The prediction of machined surface parameters is an important factor in machining centre development. There is a great need to elaborate a method for on-line surface roughness estimation [1-7]. Among various measurement techniques, optical methods are considered suitable for in-process measurement of machined surface roughness. These techniques are non-contact, fast, flexible and tree-dimensional in nature.

The optical method suggested in this paper is based on the vision system created to acquire an image of the machined surface during the cutting process. The acquired image is analyzed to correlate its parameters with surface parameters. In the application of machined surface image analysis, the wavelet methods were introduced. A digital image of a machined surface was described using the one-dimensional Digital Wavelet Transform with the basic wavelet as Coiflet. The statistical description of wavelet components made it possible to develop the quality measure and correlate it with surface roughness [8-11].

For an estimation of surface roughness a neural network estimator was applied [12-16]. The estimator was built to work in a recurrent way. The current value of the Ra estimation and the measured change in surface image features were used for forecasting the surface roughness Ra parameter. The results of the analysis confirmed the usability of the application of the proposed method in systems for surface roughness monitoring.

Combining surface measurement data from individual measurements of surface fragments is an issue that has been recognized for flat surfaces. The connection takes place on the principle of making ‘overlap’ measurements according to a specific measurement strategy, and then the algorithm synthesizes the measurement data for the common part (data fusion). This paper presents a method of combining partial data into one larger set using image processing methods. The purpose of the analysis is to combine surface data of a more complex shape in terms of surface roughness and waviness. A successful attempt was made to combine surface measurement data located on a cylindrical surface – convex surface. A rotated table was designed and used for surface data acquisition. The datasets were acquired with the use of CCI 6000 (366 μm – 366 μm) with the assumed overlapping of at least 20%. The measurement datasets were first pre-processed: filling in non-measured points, levelling and form re- moving were applied. For such processed datasets, the common part was identified (data registration) and then the data fusion was performed. An example of stitching the surface datasets shows usefulness of the presented methodology.

This article concerns a new method of assessing the thread cutting process and the quality of the formed thread using the method of optical observation of the workpiece during machining. A series of digital images of the thread profile was taken in transmitted light for each tool infeed. Such images, obtained with high resolution for three angular positions, were binarized, with the space between ridges taken in successive infeeds identified as “void”, and its projection was then parameterized. Two of these parameters, area of void and aspect ratio, were used as indicators of the technological quality of the thread. The suitability of the selected parameters for technological description of the thread was verified using the example of titanium alloy thread turning under ambient and cryogenic conditions.

Industrial applications require functional surfaces with a strictly defined micro-texture. Therefore engineered surfaces need to undergo a wide range of finishing processes. One of them is the belt grinding process, which changes the surface topography on a range of roughness and micro-roughness scales. The article describes the use of machined surface images in the monitoring process of micro-smoothing. Machined surface images were applied in the estimation of machined surface quality. The images were decomposed using two-dimensional Discrete Wavelet Transform. The approximation component was analyzed and described by the features representing the geometric parameters of image objects. Determined values of image features were used to create the model of the process and estimation of appropriate time of micro-smoothing.