Intelligent concrete detection method and system based on machine learning

By decomposing and reconstructing the machine learning network, building mapping connections of network modules, and performing parameter learning, the problems of limited generalization ability and detection accuracy of concrete intelligent detection models in the existing technology are solved, and more efficient and accurate concrete detection is achieved.

CN119649194AActive Publication Date: 2025-03-18SICHUAN CLOUD INSPECTION TECH DEV CO LTD
4 Cites 1 Cited by

Patent Information

Application Number
CN202510180267.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-02-19
Publication Date
2025-03-18
Estimated Expiration
2045-02-19

AI Technical Summary

Technical Problem

The existing intelligent concrete detection methods based on machine learning are difficult to make full use of prior knowledge and domain experience, resulting in limited generalization capabilities and detection accuracy of the model.

Method used

By decomposing the prior machine learning network and the foundation concrete intelligent detection network into multiple network module units, and building mapping contacts, using sample data to perform module parameter learning and transfer parameter learning, the target concrete intelligent detection network is generated.

Benefits of technology

It significantly improves the learning efficiency and accuracy of concrete characteristics, improves the intelligence level and detection accuracy of concrete inspection, and enhances the generalization ability of the model.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention provides an intelligent concrete detection method and system based on machine learning, and the method comprises the steps: carrying out the one-by-one parameter learning of each network module unit of a basic intelligent concrete detection network through a first sample concrete sample image data sequence, and generating a temporary intelligent concrete detection network; the learning efficiency and accuracy of concrete characteristics are remarkably improved, then key module units in a prior machine learning network and a temporary concrete intelligent detection network are fused through module interaction processing, and a plurality of fusion network module units are formed. And finally, migration parameter learning is performed on a second interaction network module unit in the fusion network module unit by using the second sample concrete sample image data sequence until the training convergence requirement is met, and a target concrete intelligent detection network for completing module parameter learning is generated, so that the intelligent level of concrete detection is improved, and the detection efficiency is improved. And the generalization capability and the detection precision of the concrete intelligent detection network are obviously enhanced.
Need to check novelty before this filing date? Find Prior Art