文章信息/Info

Title:

Research on Weld Region Extraction and Contrast Enhancement Based on Image Morphology

文章編號(hào):

1674-3962(2024)02-0175-06

作者:

劉歡; 劉驍佳; 趙耀邦; 王寧; 羅志強(qiáng); 危荃

上海航天精密機(jī)械研究所，上海 201600

Author(s):

LIU Huan;  LIU Xiaojia;  ZHAO Yaobang;  WANG Ning;  LUO Zhiqiang;  WEI Quan

Shanghai Aerospace Precision Machinery Research Institute, Shanghai 201600, China

關(guān)鍵詞:

圖像處理; X射線; 焊縫區(qū)域; 梯度; 對(duì)比度

Keywords:

image processing;  X-ray;  weld area;  gradient;  contrast ratio

分類(lèi)號(hào):

TG441.7；TP751.1

DOI:

10.7502/j.issn.1674-3962.202108021

文獻(xiàn)標(biāo)志碼:

A

摘要:

鋁板焊接后可使用X射線進(jìn)行焊縫內(nèi)部缺陷檢測(cè)，由于X射線成像存在噪聲多、對(duì)比度低、焊縫區(qū)域邊緣模糊、焊縫區(qū)域與背景區(qū)域灰度分布較為相似等問(wèn)題，使用傳統(tǒng)邊緣檢測(cè)方法效果欠佳。提出了一種自動(dòng)提取焊縫區(qū)域并提高其對(duì)比度的方法，通過(guò)去噪處理、邊緣檢測(cè)、形態(tài)學(xué)操作等步驟，最終確定焊縫區(qū)域位置，并對(duì)焊縫區(qū)域內(nèi)像素值進(jìn)行線性變換，提高了焊縫區(qū)域的對(duì)比度。實(shí)驗(yàn)結(jié)果表明，該方法針對(duì)大批量的鋁板X(qián)射線圖片，都能夠自動(dòng)提高焊縫區(qū)域的對(duì)比度,具有良好的魯棒性和準(zhǔn)確性。

Abstract:

After aluminum plate is welded, X-ray can be used to detect the internal defects of weld. Due to the problems of noise, low contrast, fuzzy edge of weld area and uneven brightness distribution in X-ray image, the effect of traditional edge detection method is not good. Therefore, this paper proposes a method to automatically extract weld area and improve its contrast. Through the steps of denoising, edge detection and morphology operation, the weld area position is finally determined, and the pixel value in the weld area is linearly transformed to improve the contrast of the weld area. The experimental results show that this method can automatically improve the contrast of weld area for a large number of aluminum plate X-ray images, and has good robustness and accuracy.