Research On Robust Algorithm Of Flatness Detecti On System For Medium Plate Based On Digital Image Processing

Medium plate is a commonly used major steel product in industrial produc tion.The lack of shape detection of defects in high-precision automated testing equipment.Flatness is an important indicator of plate-shaped defects of mediu m steel plate,which directly affects the production efficiency and material utili zation level of steel production units.Such automated plate detection equipment perennial monopoly by foreign companies.The development of reliable domes tic testing equipment can break the monopoly and reduce the cost of China's i ron and steel enterprises.Therefore,research and development of such automati on equipment is of great significance.With the rapid development of machine vision technology,flatness detecting of non-contact medium plate steel has bec ome a hot spot in China.However,at present,there are not many practical pr oducts in China.One of the important reasons is the lack of image processing algorithms with good robustness.Based on the flatness detection system of plate,this thesis focuses on disc ussing and putting forward an algorithm to enhance the robustness of system d etection,and verify and analyze its performance.Try to use a variety of techni ques,including software and hardware to broaden the ways of research,reduci ng the difficulty of conducting research.Explore the possibility of using compu ter software to virtually verify some of the key components of the overall syst em.Finally,a laser spot centering algorithm with good robustness is proposed,and the performance of the algorithm is evaluated by the actually used flatnes s detection.The overall contents of the thesis are as follows:(1)Explain the background and significance of the research,introduce theresearch status at home and abroad.The emphasis of the research is to e xplore the establishment of strong robustness of the image feature extraction al gorithm.(2)The principle of laser triangulation is deduced,and the robustness of d etection when it is set with smaller error is discussed.The beam shaping of th e laser stripe is deduced,and the optical simulation software Trace Pro is intro duced to simulate the laser spot and analyze its effect.(3)Construct the surface scattering model for medium plate and describe i ts surface characteristics.Through experiments and other means to collect data,finally conduct simulation of laser spot in Trace Pro,and explore whether the effect of its engineering significance.(4)Introduce and analyze the main spot center location algorithm,and pro pose its own robust center extraction algorithm for droplet interference.And int ends to compare the algorithms from the point of view of repeatability of liqui d droplet positioning accuracy.(5)Through the construction of experimental platform,the actual comparis on and verification of the proposed algorithm and the traditional algorithm in t he performance difference.And from the use of computer simulation technolog y on the moving state of the surface of the steel plate possible deformation of their own views.