Three-Dimensional Detection Of End-Milling Surface Roughness Based On Single Image

Surface roughness is an important indicator of surface quality and has a direct influence on the service performances,accuracy,corrosion resistance,lives and safety of machine parts,especially for parts with special functions.Therefore,detecting the surface roughness of machine parts fastly,accurately and non-destructively is very important for the application of the parts and the safety of the system.Based on the theory of machine vision,the non-contact and nondestructive detection for end-milling was achieved by shape from shading(SFS).The main contents of this thesis are as follows:(1)The samples of End-milling surface roughness samples were collected.Under the same illumination and magnification,the image were collected by the hardware platform.(2)Original images of the surfaces of machine parts were gathered,then grayed,denoised.Denoise processing is achieved by median filtering.(3)Improved algorithm of 3D reconstruction,a new improved algorithm for SFS was proposed,which was improved based on Sobel gradient operator.The difference achieved by Sobel differential gradient operator was used to replace the Lambertian illumination model.(4)Three dimensional reconstruction of end milling surface was achieved based on improved 3D reconstruction algorithm.The feasibility of the proposed algorithm is verified by comparing the 2D contour curve of the vertical texture direction in the reconstructed surface with the measured contour curve.(5)A surface roughness measurement system is established.Based on the image information of 3D reconstruction,the 2D and 3D feature parameters of the surface roughness were extracted,and then the relationships between surface roughness parameters and different roughness levels were analyzed to enrich the surface roughness characterization.The MATLAB software was used to develop a simple GUI user interface of surface roughness detection system which could quickly,accurately and non-destructively detect the surface roughness.In summary,through the analysis of a non-contact precision measuring method of End Milling surface roughness,the surface microstructure of 3D reconstruction was obtained.The 3D roughness information and workpiece surface roughness 3D characteristic parameters were picked up,and the corresponding relationship between different grade roughness and the 3D feature parameters were defermied.The surface roughness detection system was established based on MATLAB software,and the quantitative detection of surface roughness and the reliability of the test results were finished.