Journal of Engineering Research

As an important component of the flood control system, the safety of dike engineering is directly related to the safety of people's lives and property. This paper focuses on the research of "intelligent detection technology for hidden dangers in dike engineering based on machine learning", and proposes to introduce modern machine learning methods into the field of traditional water conservancy engineering detection to improve the accuracy and efficiency of hidden danger identification. By constructing a data model suitable for the structural characteristics of dikes, and combining deep learning with image recognition technology, the automatic identification and classification of typical hidden dangers such as cracks and seepage are realized. The technical scheme shows high recognition accuracy in the experimental environment and has good application prospects.

Machine learning; Dike engineering; Hidden danger detection; Intelligent recognition; Data model