Ultrasonic phased array ndt image segmentation method based on watershed and cv model

Abstract

The invention relates to an ultrasonic phased array NDT image segmentation method based on a watershed and a CV model. The method includes: performing watershed transformation on an image to be segmented based on a watershed algorithm, obtaining a watershed image, counting pixel information of the watershed image, and obtaining a weight matrix Perform filtering and denoising processing on the image to be segmented to obtain a pre-processed image, and set the initial parameters of the CV model based on the pre-processed image; Introduce the weight matrix into the CV model to establish a weight CV model; Based on the weight CV The model performs level set function iteration to obtain the final segmentation result. Compared with the prior art, the present invention has the advantages of high segmentation efficiency and high segmentation accuracy of the ultrasonic phased array NDT image.

Description

technical field [0001] The invention belongs to the field of non-destructive testing and image processing, relates to an automatic image segmentation method, in particular to an ultrasonic phased array NDT image segmentation method based on watershed and CV models. Background technique [0002] With the development of non-destructive testing (NDT) technology, ultrasonic phased array non-destructive testing has become one of the most widely used non-destructive testing methods due to its advantages of strong penetration, no pollution, and no harm to human body. The application of ultrasonic phased array NDT technology puts forward higher requirements for the processing technology of detection images. In the three-dimensional reconstruction of ultrasonic phased array NDT image sequences, there are high requirements for the segmentation accuracy, operation efficiency and edge integrity of the target. [0003] The segmentation method based on the level set algorithm is suitable...

AI Technical Summary

Problems solved by technology

However, the segmentation method based on the level set algorithm is often inefficient, and the evolution of the contour curve is carried out with a fixed iterative step size, which further leads to the reduction of its segmentation accuracy.

Due to the limitation of ultrasonic imaging mechanism and detection environment, ultrasonic phased array NDT images often contain a lot of noise, and the gray scale is uneven, which brings great difficulties to the segmentation of ultrasonic phased array NDT images

When using watershed transform to segment ultrasonic phased array NDT images, accurate edge information cannot be obtained due to the existence of noise

There are a large number of images in the image sequence. Although manual segmentation can guarantee the accuracy, the efficiency is too low. In order to meet the three-dimensional reconstruction of ultrasonic phased array NDT images, it is necessary to propose an automatic segmentation method with high precision and fast efficiency.

Images