In-situ layerwise imaging measurement of laser powder bed fusion(LPBF)provides a wealth of forming and defect data which enables monitoring of components quality and powder bed homogeneity.Using high-resolution camera...In-situ layerwise imaging measurement of laser powder bed fusion(LPBF)provides a wealth of forming and defect data which enables monitoring of components quality and powder bed homogeneity.Using high-resolution camera layerwise imaging and image processing algorithms to monitor fusion area and powder bed geometric defects has been studied by many researchers,which successfully monitored the contours of components and evaluated their accuracy.However,research for the methods of in-situ 3D contour measurement or component edge warping identification is rare.In this study,a 3D contour mea-surement method combining gray intensity and phase difference is proposed,and its accuracy is verified by designed experiments.The results show that the high-precision of the 3D contours can be achieved by the constructed energy minimization function.This method can detect the deviations of common ge-ometric features as well as warpage at LPBF component edges,and provides fundamental data for in-situ quality monitoring tools.展开更多
For a vision measurement system consisted of laser-CCD scanning sensors, an algorithm is proposed to extract and recognize the target object contour. Firstly, the two-dimensional（2D） point cloud that is output by th...For a vision measurement system consisted of laser-CCD scanning sensors, an algorithm is proposed to extract and recognize the target object contour. Firstly, the two-dimensional（2D） point cloud that is output by the integrated laser sensor is transformed into a binary image. Secondly, the potential target object contours are segmented and extracted based on the connected domain labeling and adaptive corner detection. Then, the target object contour is recognized by improved Hu invariant moments and BP neural network classifier. Finally, we extract the point data of the target object contour through the reverse transformation from a binary image to a 2D point cloud. The experimental results show that the average recognition rate is 98.5% and the average recognition time is 0.18 s per frame. This algorithm realizes the real-time tracking of the target object in the complex background and the condition of multi-moving objects.展开更多
基金This work was supported by the foundation of Key Research and Development Program of Hubei Province(2020BAB137)Shen-zhen Fundamental Research Program(JCYJ20210324142007022).
文摘In-situ layerwise imaging measurement of laser powder bed fusion(LPBF)provides a wealth of forming and defect data which enables monitoring of components quality and powder bed homogeneity.Using high-resolution camera layerwise imaging and image processing algorithms to monitor fusion area and powder bed geometric defects has been studied by many researchers,which successfully monitored the contours of components and evaluated their accuracy.However,research for the methods of in-situ 3D contour measurement or component edge warping identification is rare.In this study,a 3D contour mea-surement method combining gray intensity and phase difference is proposed,and its accuracy is verified by designed experiments.The results show that the high-precision of the 3D contours can be achieved by the constructed energy minimization function.This method can detect the deviations of common ge-ometric features as well as warpage at LPBF component edges,and provides fundamental data for in-situ quality monitoring tools.
文摘For a vision measurement system consisted of laser-CCD scanning sensors, an algorithm is proposed to extract and recognize the target object contour. Firstly, the two-dimensional（2D） point cloud that is output by the integrated laser sensor is transformed into a binary image. Secondly, the potential target object contours are segmented and extracted based on the connected domain labeling and adaptive corner detection. Then, the target object contour is recognized by improved Hu invariant moments and BP neural network classifier. Finally, we extract the point data of the target object contour through the reverse transformation from a binary image to a 2D point cloud. The experimental results show that the average recognition rate is 98.5% and the average recognition time is 0.18 s per frame. This algorithm realizes the real-time tracking of the target object in the complex background and the condition of multi-moving objects.
