Core box corner point extraction and perspective distortion correction method based on multi-scale retinex fusion

By combining multi-scale Retinex fusion, topological constraint Shi-Tomasi and graph cut optimization, thin plate spline transformation and Markov random field, the problems of inaccurate corner point extraction, insufficient image preprocessing, unsuitable perspective correction and low batch processing efficiency in core box image correction in the field are solved. It realizes efficient and accurate core box image correction and batch processing, which meets the digital needs of field core exploration.

CN122023206BActive Publication Date: 2026-06-23ZIJIN MINING GRP SOUTHWEST GEOLOGICAL EXPLORATION CO LTD
View PDF 1 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
ZIJIN MINING GRP SOUTHWEST GEOLOGICAL EXPLORATION CO LTD
Filing Date
2026-04-15
Publication Date
2026-06-23

AI Technical Summary

Technical Problem

Existing core box image correction technology suffers from problems such as inaccurate corner point extraction, insufficient image preprocessing, unsuitable perspective correction, incomplete verification of correction results, low batch processing efficiency, and non-lightweight algorithms in field exploration, making it difficult to meet the high precision and high efficiency requirements of field core exploration.

Method used

Image preprocessing is performed using an iterative algorithm with multi-scale Retinex fusion. Corner points are extracted by combining topological constraints Shi-Tomasi and graph cut optimization. Perspective correction is performed by fusing thin plate spline transformation and Markov random field. Full-dimensional accuracy is verified by Hough transform line fitting and geometric moment invariants, enabling adaptive parameter transfer for batch images and lightweight deployment of the algorithm.

Benefits of technology

It achieves accurate extraction of core box corner points and efficient correction of image distortion, ensuring the full-dimensional accuracy and stability of the correction results, improving batch processing efficiency, and enabling the algorithm to be deployed and applied on field edge equipment, thus meeting the digital processing needs of field core exploration.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122023206B_ABST
    Figure CN122023206B_ABST
Patent Text Reader

Abstract

The present application provides a field core tank corner point extraction and perspective distortion correction method based on multi-scale Retinex fusion, comprising: S1. Core tank image preprocessing based on anisotropic diffusion and multi-scale Retinex fusion; S2. Core tank corner point extraction based on topological constraint Shi-Tomasi and graph cut optimization interaction; S3. Perspective correction based on thin plate spline transformation and Markov random field fusion; S4. Correction result precision verification and closed loop feedback based on Hough transform straight line fitting and geometric moment invariant fusion. The present application realizes accurate extraction of core tank corner points and efficient distortion correction of images, and simultaneously completes full-dimensional precision verification of the correction results, adaptive migration of processing parameters of batch images and lightweight field deployment of the algorithm.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Rock core photograph automatic correction and cutting method based on YOLOv8-seg and perspective transformation

    CN121707887A