A method, system, device and medium for clustering large rocks based on deep learning

By applying the deep learning-based large stone clustering method in front-line construction projects, the accuracy and efficiency problems of detecting and analyzing large stones in the existing technology are solved, real-time accurate detection and clustering analysis of large stones on site are realized, and surveying costs are reduced.

CN119741520APending Publication Date: 2025-04-01SHENZHEN TIEYUE ELECTRIC CO LTD
0 Cites 0 Cited by

Patent Information

Application Number
CN202411467721.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-10-21
Publication Date
2025-04-01

AI Technical Summary

Technical Problem

It is difficult for the prior art to accurately detect and analyze whether there are large stones on site and their aggregation degree in front-line construction projects, resulting in high survey costs and low efficiency.

Method used

Using a large stone clustering method based on deep learning, the stone target box is positioned through the target detection network, and cluster analysis is performed. The target box with a size smaller than the preset threshold is deleted, and the target box with intersection is fused to output the clustering results.

Benefits of technology

Real-time accurate detection and cluster analysis of large stones on site are achieved, which greatly reduces the labor cost of on-site surveys and improves detection efficiency.

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

Abstract

The embodiments of the present invention relate to the field of intelligent construction technology, and provide a method, system, device and medium for clustering large stones based on deep learning. The present invention detects whether large stone clusters appear by using target detection reasoning and cluster analysis methods, designs a unique clustering algorithm based on the characteristics of large stones, and introduces the size of the target image as a reference for selecting the stone target frame, so that the clustering algorithm of the present invention is more generalized.
Need to check novelty before this filing date? Find Prior Art