Metal surface rusty spot detection method fused with multi-direction multi-element universe local segmentation

The invention relates to a metal surface rust spot detection method fused with multidirectional multivariate universe local segmentation, which comprises the following steps: firstly, acquiring a metal surface rust spot image to be detected, and carrying out median filtering and homomorphic filterin...

Full description

Saved in:
Bibliographic Details
Main Authors ZHU KUNHUA, YAN XIN, LIAO YIPENG
Format Patent
LanguageChinese
English
Published 24.11.2023
Subjects
CALCULATING
COMPUTING
COUNTING
ELECTRIC DIGITAL DATA PROCESSING
IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
PHYSICS
Online AccessGet full text

Cover

More Information
Summary:The invention relates to a metal surface rust spot detection method fused with multidirectional multivariate universe local segmentation, which comprises the following steps: firstly, acquiring a metal surface rust spot image to be detected, and carrying out median filtering and homomorphic filtering preprocessing on the image; secondly, carrying out global multi-direction gray scale scanning analysis on the image, determining a main gray scale distribution curve, and carrying out peak valley point searching and positioning; then, a multi-universe optimization algorithm is adopted to optimize fluctuation amplitude threshold value and floating threshold value parameters of peak-valley interval segmentation, and optimal local segmentation in all directions is obtained; and finally, carrying out compression fusion on the multi-directional optimal local segmentation image by adopting a principal component analysis method to obtain a final rust spot detection result. The method is beneficial for improving the dete
Bibliography:Application Number: CN202310386231