In contemporary industrial operations, machining precision has significantly improved, making surface roughness measurement a critical aspect of quality assessment. However, current methodologies for surface roughness measurement are time-consuming. Moreover, these methods require measuring directly on the sample and removing it from the processing machine, leading to potential errors and loss of machining standards. This project aims to minimize the time required for inspecting product quality after machining by assessing surface roughness through images of the machined surface. This study explores the application of deep learning models using MATLAB software to diagnose surface roughness in machining. The results indicate positive signals and a high level of feasibility in diagnosing and checking surface quality. To enhance the reliability of the results, it is necessary to increase the amount of data fed into the model.