Harbin, China
Harbin, China

Time filter

Source Type

Liu F.-Q.,Harbin Engineering University | Liu F.-Q.,Harbin XiRobot Co. | Wang Z.-Y.,Harbin Engineering University | Wang Z.-Y.,Harbin XiRobot Co. | And 2 more authors.
Advances in Intelligent Systems and Computing | Year: 2015

In order to guarantee that a workpiece is securely fixed during a welding process, some areas of the workpiece should be tack welded before the formal welding process starts. Weld seam images captured at these tacked weld points, however, will be different from normal weld seam images. These differences present a difficulty when trying to obtain accurate feature points from an image. Tacked weld point recognition is the key factor of the robustness and stability in the tracking process of welding. It is easy to distinguish tacked weld points on regular fillet weld seam images, rather than irregular fillet and overlapped weld seam images. This paper analyzes the geometrical features of tacked weld points and introduces the methods to recognize these tacked weld points. Experimental results show that the methods proposed by this paper are robust and the weld point recognition is accurate. © Springer International Publishing Switzerland 2015.


Lin X.-T.,Harbin Engineering University | Wang Z.-Y.,Harbin Engineering University | Wang Z.-Y.,Harbin XiRobot Co. | Ji Y.,Harbin Engineering University | Ji Y.,Harbin XiRobot Co.
Advances in Intelligent Systems and Computing | Year: 2015

There are problems needed to be solved when extracting centerline of structured light strip on multi-pass seam. The problems include complex shape, non-uniform intensity distribution, disconnected caused by welding beans or occlusion and environmental noise etc. To solve these problems, a novel centerline extraction algorithm is proposed based on anisotropic heat diffusion. Putting pixels gray values as the temperature, the algorithm could get centerline composed by highest energy pixels, through heat diffusion along same temperature curve normal directions. The higher accuracy and efficiency of our algorithm is proved by experiments. © Springer International Publishing Switzerland 2015.


Luo X.,Harbin Engineering University | Wang Z.-Y.,Harbin Engineering University | Jia C.-B.,Harbin Xirobot Co.
Advances in Intelligent Systems and Computing | Year: 2015

In order to get real-time visual information about the extremely bright weld arc and the neighboring dark areas in the open arc welding process, a low-cost vision system is proposed. The vision system contains three main components: a single off-the-shelf CCD/CMOS digital camera, a welding glass and an IPC (Industrial Personal Computer). A software-based method of capturing multiexposure images is investigated; it changes exposure gain and shutter time by sending exposure parameters through PC program. By using software trigger method, the system captures sets of two exposures, such as “dark”, “light”, repeatedly. To obtain an image with sufficient details of the weld process, a simple and little time-consuming algorithm which adds every two continuous multi-exposure images is discussed. The approach can render videos of 640 by 480 pixel resolution at up to 22 frames per second. © Springer International Publishing Switzerland 2015.

Loading Harbin XiRobot Co. collaborators
Loading Harbin XiRobot Co. collaborators