融合改进FT显著性与Grabcut的图像目标分割算法
|
摘要
针对目前复杂度较大的图像中目标分割速度较慢、显著性边界分割不明确等问题,提出了一种融合改进的FT(Frequency-tuned)显著性检测与Grabcut的图像分割算法。该算法首先通过改进基于频率调谐的FT显著性检测方法得到图像中显著性较高的区域,并利用SLIC(Simple Linear Iter-ative Clustering)算法对显著图进行预处理得到超像素图,能够有效改善边界的分割效果,然后通过以图论GraphCut算法为基础改进的Grabcut算法建立高斯混合模型。为了提高算法效率,通过聚类以超像素代替原像素,并反复迭代高斯混合模型(Gaussian Mixed Model,GMM)参数,最后利用最大流最小割算法得到最优目标分割结果。实验结果表明所提算法能够更准确更高效率地分割图像中的显著性目标,对高分辨率图像也有很好的适用效果,相比于其他算法在分割精度上提高10%左右,并具有较高的分割效率。
In order to solve the problems of slower target segmentation and ambiguous boundary separation in images with large complexity,an improved image segmentation algorithm based on frequency-tuned(FT)saliency and Grabcut is proposed.The algorithm firstly obtains a region with higher significance in the image by improving the FT saliency detection method based on frequency tuning,then uses the simple linear iterative clustering(SLIC) algorithm to preprocess the significant images to get a superpixel map,which can effectively improve the segmentation effect of the boundary,and then a Gaussian mixture model is built by using the improved Grabcut algorithm based on Graph algorithm.In order to improve the efficiency of the algorithm,the original pixels are replaced by superpixels by clustering,and the Gaussian mixed model(GMM) parameters are iterated repeatedly.Finally,the optimal target segmentation result is obtained by using the maximum flow minimum cut algorithm.The experimental results show that the algorithm can more accurately and efficiently segment significant targets in images and has a good effect on high resolution images.The proposed segmentation method achieves better segmentation efficiency and its segmentation accuracy has increased by about 10% on average in comparison with that of other existing algorithms.
In order to solve the problems of slower target segmentation and ambiguous boundary separation in images with large complexity,an improved image segmentation algorithm based on frequency-tuned(FT)saliency and Grabcut is proposed.The algorithm firstly obtains a region with higher significance in the image by improving the FT saliency detection method based on frequency tuning,then uses the simple linear iterative clustering(SLIC) algorithm to preprocess the significant images to get a superpixel map,which can effectively improve the segmentation effect of the boundary,and then a Gaussian mixture model is built by using the improved Grabcut algorithm based on Graph algorithm.In order to improve the efficiency of the algorithm,the original pixels are replaced by superpixels by clustering,and the Gaussian mixed model(GMM) parameters are iterated repeatedly.Finally,the optimal target segmentation result is obtained by using the maximum flow minimum cut algorithm.The experimental results show that the algorithm can more accurately and efficiently segment significant targets in images and has a good effect on high resolution images.The proposed segmentation method achieves better segmentation efficiency and its segmentation accuracy has increased by about 10% on average in comparison with that of other existing algorithms.
引文
[1] GIDE M S,KARAM L J.A locally weighted fixation densi-ty-based metric for assessing the quality of visual saliencypredictions[J]. IEEE Transaction on Image Processing,2016,25(8):3852-3861.
[2] YANG C,ZHANG L H,LU H C,et al.Saliency detectionvia graph-based manifold ranking[C]//Proceedings of2013 IEEE Conference on Computer Vision and PatternRecognition.Portland:IEEE,2013:3166-3173.
[3] BORJI A,SIHITE D N,ITTI L.What/where to look next?modeling top-down visual attention in complex interactiveenvironments[J]. IEEE Transactions on Systems,Man,and Cybernetics Systems,2014,44(5):523-538.
[4] ITTI L,KOCH C,NIEBUR E.A model of saliency-basedvisual attention for rapid scene analysis[J]. IEEE Trans-actions on Pattern Analysis and Machine Intelligence,1998,20(11):1254-1259.
[5] HAREL J,KOCH C,PERONA P.Graph-based visual sa-liency[C]//Proceedings of the 2006 Conference on Ad-vances in Neural Information Processing System. Cam-bridge:MIT Press,2007:545-552.
[6] HOU X,ZHANG L.Saliency detection:a spectral residualapproach[C]//Proceedings of the 2007 IEEE Conferenceon Computer Vision and Pattern Recognition. Minneapo-lis:IEEE,2007:1-8.
[7] YANG C,ZHANG L H,LU H C,et al.Saliency detectionvia graph-based manifold ranking[C]//Proceedings ofthe 2013 IEEE Conference on Computer Vision and Pat-tern Recognition.Portland:IEEE,2013:3166-3173.
[8] LIU N,HAN J. DHSNet:deep hierarchical saliency net-work for salient region detection[J].IEEE Transactions onPattern Analysis and Machine Intelligence,2015,37(3):569-582.
[9] HE Y,SUN Y.An automatic image segmentation algorithmbased on Grabcut[C]//Proceedings of 6th InternationalConference on Wireless,Mobile and Multi-Media(IC-WMMN 2015).Beijing:IET,2015:185-190.
[10] LI X,LU H,ZHANG L,et al.Saliency detection via denseand sparse reconstruction[C]//Proceedings of the 26 thInternational Conference on Computer Vision and PatternRecognition.Los Alamitos:IEEE,2013:976-2983.
[11]周芹,马志强,单勇,等.自适应约束下的双边全变差正则化超分辨率重建[J].电讯技术,2015,55(6):599-604.
[12]刘辉,石小龙.结合显著性和超像素改进的Grabcut图像分割[J].红外技术,2018,40(1):55-61.
[13] ACHANATA R,HEMAMI S,ESTRADA F,et al.Frequencytuned salient region detection[C]//Proceedings of 2009IEEE Conference on Computer Vision and Pattern Recogni-tion(CVPR).Miami:IEEE,2009:1597-1604.
[14] ROTHER C,KOLMOGOROV V,BLAKE A.Grabcut:inter-active foreground extraction using iterated graph cuts[J].ACM Transactions on Graphics,2004,23(3):309-314.
[15] BOYKOV Y,JOLLY M P.Interactive graphcuts for optimalboundary and region segmentation of objects in N-D images[C]//Proceedings of 2001 International Conference onComputer Vision.Vancouver:IEEE,2 001:105-112.
[16] ACHANATA R,SHAJI A,SMITH K,et al.SLIC super-pixels compared to state-of-the-art superpixel methods[J].IEEE Transactions on Pattern Analysis and MachineIntelligence,2012,34(11):2274-2282.
[17] ACHANATA R,SUSST R,UNK S.Saliency detection u-sing maximum symmetric surround[C]//Proceedings of2010 IEEE International Conference on Image Process-ing.Hong Kong:IEEE,2010:2653-2656.
[18] TANG J,DENG C,HUANG G,et al.Compressed-do-main ship detection on spaceborne optical image usingdeep neural network and extreme learning machine[J].IEEE Transactions on Geoscience and Remote Sensing,2015,53(3):1174-1185.
[19] YANG C,ZHANG L H,LU H C,et al.Saliency detectionvia graph-based manifold ranking[C]//Proceeding of2013 IEEE Conference on Computer Vision and PatternRecognition.Portland:IEEE,2013:3166-3173.
[20]陈科尹,邹湘军,熊俊涛,等.基于视觉显著性改进的水果图像模糊聚类分割算法[J].农业工程学报,2013,29(6):157-165.
[21] BAI P,CUI Z,GAN Z,et al.A novel saliency detectionmodel based on curvelet transform[C]//Proceedings ofInternational Conference on Wireless Communications&Signal Processing.Yangzhou:IEEE,2016:1-5.
[2] YANG C,ZHANG L H,LU H C,et al.Saliency detectionvia graph-based manifold ranking[C]//Proceedings of2013 IEEE Conference on Computer Vision and PatternRecognition.Portland:IEEE,2013:3166-3173.
[3] BORJI A,SIHITE D N,ITTI L.What/where to look next?modeling top-down visual attention in complex interactiveenvironments[J]. IEEE Transactions on Systems,Man,and Cybernetics Systems,2014,44(5):523-538.
[4] ITTI L,KOCH C,NIEBUR E.A model of saliency-basedvisual attention for rapid scene analysis[J]. IEEE Trans-actions on Pattern Analysis and Machine Intelligence,1998,20(11):1254-1259.
[5] HAREL J,KOCH C,PERONA P.Graph-based visual sa-liency[C]//Proceedings of the 2006 Conference on Ad-vances in Neural Information Processing System. Cam-bridge:MIT Press,2007:545-552.
[6] HOU X,ZHANG L.Saliency detection:a spectral residualapproach[C]//Proceedings of the 2007 IEEE Conferenceon Computer Vision and Pattern Recognition. Minneapo-lis:IEEE,2007:1-8.
[7] YANG C,ZHANG L H,LU H C,et al.Saliency detectionvia graph-based manifold ranking[C]//Proceedings ofthe 2013 IEEE Conference on Computer Vision and Pat-tern Recognition.Portland:IEEE,2013:3166-3173.
[8] LIU N,HAN J. DHSNet:deep hierarchical saliency net-work for salient region detection[J].IEEE Transactions onPattern Analysis and Machine Intelligence,2015,37(3):569-582.
[9] HE Y,SUN Y.An automatic image segmentation algorithmbased on Grabcut[C]//Proceedings of 6th InternationalConference on Wireless,Mobile and Multi-Media(IC-WMMN 2015).Beijing:IET,2015:185-190.
[10] LI X,LU H,ZHANG L,et al.Saliency detection via denseand sparse reconstruction[C]//Proceedings of the 26 thInternational Conference on Computer Vision and PatternRecognition.Los Alamitos:IEEE,2013:976-2983.
[11]周芹,马志强,单勇,等.自适应约束下的双边全变差正则化超分辨率重建[J].电讯技术,2015,55(6):599-604.
[12]刘辉,石小龙.结合显著性和超像素改进的Grabcut图像分割[J].红外技术,2018,40(1):55-61.
[13] ACHANATA R,HEMAMI S,ESTRADA F,et al.Frequencytuned salient region detection[C]//Proceedings of 2009IEEE Conference on Computer Vision and Pattern Recogni-tion(CVPR).Miami:IEEE,2009:1597-1604.
[14] ROTHER C,KOLMOGOROV V,BLAKE A.Grabcut:inter-active foreground extraction using iterated graph cuts[J].ACM Transactions on Graphics,2004,23(3):309-314.
[15] BOYKOV Y,JOLLY M P.Interactive graphcuts for optimalboundary and region segmentation of objects in N-D images[C]//Proceedings of 2001 International Conference onComputer Vision.Vancouver:IEEE,2 001:105-112.
[16] ACHANATA R,SHAJI A,SMITH K,et al.SLIC super-pixels compared to state-of-the-art superpixel methods[J].IEEE Transactions on Pattern Analysis and MachineIntelligence,2012,34(11):2274-2282.
[17] ACHANATA R,SUSST R,UNK S.Saliency detection u-sing maximum symmetric surround[C]//Proceedings of2010 IEEE International Conference on Image Process-ing.Hong Kong:IEEE,2010:2653-2656.
[18] TANG J,DENG C,HUANG G,et al.Compressed-do-main ship detection on spaceborne optical image usingdeep neural network and extreme learning machine[J].IEEE Transactions on Geoscience and Remote Sensing,2015,53(3):1174-1185.
[19] YANG C,ZHANG L H,LU H C,et al.Saliency detectionvia graph-based manifold ranking[C]//Proceeding of2013 IEEE Conference on Computer Vision and PatternRecognition.Portland:IEEE,2013:3166-3173.
[20]陈科尹,邹湘军,熊俊涛,等.基于视觉显著性改进的水果图像模糊聚类分割算法[J].农业工程学报,2013,29(6):157-165.
[21] BAI P,CUI Z,GAN Z,et al.A novel saliency detectionmodel based on curvelet transform[C]//Proceedings ofInternational Conference on Wireless Communications&Signal Processing.Yangzhou:IEEE,2016:1-5.