A Design And Implementation Of Bridge Disease Identification System Based On Machine Learning

With the development of computer science and technology,image recognition technology is widely used in all aspects of production and life,and image recognition can also play its corresponding role in the routine inspection and maintenance of bridges.Improve the practicability and superiority of the application of image recognition technology through the study of machine learning,establish a digital model of bridge disease picture information,realize automatic recognition of bridge disease images,strengthen the intelligent management of bridges,and improve bridge inspection and maintenance The topics of work efficiency and improving the accuracy of bridge disease location in bridge inspections are of great significance.Starting from the image recognition technology based on machine learning,the paper analyzes the bridge disease inspection and identification process in detail during the bridge inspection process.Based on the bridge inspection content,from the perspective of intelligent identification of bridge diseases,a paper is developed.A bridge disease identification system that provides bridge disease identification and record for bridge inspectors.Based on the structure of the bridge and the location of the occurrence of the bridge disease,the rules of the types of bridge disease found during the bridge inspection are summarized.Based on this,the types of the bridge disease location are divided into seven detailed categories,and eight bridges corresponding to the location are defined.According to the disease category,a mathematical model of bridge disease picture information is established based on a large amount of collected bridge disease picture information and matching algorithms.The newly collected bridge disease picture can be identified to obtain the corresponding type of bridge disease.At the same time,according to the actual situation during the bridge inspection process,the bridge disease picture identification module was deployed on the server side,and the WEB-side management module and the mobile APP side inspection module were developed.The bridge was identified by calling the bridge disease picture identification interface to implement the bridge Intelligent identification of bridge diseases during inspections.The system uses Python and C # as the main development languages,and uses a machine learning-based model.Through training on more than 20,000 bridge disease pictures,a classification model of bridge diseases is obtained.Developed the WEB platform and bridge disease identification APP,deployed the bridge disease identificationmodule on the server,provided bridge disease picture recognition for the WEB platform and APP,and deployed both through the WEB platform and APP to build a bridge inspection disease identification system The practical application provides intelligent bridge disease identification and recording for bridge inspection personnel.