Non-invasive optical characterization and estimation of Zn porosity in gas tungsten arc welding of Fe–Al joints using CR model and OES measurements

In this study,we employed a non-invasive approach based on the collisional radiative(CR)model and optical emission spectroscopy(OES)measurements for the characterization of gas tungsten arc welding(GTAW)discharge and quantification of Zn-induced porosity during the GTAW process of Fe–Al joints.The OES measurements were recorded as a function of weld current,welding speed,and input waveform.The OES measurements revealed significant line emissions from Zn-I in 460–640 nm and Ar-I in 680–800 nm wavelength ranges in all experimental settings.The OES coupled CR model approach for Zn-I line emission enabled the simultaneous determination of both essential discharge parameters i.e.electron temperature and electron density.Further,these predictions were used to estimate the Zn-induced porosity using OES-actinometry on Zn-I emission lines using Ar as actinometer gas.The OES-actinometry results were in good agreement with porosity data derived from an independent approach,i.e.x-ray radiography images.The current study shows that OES-based techniques can provide an efficient route for real-time monitoring of weld quality and estimate porosity during the GTAW process of dissimilar metal joints.

Weld extraction from real-time radiograph based on bidirectional accumulator Hough Transform

An effective approach is presented to extract welds from real-time radiographs, Firstly an algorithm based on an adaptive bidirectional threshold was proposed to segment the gradient image into ternary image, and then the bidirectional accumulator Hough Transform was developed to extract weld edges from the ternary image. Different values of the coefficient proposed in the threshold algorithm were tested, and the proposed approach was applied to extract welds from real-time radiographic images of different types of welds with defects. Results show that the proposed method is adaptive and effective to extract welds from real-time radiographs of linear welds.

Multi-slice CT angiography by triple-phase enhancement in preoperative evaluation of hepatocellular carcinoma

Background Triple-phase enhancement of multi-slice computed tomography (MSCT) has markedly improved the diagnostic accuracy of hepatocellular carcinoma (HCC), and MSCT angiography (MSCTA) has been proved useful in detecting vascular anatomy noninvasively. This study aimed to explore the value of MSCTA by triple-phase enhancement in preoperative evaluation of HCC.Methods Fifty-six consecutive cases of primary HCC scheduled for resection were studied with MSCTA by triple-phase enhancement. The raw data images were processed on a workstation for multiplanar reconstruction (MPR) and three-dimensional (3D) reconstruction. The findings after processing of the data were compared with those after surgery or intraoperative sonography. Results The false positive rate of MSCTA by triple-phase enhancement was 10.1% and its false negative rate was 4.3% in detecting HCC. No significant difference was observed in MSCTA and surgery or intraoperative sonography in detecting vascular anatomy anomalies and pathologic variations, whereas significant difference was found in detecting bile duct invasion with MSCT compared to intraoperative sonography.Conclusions MSCTA by triple-phase enhancement not only improves the detection of HCC, but also provides valuable preoperative information about hepatic vascular architecture and parenchyma. MSCTA by triple-phase enhancement is worthy of application as a non-invasive method in preoperative evaluation of HCC.