Detection of Stainless Steel Corrosion Based on Convolutional Neural Network

The development of the stainless steel pipe industry in Indonesia has shown significant growth, driven by increasing demand across sectors such as construction, oil and gas, food and beverage, and automotive. Despite its advantages in corrosion resistance, stainless steel pipes remain vulnerable to corrosion due to the reduced composition of essential metals during the manufacturing process. Corrosion-related damage can have serious impacts on safety and operational costs, while manual inspection methods are often considered inefficient and inaccurate. This study aims to develop a web-based corrosion detection system using the Convolutional Neural Network (CNN) method. CNN was chosen for its ability to effectively extract image features and its widespread use across various fields such as healthcare, transportation, and manufacturing. By leveraging a CNN model, the system can automatically classify pipe conditions as either ‘corroded’ or ‘not corroded’ through image analysis. The results of this research are expected to make a meaningful contribution to the monitoring of stainless steel pipe corrosion in a faster, more accurate, and efficient manner, offering an alternative to conventional methods.