CPC G06T 7/001 (2013.01) [G06F 18/22 (2023.01); G06V 10/30 (2022.01); G06V 10/751 (2022.01); G06V 10/761 (2022.01); G06T 2207/10016 (2013.01); G06T 2207/20076 (2013.01); G06T 2207/30108 (2013.01); G06T 2207/30181 (2013.01)] | 10 Claims |
1. A method for matching image features applied to mine machine vision, comprising:
a processor de-noising an image to be analyzed using CBDNet network, wherein the image to be analyzed is a captured mine video frame image;
a processor performing super-pixel segmentation on the image to be analyzed obtained by de-noising to obtain a plurality of image blocks;
a processor calculating the information entropy of each image block to obtain an image block with information entropy greater than a first preset threshold;
a processor using SURF algorithm to extract feature points of the image block with information entropy greater than a first preset threshold, to obtain a feature point set of the image to be analyzed;
a processor using a Harr wavelet method to describe each feature point in the feature point set to obtain a feature point descriptor set of the image to be analyzed; and
a processor matching the feature points in the feature point set of the image to be analyzed with feature points of a target image based on the feature point descriptor set of the image to be analyzed to confirm whether an aspect of the mine is abnormal, wherein the target image is an image containing information related to potential abnormalities of the mine.
|