基于多分辨率和自适应分数阶的Active Demons算法

张桂梅; 郭黎娟; 熊邦书; 储珺

doi:10.7544/issn1000-1239.2018.20170523

基于多分辨率和自适应分数阶的Active Demons算法

¹(南昌航空大学计算机视觉研究所南昌 330063)
²(南昌航空大学信息工程学院南昌 330063) (guimei.zh@163.com)

基金项目: 国家自然科学基金项目(61462065，61462063)；江西省科技支撑计划基金项目(20161BBF60091)

详细信息

中图分类号: TP391.41
计量
- 文章访问数: 911
- HTML全文浏览量: 0
- PDF下载量: 281
出版历程
- 发布日期: 2018-11-30

Active Demons Algorithm Based on Multi-Resolution and Adaptive Fractional Differential

¹(Computer Vision Institute, Nanchang Hangkong University, Nanchang 330063)
²(School of Information Engineering, Nanchang Hangkong University, Nanchang 330063)

摘要

摘要: 分数阶Active Demons(fractional active demons, FAD)算法是图像非刚性配准的有效方法，并且能解决灰度均匀和弱纹理图像配准精度低，优化易陷入局部极小而导致的配准速度缓慢问题，但是该算法中分数阶最佳阶次的寻找需要通过多次实验人工选取，缺乏阶次自适应性.针对该问题，提出了基于多分辨率和自适应分数阶的Active Demons算法，该算法首先根据图像梯度模值和信息熵，构建了自适应分数阶阶次的数学模型，基于该模型自动计算出分数阶的最佳阶次和微分动态模板；然后将多分辨率策略加入到自适应分数阶Active Demons算法中，进一步提高了图像配准效率.理论分析和实验结果均表明：提出的算法可用于灰度均匀、弱边缘和弱纹理图像的配准，能根据图像的局部特征自适应计算最佳分数阶阶次，并避免了算法陷入局部最优，从而提高了图像配准的精度和效率.
- R-L(Riemann-Liouville)分数阶微分 /
- Active Demons (AD)算法 /
- 自适应模型 /
- 多分辨率 /
- 图像配准
Abstract: Active Demons algorithm based on fractional differential has been proved to be effective for non-rigid image registration, and can solve the problem of low accuracy and low efficiency in image registration for intensity uniformity or weak texture region. However, the optimal order of fractional differential operator in the algorithm needs to be selected manually by multiple experiments, lack of order adaptive in image registration. Aiming at the problem, this paper proposes a new Active Demons algorithm based on multi-resolution and adaptive fractional differential. Firstly, an adaptive fractional order mathematical model is constructed based on the gradient magnitude and information entropy of image, therefore the optimal order and differential dynamic template are adjusted adaptively. Additionally, multi-resolution strategy is introduced to adaptive fractional differential Active Demons algorithm, therefore the registration efficiency is improved. Theory analysis and experimental results show that the proposed algorithm is capable of registrating images with intensity uniformity, weak edge and weak texture. And the optimal order of fractional differential operator can be calculated adaptively. Furthermore, the presented method can avoid falling into local optimum, thus the accuracy and efficiency of registration can be improved.
- R-L(Riemann-Liouville) fractional differential /
- Active Demons (AD) algorithm /
- adaptive model /
- multi-resolition /
- image registration

HTML全文

参考文献(0)

施引文献(11)

期刊类型引用(5)

1.	王明，张倩. 我国基于深度学习的图像识别技术在农作物病虫害识别中的研究进展. 中国蔬菜. 2023(03): 22-28 . 百度学术
2.	覃伟荣，劳燕玲. 基于3D关联规则深度学习的异构遥感数据检测. 计算机仿真. 2023(09): 482-486 . 百度学术
3.	吕晓洁. 基于深度学习的分布式光伏发电系统电压稳定性评估. 电子设计工程. 2022(17): 114-118 . 百度学术
4.	宋美佳，贾鹤鸣，林志兴，卢仁盛，刘庆鑫. 自适应学习率梯度下降的优化算法. 三明学院学报. 2021(06): 36-44 . 百度学术
5.	郑俊浩. 基于深度学习的乳腺癌MRI影像预处理. 智能计算机与应用. 2020(01): 231-232+236 . 百度学术