In this paper, a method for predicting progressive tool flank wear using extracted features from turned surface images has been proposed. Acquired turned surface images are analyzed by using texture analyses, viz., gray level co-occurrence matrix (GLCM), Voronoi tessellation (VT), and discrete wavelet transform (DWT) based methods to obtain information about waviness, feed marks, and roughness from machined surface images for describing tool flank wear. Two features from each texture analyses are extracted and fed into support vector machine (SVM) based regression models for predicting progressive tool flank wear. Mean correlation coefficient between the measured and predicted tool flank wear is found as 0.991.
Automatic detection of corrosion inhibition efficacy by processing the microscopic images of the sample surface by using image texture analysis is a very useful technique in the contemporary research field. The reported technique is less time consuming with the capability of pattern recognition via quantification with preferences of suitable software, coding and various numerical methods ultimately represents data (vide acquisition) more precisely and accurately at agiven condition. An optimal anisotropic diffusion-based pre-processing for image enhancement and run length statistics (RLS) based texture analysis for feature extraction are performed on the microscopic images of mild steel surface treated in 1 M HCl solution with different inhibitors viz., (E)-1-(2,4-dinitrophenyl)-2-(pyridin-4-yl)methylene)hydrazine (abbreviated as DHMP and represented as L1), (E)-1-(4-isocyanophenyl)-2-((pyridin-4-yl)methylene)hydrazine) (abbreviated as IMH and represented as L2) and (E)-1-(2,4-dinitrophenyl)-2-((pyridin-4-yl)methylene)phenol (abbreviated as DMP and represented as L3). Extracted normalized or relative features can be quantified the corrosion inhibition efficiency in an excellent way using multi-regression analysis having R-square values in the range of 0.997 to 0.999.