基于Hough变换和GVF Snake模型的脑肿瘤分割方法
|
摘要
脑MR图像中肿瘤区域的精确分割对后续的治疗与诊断十分关键,提出了一种基于Hough变换定位与遗传算法优化GVF Snake模型的脑肿瘤分割方法。首先,利用Hough变换和阈值处理自动确定肿瘤区域;然后,利用GVF Snake模型对肿瘤区域进行分割;同时,为了防止GVF Snake在分割时易出现局部极小值的问题,进一步利用遗传算法的全局优化特性,对GVF Snake模型分割的结果进行优化。实验结果表明,提出的模型能实现对肿瘤区域的自动定位,同时也避免了GVF Snake模型在分割时易陷入局部最优的问题,使分割的结果更加精确。
The precise segmentation of the tumor area in the brain MRI image is critical important for the subsequent treatments and diagnoses. This paper proposed a new segmentation method,which based on Hough transform location and genetic algorithm optimization of GVF Snake model. Firstly,it used Hough transform and threshold processing to locate the tumor areas automatically. Then it applied the GVF Snake model to segment the tumor areas. At the same time,to address the problem that the algorithm is easily trapped into local minimum,the results of GVF Snake segmentation are further optimized by using the global optimization of genetic algorithm. The experimental results show that the proposed model is then done automatically with no manual intervention,and it can also avoid the local convergence problems of GVF Snake model,then obtains a more precise result.
The precise segmentation of the tumor area in the brain MRI image is critical important for the subsequent treatments and diagnoses. This paper proposed a new segmentation method,which based on Hough transform location and genetic algorithm optimization of GVF Snake model. Firstly,it used Hough transform and threshold processing to locate the tumor areas automatically. Then it applied the GVF Snake model to segment the tumor areas. At the same time,to address the problem that the algorithm is easily trapped into local minimum,the results of GVF Snake segmentation are further optimized by using the global optimization of genetic algorithm. The experimental results show that the proposed model is then done automatically with no manual intervention,and it can also avoid the local convergence problems of GVF Snake model,then obtains a more precise result.
引文
[1]万俊,聂生东,王远军.基于MRI的脑肿瘤分割技术研究进展[J].中国医学物理学杂志,2013,30(4):4266-4271.
[2]刘岳,王小鹏,王金全,等.基于形态学重建和梯度分层修正的分水岭脑肿瘤分割[J].计算机应用研究,2015,32(8):2487-2491.
[3]徐天芝,张贵仓,贾园.基于形态学梯度的分水岭彩色图像分割[J].计算机工程与应用,2016,52(11):200-203,208.
[4]吴方,何尾莲.基于改进粗糙集概率模型的鲁棒医学图像分割算法[J].计算机应用研究,2017,34(8):2546-2550,2556.
[5] Cabria I,Gondra I. MRI segmentation fusion for brain tumor detection[J].InformationFusion,2017,36(7):1-9.
[6] Benson C C,Lajish V L,Rajamani K. Brain tumor extraction from MRI brain images using marker based watershed algorithm[C]//Proc of International Conference on Advances in Computing,Communications and Informatics. Piscataway,NJ:IEEE Press,2015:318-323.
[7] Moallem P,Tahvilian H,Monadjemi S A. Parametric active contour model using Gabor balloon energy for texture segmentation[J]. Signal Image&Video Processing,2015,10(2):1-8.
[8] Zhu Shiping,Gao Ruidong. A novel generalized gradient vector flow snake model using minimal surface and component-normalized method for medical image segmentation[J]. Biomedical Signal Processing&Control,2016,26(4):1-10.
[9] Xu Chenyang,Prince J L. Gradient vector flow:a new external force for snakes[C]//Proc of Conference on Computer Vision and Pattern Recognition. Washington DC:IEEE Computer Society,1997:66.
[10] Zohra B F,Nacéra B,Abdelmalik T A. Adjustment of active contour parameters in brain MRI segmentation using evolution strategies[C]//Proc of International Conference on Electrical Engineering. Piscataway,NJ:IEEE Press,2016:1-7.
[11] Wang Guoqiang,Wang Dongxue. Segmentation of brain MRI image with GVF snake model[C]//Proc of the 1st International Conference on Pervasive Computing Signal Processing and Applications. Piscataway,NJ:IEEE Press,2010:711-714.
[12]李丹仪,李卫锋,廖庆敏.基于对称信息和主动轮廓模型的脑肿瘤分割系统[J].清华大学学报:自然科学版,2013,53(7):995-1000.
[13]Mukhopadhyay P,Chaudhuri B B. A survey of Hough transform[J].Pattern Recognition,2015,48(3):993-1010.
[14]Liu Leiming,Yang Ning,Lan Jinhui,et al. Image segmentation based on gray stretch and threshold algorithm[J]. Optik-International Journal for Light and Electron Optics,2015,126(6):626-629.
[15]毕晓君,肖婧.差分进化算法GVF Snake模型在PET图像分割中的应用[J].中国图象图形学报,2011,16(3):382-388.
[16]高宜录,顾志凯,陈建,等.颅内脑膜瘤大小分型的建议及其意义[J].中国临床神经外科杂志,2004,9(3):170-172.
[17]葛婷,牟宁,李黎.基于softmax回归与图割法的脑肿瘤分割算法[J].电子学报,2017,45(3):644-649.
[2]刘岳,王小鹏,王金全,等.基于形态学重建和梯度分层修正的分水岭脑肿瘤分割[J].计算机应用研究,2015,32(8):2487-2491.
[3]徐天芝,张贵仓,贾园.基于形态学梯度的分水岭彩色图像分割[J].计算机工程与应用,2016,52(11):200-203,208.
[4]吴方,何尾莲.基于改进粗糙集概率模型的鲁棒医学图像分割算法[J].计算机应用研究,2017,34(8):2546-2550,2556.
[5] Cabria I,Gondra I. MRI segmentation fusion for brain tumor detection[J].InformationFusion,2017,36(7):1-9.
[6] Benson C C,Lajish V L,Rajamani K. Brain tumor extraction from MRI brain images using marker based watershed algorithm[C]//Proc of International Conference on Advances in Computing,Communications and Informatics. Piscataway,NJ:IEEE Press,2015:318-323.
[7] Moallem P,Tahvilian H,Monadjemi S A. Parametric active contour model using Gabor balloon energy for texture segmentation[J]. Signal Image&Video Processing,2015,10(2):1-8.
[8] Zhu Shiping,Gao Ruidong. A novel generalized gradient vector flow snake model using minimal surface and component-normalized method for medical image segmentation[J]. Biomedical Signal Processing&Control,2016,26(4):1-10.
[9] Xu Chenyang,Prince J L. Gradient vector flow:a new external force for snakes[C]//Proc of Conference on Computer Vision and Pattern Recognition. Washington DC:IEEE Computer Society,1997:66.
[10] Zohra B F,Nacéra B,Abdelmalik T A. Adjustment of active contour parameters in brain MRI segmentation using evolution strategies[C]//Proc of International Conference on Electrical Engineering. Piscataway,NJ:IEEE Press,2016:1-7.
[11] Wang Guoqiang,Wang Dongxue. Segmentation of brain MRI image with GVF snake model[C]//Proc of the 1st International Conference on Pervasive Computing Signal Processing and Applications. Piscataway,NJ:IEEE Press,2010:711-714.
[12]李丹仪,李卫锋,廖庆敏.基于对称信息和主动轮廓模型的脑肿瘤分割系统[J].清华大学学报:自然科学版,2013,53(7):995-1000.
[13]Mukhopadhyay P,Chaudhuri B B. A survey of Hough transform[J].Pattern Recognition,2015,48(3):993-1010.
[14]Liu Leiming,Yang Ning,Lan Jinhui,et al. Image segmentation based on gray stretch and threshold algorithm[J]. Optik-International Journal for Light and Electron Optics,2015,126(6):626-629.
[15]毕晓君,肖婧.差分进化算法GVF Snake模型在PET图像分割中的应用[J].中国图象图形学报,2011,16(3):382-388.
[16]高宜录,顾志凯,陈建,等.颅内脑膜瘤大小分型的建议及其意义[J].中国临床神经外科杂志,2004,9(3):170-172.
[17]葛婷,牟宁,李黎.基于softmax回归与图割法的脑肿瘤分割算法[J].电子学报,2017,45(3):644-649.