Tool wear is intimately related to intelligent operation and maintenance of automated production, workpiece surface quality and dimension accuracy, and tool life. Therefore, it is necessary to improve the production efficiency and quality by predicting tool wear values. The purpose of this research is to use machine learning and image processing methods to estimate tool wear when turning AA7075 aluminum alloy. In addition, the in-depth analysis of cutting tools at different parameters were examined with SEM and EDS analysis. The results shown that there was a 44.40% increase in tool wear when the cutting speed was raised by 100% while feed rates were maintained at the same level. On the other hand, there was a 22.78% increase in tool wear seen when the cutting speed was maintained at the same level. In addition, the difference between the actual values and the image processing model is 3.5%, and the difference between the Multilayer Perception (MLP) model and the linear regression (LR) model is 5% and 7%, respectively.