The corrosion of metal structures and equipment in use not only lead tosignificant economic losses，but also may cause environmental pollution and potential safety hazards. Therefore，it is particularly important to adopt rapid corrosion detection and analysistechniques. With the continuous progress of technology，artificial intelligence（AI）algorithmsshowed great potential in the fields of corrosion image detection and analysis. In this paper，the development of AI，especially computer vision technology and deep learning was reviewed，and how they could change the pattern of traditional detection methods was discussed. Through themethods of automation，data analysis and feature extraction，many challenges in corrosion image detection and analysis could be solved. In short，the integration of AI with deep learningcould facilitate automated feature extraction and classification，substantially enhance the detection precision and efficiency. Additionally，some key challenges to solve in this field werealso summarized. The purpose was to explore the potential applications of AI in corrosionresearch，provide a new perspective for the future research of detection and evaluationtechnologies，and promote the wider adoption of AI.