The analysis of various processes in a continuous casting plant can aid in reducing costs and defects in the production of steel slabs. As the quality of the slabs can only be determined at the end of the solidification process, this thesis focuses on analyzing the surface movements on the mold using a variety of methods. The primary objective of this study is to develop a graphical user interface and implement deep learning methods for automated inspection in a continuous casting steel plant. The developed user interface is designed to visualize recorded image data of the mold and perform statistical analysis using techniques such as histogram and optical flow. The results of the analysis are displayed directly in the software, and tests have demonstrated its effectiveness in identifying asynchronous movements between the right and left sides of the mold. Moreover, the study utilizes a deep neural network method on a publicly available labeled steel dataset with defects. The applied model, Mask R-CNN, can analyze steel defects and provide insight into the quality of the steel end-products. This research demonstrates the potential for combining graphical user interface and deep learning techniques to enhance the inspection process in continuous casting steel plants.