A method for evaluating the effect of generating images with the same viewing angle before and after damage

By registering and spatial conversion of the three-dimensional model, pre-destruction images are generated and evaluating the three-dimensional structural differences, the problem of inconsistency in image angle and position in the damage assessment method in the prior art is solved, and the evaluation efficiency and accuracy are improved.

CN119338813BActive Publication Date: 2025-03-18BEIJING INST OF TECH
1 Cites 0 Cited by

Patent Information

Application Number
CN202411876750.5
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-12-19
Publication Date
2025-03-18
Estimated Expiration
2044-12-19

AI Technical Summary

Technical Problem

Among the existing damage assessment methods, the image comparison method cannot ensure that the angle and position of the image taken by the drone before and after the damage is consistent, making it difficult to acquire paired images and high test costs. The traditional two-dimensional image registration method is poorly robust to occlusion, noise, and light changes, which affects the evaluation accuracy.

Method used

By registering the three-dimensional model, the pre-damage model and post-damage model are spatially transformed, and pre-damage images corresponding to the post-damage are generated, the test cost is reduced, and the three-dimensional structural differences are used as an indicator for destruction assessment.

Benefits of technology

It breaks the requirements for paired images of the same perspective, improves the evaluation efficiency, reduces the difficulty and cost of data acquisition, enhances the adaptability of the method, and improves the accuracy of damage assessment.

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

Abstract

The present invention discloses an effect evaluation method for generating images of the same viewing angle before and after damage, and the method belongs to the technical field of damage effect evaluation. The present invention constructs pre-damage and post-damage data sets based on videos shot by a drone, including images, image shooting postures, and sparse point clouds; then, the pre-damage model and the post-damage model are spatially transformed according to the sparse point cloud registration; next, the posture, point cloud, and image generated before the damage obtained after the spatial transformation are used as input, and trained through a differentiable rendering model to obtain a pre-damage model; the trained model can generate images before the damage according to the shooting angle after the damage, and obtain an evaluation of the degree of damage by comparing the differences between the two images and point clouds before and after the damage.
Need to check novelty before this filing date? Find Prior Art