Surface finish control in machining processes using haralick descriptors and neuronal networks
CompIMAGE'10 Proceedings of the Second international conference on Computational Modeling of Objects Represented in Images
3D surface roughness measurement for scaliness scoring of psoriasis lesions
Computers in Biology and Medicine
| Hi-index | 0.00 |
This paper presents a feasibility assessment of acquiring vision-based surface roughness parameters for specimens produced by different machining type processes. Twenty produced specimens using five of the most common machining operations were employed in the investigation. Stylus-based measurements of these specimens are performed and results are compared to vision-based measurements using amplitude, spacing and hybrid roughness parameters. The applied filtering technique was based on 0.8mm cut-off value and was maintained identical for both stylus and vision data to enable comparable results. Two models of surface irradiance were employed in the interpretation of the vision data. A third model was also established by utilising an additional filtering operation on the vision data. This allows assessment of the impact of this case on acquired values of surface parameters. Ambient lighting was used to minimize the effects of specular type of reflection. Results showed that although there is a considerable change in the parameter values acquired from vision data compared to stylus data, it is possible to obtain roughness parameters using vision-based method to a certain limit of accuracy. However, some roughness evaluation parameters proved to be more reliable than others. The overall results of this paper would encourage further developments in this area to achieve robust 3D vision based roughness measurement systems for industrial use.