOv8n network. In the Backbone, the C2f-DBB module of re-parameterized deep feature extraction was introduced, and the attention module was designed to improve the accuracy of information extraction. In the Neck part, a Bi-Afpn multiscale feature fusion strategy is designed to facilitate information exchange between features at different scales. Finally, Focal-CIoU is employed as the bounding box loss function to enhance the network’s localization performance and accuracy for defects. Experimentally, it is proved that the improved network in this paper improves the Recall, Precision, and mAP0.5 by 1.2%, 2.1%, and 1.9%, respectively, on the original basis, which is better than other network models.

Characterization of Thickness Loss in a Storage Tank Plate with Piezoelectric Wafer Active Sensors

In terms of the structural health inspection of storage tanks by ultrasonic guided wave technology, many scholars are currently focusing on the tanks’ floor and walls, while little research has been conducted on storage tank roofs. However, the roof of a storage tank is prone to corrosion because of its complex structure and unique working environment. For this purpose, this paper proposes a reflection/transmission signal amplitude ratio (RTAR) coefficient method for corrosion depth assessment. We studied the relationship between the RTAR coefficient, the corrosion depth, and the guided wave frequency to establish a depth assessment model. More importantly, unlike the traditional reflection coefficient method, the characteristics of guided wave signals, including the propagation and attenuation, are introduced in this model for accurate assessment. To eliminate the interference of residual vibration and improve the detection accuracy of defects, we built a corrosion detection system by using piezoelectric sensors and carried out field tests to verify the performance of the proposed method. We demonstrate that corrosion defects with a minimum depth of 0.2 mm can be quantitatively evaluated.

Estimate the Precision of Defects Based on Reports Duplication in Crowdsourced Testing

When analyzing the defects of crowdsourced testing, the testing reports need to be preprocessed, including removing duplicate and false positives. At present, most crowdsourced testing research focuses on duplication of the reports, which has achieved high accuracy. However, studies on reducing false positives of defects have rarely been conducted. Starting from the duplication of defects in the reports, this paper discusses the relationship between the duplication and the accuracy of defects and proposes an estimation approach based on the defect distribution in historical crowdsourced testing projects. Experiments have shown that our approach can provide a priori knowledge of defects and exhibits good stability. We applied this approach to the defect population estimation of crowdsourced testing. With better experimental results, our improved model is more accurate than the original model. We attempt to apply it to the estimation of the defect population of crowdsourced testing, which is more accurate than the original model.

Detection of defects in building walls using modified OptD method for down-sampling of point clouds

ABSTRACT Terrestrial laser scanning is a simple and nondestructive method for the high-accuracy, three-dimensional mapping of buildings and structures. It yields a high-resolution point cloud, allowing for comprehensive and reliable diagnosis of the target building. However, there are difficulties in processing such large datasets. Commercial software typically reduces the datasets using random methods, resulting in the loss of useful information. Herein, we propose a modified optimum dataset (OptD) method for performing diagnostic measurements on buildings. The modified OptD method allows the retention of more points corresponding to areas of interest, such as those with cracks, cavities, and other surface imperfections, and removal of redundant information related to flat and homogeneous surface walls. We propose two approaches for reducing the size of the datasets while simultaneously detecting the imperfections in building walls. The first is to down-sample the datasets in the OXYZ coordinate system to improve the detection of defects corresponding to geometric changes (e.g. cracks and cavities). The second is to down-sample the datasets in the OXYI coordinate system (where I is the laser intensity) to improve the detection accuracy for defects corresponding to changes in the physicochemical properties of the surface (e.g. moisture content, weathering, salt blooming, and biodeterioration).

Importance of Voxel Size in Defect Localization Using Gamma-Ray Scattering

The voxel, defined as the volume of the intersection between incident (primary) and scattered beams, plays an important role in the localization of defects in samples having several interests. In this work, the gamma rays emitted from a 137Cs radioactive source (having the strength of 222 GBq) are scattered from various regions of a wood sample. The scattered gamma flux is detected by an NaI(Tl) scintillation detector placed at 110 deg to the primary gamma-ray beam. Defect (decay) in the wood is simulated by drilling two collinear cylindrical flaws (having diameters of 0.8 and 1.2 cm) in the wood sample and then filling it with a mixture of sawdust and glue. Three sets of collimators with diameters of 6, 7, and 8 mm for the source and detector are used to vary the voxel size (volume). It has been found that better contrast (29.43% for a 1.2-cm defect and 16.37% for an 0.8-cm defect) is achieved for the smallest voxel (16.13 cm3) in comparison to the other two voxels (25.65 and 38.36 cm3). Further, better contrast for the smallest voxel is confirmed by comparing gray images obtained using MATLAB for all three voxel sizes at different scan positions. It has been concluded that for a given experimental setup, the accuracy of defect (decay) detection demands reduced voxel size.

Novel Method to Improve the Imaging Accuracy of Defect Based on Zero-Signal Filling for the Plate

Many effective image-based imaging algorithms are available for detecting and locating the defects for plate-like structures. The discrete ellipse imaging algorithm is widely applied in plate-like structures. However, the scattering-wave extracted by the baseline subtraction may easily be impacted by noise and the reflect wave from boundary, which significantly reduces the signal-to-noise (STN) of the scattering wave, and then leads to the degrading of imaging accuracy. To enhance the STN and improve imaging accuracy, a novel method of zero-signal filling is used in the discrete ellipse imaging in this paper. To verify its effectiveness, the imaging experiment with a circular sparse array has been conducted. The experimental results show that the method can effectively resolve the signal delay and enhance the STN of the scattering wave. The imaging accuracy can be improved obviously by this method and the defect can be located accurately.

Thermography data fusion and nonnegative matrix factorization for the evaluation of cultural heritage objects and buildings

The application of the thermal and infrared technology in different areas of research is considerably increasing. These applications involve nondestructive testing, medical analysis (computer aid diagnosis/detection—CAD), and arts and archeology, among many others. In the arts and archeology field, infrared technology provides significant contributions in terms of finding defects of possible impaired regions. This has been done through a wide range of different thermographic experiments and infrared methods. The proposed approach here focuses on application of some known factor analysis methods such as standard nonnegative matrix factorization (NMF) optimized by gradient-descent-based multiplicative rules (SNMF1) and standard NMF optimized by nonnegative least squares active-set algorithm (SNMF2) and eigen-decomposition approaches such as principal component analysis (PCA) in thermography, and candid covariance-free incremental principal component analysis in thermography to obtain the thermal features. On the one hand, these methods are usually applied as preprocessing before clustering for the purpose of segmentation of possible defects. On the other hand, a wavelet-based data fusion combines the data of each method with PCA to increase the accuracy of the algorithm. The quantitative assessment of these approaches indicates considerable segmentation along with the reasonable computational complexity. It shows the promising performance and demonstrated a confirmation for the outlined properties. In particular, a polychromatic wooden statue, a fresco, a painting on canvas, and a building were analyzed using the above-mentioned methods, and the accuracy of defect (or targeted) region segmentation up to 71.98%, 57.10%, 49.27%, and 68.53% was obtained, respectively.

Adaptive two-SVM multi-objective cuckoo search algorithm for software defect prediction

Two-support vector machine is a new prediction model for software defect. For this model, one multi-objective oriented cuckoo search is designed to optimise several objects simultaneously to improve the defect accuracy, and the ratio of dataset plays an important role to determine the number of big/small modules. In this paper, we provide one extension for the multi-objective oriented cuckoo search, so that it can also adaptive optimise this ratio. Simulation results show our modification achieves the best performance when compared with two other software defect prediction models.

Research on the Nondestructive Detection of Ultrasonic Guided Wave Based on the Time Reversal Method

Because there are many mode transformations when the ultrasonic guided waves run into defects in the pipeline, the reflected signals got by means of the traditional ultrasonic guided wave technique are complex and the amplitudes are small. The time reversal method is a way to intercept the reflected signals in the initial results with a certain bandwidth and excite the time reversal guided waves on the corresponding nodes. In this way, energies of the guided wave are focused in time and space. By comparing the accuracy of defects identification in the straight pipes and the bent pipes with the traditional ultrasonic guided wave method and the time reversal method, this paper proves that the time reversal method has many advantages over the tradition ultrasonic guided wave method. The time reversal method overcomes disadvantages of the traditional guided wave technique and improves the identification degree and accuracy of defects effectively. It lays the foundation for the final defect identification.

A case study on detection and sizing of defects in steam generator tubes using eddy current testing

Eddy current testing (ECT) method is widely used to detect various types of defects occurring in nuclear steam generator tubes. Therefore, the reliability of its detection and sizing accuracy for defects should be validated. For this purpose, two tubes with defect signals were pulled from an operating steam generator and destructively examined. The defect type was a circumferential crack for one tube and an intergranular attack (IGA) for the other tube. The plus point coil probe showed a better capability to detect and size both a circumferential crack and a volumetric IGA than pancake and bobbin coil probe. The destructive results are correlated with the ECT results obtained during the in-service inspection.