基于信息熵的自适应尺度活动轮廓图像分割模型

发布时间：2018-06-06 01:29

本文选题：图像分割 + 活动轮廓模型　；参考：《西北工业大学学报》2017年02期

【摘要】：针对固定尺度活动轮廓模型无法快速准确分割灰度不均匀图像的问题,提出了一种基于信息熵的自适应尺度活动轮廓图像分割模型。首先,利用最大后验概率(MAP)以及贝叶斯分类准则,提出了一种新型能量泛函,提高了模型对灰度信息的提取能力,进而极大提高了模型对灰度不均匀图像的分割准确度。其次,利用图像信息熵构造了自适应尺度算子,使模型能根据图像灰度不均程度自动调整尺度,提高了模型对灰度不均匀图像的分割速度。最后,为验证文中模型的优越性,将该模型与LGDF模型进行了对比,并通过分割时间、迭代次数以及相似度等指标,对分割结果进行了客观、定量分析。最终结果表明,该模型不但对初始轮廓具有较高鲁棒性,而且对灰度不均匀图像具有较高的分割准确性与分割效率。
[Abstract]:An adaptive active contour image segmentation model based on information entropy is proposed to solve the problem that the fixed scale active contour model can not segment gray-scale non-uniform image quickly and accurately. Firstly, using the maximum posterior probability MAP( map) and Bayesian classification criterion, a new energy functional is proposed, which improves the ability of extracting gray level information and greatly improves the segmentation accuracy of the model for grayscale uneven images. Secondly, the adaptive scale operator is constructed by using image information entropy, so that the model can automatically adjust the scale according to the uneven degree of gray level of the image, and improve the segmentation speed of the model to the non-uniform image of gray scale. Finally, in order to verify the superiority of the proposed model, the model is compared with the LGDF model, and the segmentation results are analyzed objectively and quantitatively by using the segmentation time, iteration times and similarity. The results show that the proposed model is not only robust to the initial contour, but also has high segmentation accuracy and efficiency for non-uniform gray-scale images.
【作者单位】：西北工业大学自动化学院;西北工业大学机电学院;西安交通大学人工智能与机器人研究所;
【基金】：中央高校基本科研业务费专项资金(3102016ZY013)资助
【分类号】：TP391.41

【相似文献】