Study On Microstructure And Properties Of Laser Cladding Based On Edge Detection Algorithm

Laser cladding technology is one of the most promising manufacturing technology in rapid prototyping field. It uses high energy density of laser beam to fastly melt the alloy and substrate surface which have different component and performance. Then, different composition and properties of alloy layer are formed in between the base material substrate and it’s surface. In recent years, the study of laser cladding technology are mainly concentrated on the basic theory, process parameters of laser cladding, etc. But using image processing and analysis to improve the study of laser cladding technology are very scanty.This paper simply introduces the technology of laser cladding and the experimental materials, methods, and equipment of laser cladding tissue image. Ni base alloy with B4C powder as the cladding material. By different content of B4C powder to study the morphology, microstructure and performance of the cladding layer. The original laser cladding tissue images exist many shortcomings, for example complex background, much noise and low contrast. Therefore, it is necessary to pretreat the original image. Image edge is one of the most basic and important features of the image, which contains most of information in the image. Image edge detection has been a hot topic in the research of image processing field. The experimental results show that neither the classical edge dretection algorithms nor the traditional SUSAN algorithm can accurately extract the edge of the ceramic particles in the laser cladding tissue image.This paper puts forward an improved SUSAN edge detection method. Because of small computational complexity and strong anti-noise ability7, SUSAN algorithm, which can better balance detective precision and computation complexity, is very suitable to detect the laser cladding organization image. Extracting candidate edge points and realizing the adaptive threshold about t before detecting edge to overcome the shortcomings which SUSAN algorithm need artificial threshold. The experimental results show that the improved SUSAN algorithm can more accurately detect the edge information, reduce the amount of calculation, and do not need artificial threshold.This paper proposes a kind of edge connection algorithm based on active growth. If only by extracting the edge, without connecting edge. we could only detect sporadic pixels or partial edges of the real edge. It’s easy to happen edge break in laser cladding tissue edge image after corroding alloy coating. Therefore, edge joint algorithm is proposed in this paper to connecte the edge of the ceramic particles, which is conducive to calculate grain area and summarizes the particle distribution at late image analysis work.