CAJ | 학술논문

提出了一种新的交互式图像分割方法，该方法通过空间曲面拟合方法生成背景能量项，并将其融入Graph Cuts的分割模型中，以实现对背景灰度不均匀的PCB CT图像分割。本文的方法通过概率模型将自顶向下的全局信息和传统图割模型中自底向上的局部信息有效结合起来，其中背景能量项依据用户添加的背景种子点自适应地生成。通过对多组 PCB CT 图像进行分割实验，结果表明与经典的Graph Cuts 和Grabcut等算法相比，本文方法能得到更准确的分割结果。
제출료일충신적교호식도상분할방법，해방법통과공간곡면의합방법생성배경능량항，병장기융입Graph Cuts적분할모형중，이실현대배경회도불균균적PCB CT도상분할。본문적방법통과개솔모형장자정향하적전국신식화전통도할모형중자저향상적국부신식유효결합기래，기중배경능량항의거용호첨가적배경충자점자괄응지생성。통과대다조 PCB CT 도상진행분할실험，결과표명여경전적Graph Cuts 화Grabcut등산법상비，본문방법능득도경준학적분할결과。
This paper proposes a new interactive segmentation method for non -uniform PCB CT image combining background potential which generated from surface fitting and graph cuts framework .We employ a probabilistic model which includes background potentials to incorporate top -down information that is global across the image ,in addition to the grid clique potentials which provide the bottom -up information used in previous approaches .Furthermore ,the background potential is generated following user input .Through a comparison of different PCB image segmentation experiments ,results demonstrate the proposed method have better performance compared with the state of art method such as Graph Cuts and Grabcut in terms of segmentation accuracy .