基于环绕数约束模型的轮廓识别算法研究
|
摘要
为了解决当前图像轮廓识别算法中由于区域标记和轮廓标记性质不同,导致难以将多标记融合识别技术应用于图像轮廓识别中的问题,本文提出了一种基于环绕数约束的能量最小化模型,用以精确识别目标轮廓.在这种模型中,区域标记(如颜色和纹理均匀性)和轮廓标记(如局部对比度和连续性)通过一个目标函数进行描述,实现多标记融合识别.首先,将环绕数作为约束,将其引入到能量最小化模型中,得到区域标记与轮廓标记的线性约束;然后,对区域标记、轮廓标记以及曲率标记进行融合,实现对图像中目标轮廓的识别;最后,将能量最小化模型与标记相结合,通过比率能量函数对算法进行实例应用分析,验证算法的有效性.实验结果表明:与传统轮廓识别算法相比,所提算法具有更高的轮廓识别精度.
In order to solve the current image contour recognition algorithm, it is difficult to realize the multi-label fusion recognition because of the difference of the contour cues and the region cues, and the recognition effect is not good. In this paper, an energy minimization model has been proposed on the basis of wrapping number constraint, which can be used to accurately identify the target contour. In this model, the regional markers(such as color and texture uniformity) and contour markers(such as local contrast and continuity) are described by an objective function to implement multi-label fusion recognition. Firstly, the winding number is introduced as a constraint into energy minimization framework, and the linear constraints of the region cues and the contour cues are obtained. Secondly, the contour recognition of the target image is carried out by integrating the region cues with the contour cues and the curvature cues. And lastly, the energy minimization framework is combined with the cues and instantiated by the energy ratio function, verify the validity of the algorithm. The experimental results show that the contour recognition algorithm based on the winding number constraints model proposed in this paper is applied to the fusion of the region cues and the contour cues, and the ideal image processing effect can be obtained.
In order to solve the current image contour recognition algorithm, it is difficult to realize the multi-label fusion recognition because of the difference of the contour cues and the region cues, and the recognition effect is not good. In this paper, an energy minimization model has been proposed on the basis of wrapping number constraint, which can be used to accurately identify the target contour. In this model, the regional markers(such as color and texture uniformity) and contour markers(such as local contrast and continuity) are described by an objective function to implement multi-label fusion recognition. Firstly, the winding number is introduced as a constraint into energy minimization framework, and the linear constraints of the region cues and the contour cues are obtained. Secondly, the contour recognition of the target image is carried out by integrating the region cues with the contour cues and the curvature cues. And lastly, the energy minimization framework is combined with the cues and instantiated by the energy ratio function, verify the validity of the algorithm. The experimental results show that the contour recognition algorithm based on the winding number constraints model proposed in this paper is applied to the fusion of the region cues and the contour cues, and the ideal image processing effect can be obtained.
引文
[1] 曹家梓,宋爱国.基于马尔科夫随机场的纹理图像分割方法研究 [J].仪器仪表学报,2015,36(4):776-786.
[2] CAO X C,TAO Z Q,ZHANG B,et al.Self-adaptively Weighted Co-Saliency Detectionvia Rank Constraint [J].IEEE Transactions on Image Processing,2014,23(9):4175-4186.
[3] 沈宋衍,陈莹.基于在线回归学习的轮廓跟踪算法 [J].计算机工程,2016,42(5):230-234.
[4] CHEN N,XU Z S,XIA M M.Hierarchical Hesitant Fuzzy K-Means Clustering Algorithm [J].Applied Mathematics-a Journal of Chinese Universities,2014,29(1):1-17.
[5] 李明.曲面上一类闭曲线的旋转指标公式 [J].西南大学学报(自然科学版),2013,35(8):97-100.
[6] TAN H C,ZHOU Y,ZHU Y,et al.Improved River Flow and Random Sample Consensus for Curve Lane Detection [J].Advances in Mechanical Engineering,2015,7(7):1-12.
[7] ROMANO Y,PROTTER M,ELAD M.Single Image Interpolation Via Adaptive Nonlocal Sparsity-Based Modeling [J].IEEE Transactions on Image Processing,2014,23(7):3085-3098.
[8] 杜超.基于广度优先遍历的不信任网络扩展推荐算法 [D].长春:吉林大学,2017.
[9] 凡时财,曾祥峰,周雪,等.融合超像素分割与码本模型的目标检测算法 [J].电子科技大学学报,2017,46(4):572-578.
[10] 王淋,何坤金,陈正鸣,等.基于特征点映射关系的接骨板系列化设计方法 [J].计算机辅助设计与图形学学报,2016,28(9):1587-1597.
[11] 闫超,张建州,姜正茂.基于外区抑制与马尔可夫模型的轮廓检测 [J].电子科技大学学报,2016,45(1):135-140.
[12] SHARMA A,GHOSH J K.Saliency Based Segmentation of Satellite Images [J].ISPRS Annals of Photogrammetry,Remote Sensing and Spatial Information Sciences,2015,II-3/W4:207-214.
[13] FU B L,LI Y,ZHANG B,et al.Dynamic Evaluation of Ecosystem Service Value of the Riparian Zone Based on Remote Sensing from 1986 to 2012 [J].Solid Earth Discussions,2015,7(3):2151-2184.
[14] 王田,邹子龙,乔美娜.基于图像特征分析的物体轮廓提取 [J].北京航空航天大学学报,2016,42(8):1762-1768.
[15] LEVINSHTEIN A,SMINCHISESCU C,DICKINSON S.Optimal Image and Video Closure by Superpixel Grouping [J].International Journal of Computer Vision,2012,100(1):99-119.
[16] YE L W,LIU Z,LI J H,et al.Co-Saliency Detection Via Co-Salient Object Discovery and Recovery [J].IEEE Signal Processing Letters,2015,22(11):2073-2077.
[2] CAO X C,TAO Z Q,ZHANG B,et al.Self-adaptively Weighted Co-Saliency Detectionvia Rank Constraint [J].IEEE Transactions on Image Processing,2014,23(9):4175-4186.
[3] 沈宋衍,陈莹.基于在线回归学习的轮廓跟踪算法 [J].计算机工程,2016,42(5):230-234.
[4] CHEN N,XU Z S,XIA M M.Hierarchical Hesitant Fuzzy K-Means Clustering Algorithm [J].Applied Mathematics-a Journal of Chinese Universities,2014,29(1):1-17.
[5] 李明.曲面上一类闭曲线的旋转指标公式 [J].西南大学学报(自然科学版),2013,35(8):97-100.
[6] TAN H C,ZHOU Y,ZHU Y,et al.Improved River Flow and Random Sample Consensus for Curve Lane Detection [J].Advances in Mechanical Engineering,2015,7(7):1-12.
[7] ROMANO Y,PROTTER M,ELAD M.Single Image Interpolation Via Adaptive Nonlocal Sparsity-Based Modeling [J].IEEE Transactions on Image Processing,2014,23(7):3085-3098.
[8] 杜超.基于广度优先遍历的不信任网络扩展推荐算法 [D].长春:吉林大学,2017.
[9] 凡时财,曾祥峰,周雪,等.融合超像素分割与码本模型的目标检测算法 [J].电子科技大学学报,2017,46(4):572-578.
[10] 王淋,何坤金,陈正鸣,等.基于特征点映射关系的接骨板系列化设计方法 [J].计算机辅助设计与图形学学报,2016,28(9):1587-1597.
[11] 闫超,张建州,姜正茂.基于外区抑制与马尔可夫模型的轮廓检测 [J].电子科技大学学报,2016,45(1):135-140.
[12] SHARMA A,GHOSH J K.Saliency Based Segmentation of Satellite Images [J].ISPRS Annals of Photogrammetry,Remote Sensing and Spatial Information Sciences,2015,II-3/W4:207-214.
[13] FU B L,LI Y,ZHANG B,et al.Dynamic Evaluation of Ecosystem Service Value of the Riparian Zone Based on Remote Sensing from 1986 to 2012 [J].Solid Earth Discussions,2015,7(3):2151-2184.
[14] 王田,邹子龙,乔美娜.基于图像特征分析的物体轮廓提取 [J].北京航空航天大学学报,2016,42(8):1762-1768.
[15] LEVINSHTEIN A,SMINCHISESCU C,DICKINSON S.Optimal Image and Video Closure by Superpixel Grouping [J].International Journal of Computer Vision,2012,100(1):99-119.
[16] YE L W,LIU Z,LI J H,et al.Co-Saliency Detection Via Co-Salient Object Discovery and Recovery [J].IEEE Signal Processing Letters,2015,22(11):2073-2077.