Coffee-ring-effect-image-analysis

Coffee ring effect images analyzed by PCA. Both raw images and feature data extracted from processed images analyzed. Raw images processed by Matlab. Best parameters selected by tuning hyperparameters of blackwhite threshold and noise removal parameter. Different parameters selected by different data sets. Boxplot and ANOVA executed by R Raw images PCA analysis conducted by python. PCA conducted on each data set, results plotted PCA conducted on whole data set results plotted. Features extracted from processed images analyzed by python. PCA conducted on each data set, results plotted. PCA conducted on each data set, results plotted. Results has shown by choosing the best parameters, the p-value of ANOVA is low which means the differnce is significant. Both raw images PCA analysis and feature PCA analysis are effective distinguish different water samples. Feature extracted method is more effectove than raw images method.