Adaptive Segmentation Method of Wear Particle Chains for Online Ferrographic Image Automatic Recognition

Abstract

The wear particle chain self-adaptive segmentation method for online ferrographic image automatic recognition, step 1, the transmitted light image and the reflected light image provided by the ferrographic sensor are respectively preprocessed and converted into a binary image and a grayscale image; Step 2: Use the reflected light image Imgf to perform rough segmentation based on grayscale morphology; Step 3: Perform fine segmentation on the binary image after the above rough segmentation—multi-scale binary morphological segmentation: for each abrasive grain chain, use The variable-scale erosion-expansion algorithm realizes the segmentation of large and small abrasive particles, and obtains a binary segmentation line; step 4, superimposes the binary segmentation line on the original transmitted light image and reflected light image to obtain the segmented abrasive particle image; The invention can not only solve the problem of segmenting the online wear chain image, but also can be applied to the automatic segmentation of the wear chain in the traditional off-line ferrogram image, which is of great significance for realizing the intellectualization and automation of the ferrogram image analysis technology.

Description

technical field [0001] The invention relates to the technical field of state monitoring of mechanical systems, to an automatic analysis technology of ferrographic images, and in particular to an adaptive segmentation method for wear particle chains oriented to automatic identification of online ferrographic images. Background technique [0002] Wear particle analysis is a key technology to obtain the wear state information of the machine by analyzing the lubricating medium and the wear particles carried by the tested machine, which plays a vital role in the diagnosis, prediction and maintenance decision-making of wear faults. [0003] As an important wear particle analysis method, traditional off-line ferrography image analysis has been successfully applied to the wear state monitoring of industrial equipment, forming a standardized wear state evaluation system. However, this technique has obvious defects: 1) The sampling and analysis cycle is long and the efficiency is low;...

AI Technical Summary

Problems solved by technology

[0013] Aiming at the defects of the prior art, the present invention provides an adaptive segmentation method of wear particle chains oriented to automatic recognition of online ferrographic images, and a method for image segmentation of wear particle chains combined with grayscale and morphological information, which realizes the wear resistance in online ferrographic images. Segmentation of grains; the invention can not only solve the problem of segmentation of online wear grain chain images, but also can be applied to the automatic segmentation of wear grain chains in traditional off-line ferrography images, which is of great significance to realize the intelligence and automation of ferrography image analysis technology

Images