Measurement Of AFM Surface Roughness Based On Digital Image

The surface roughness is refers to the surface of small spacing and tiny peak and valley roughness. The microscopic size and shape characteristic made by the process and other factors. The surface roughness directly influences the conformity, wear resistance, corrosion resistance, working accuracy of the machine, and the reliability of the instrument and its service life. With the progress of processing industry, the quality of surface processing, and higher requirements on the surface quality of the surface roughness measurement has become an indispensable important test.Through analysis of the surface roughness of the current detection method, combining with AFM characteristics, such as slit, cross passageways, hole cavity surface roughness measurement with difficulty and problems, this thesis put forward the digital image processing and the neural network technology combining method, and design AFM surface roughness measurement system. The first design and make AFM experimental samples, acquire the surface's image and the R a by video hole explore and the surface roughness meter. Then establish image processing software based on MATLAB, and extract characteristic value of image. Finally establish neural net to train and recognition eigenvalues, validate that there is a corresponding relationship between image feature and surface roughness, and thus finish qualitative measurement of surface roughness.This thesis adopted AFM machine, video hole explore and computer as hardware devices, MATLAB as software platform to establishing image processing software system. This system has all kinds of image processing function, and joined the eigenvalue extraction and neural network training function, realizing the AFM surface roughness measurement and results output.