视觉仿生轮廓检测中多尺度融合方法研究
|
摘要
从复杂自然场景中检测目标轮廓是计算机视觉的重要任务之一。研究表明,初级视皮层(V1区)中神经元对外界刺激的响应是经典感受野(CRF)和非经典感受野(nCRF)共同作用的结果,机制可有效用于消除背景纹理。结合多尺度分析,研究基于该机制的轮廓检测,提出一种多尺度融合方法。首先通过仿非经典感受野的周边抑制特性,获得不同尺度下的多幅轮廓信息二值图;接着以最小尺度轮廓信息二值图为基准,判断各像素最优方向两侧相应范围内是否存在其它尺度的下的轮廓信息,同时利用高斯函数对相关信息进行加权融合,获得图像的权重图;最后对权重图进行非极大值抑制和二值化,得到最终轮廓检测融合结果。实验结果表明,与原轮廓检测方法相比,上述方法可以有效提升轮廓检测的效果。
In computer vision,contour detection from complex scene is one of the most important tasks. Research suggests that the response of neurons in primary visual cortex(V1 area) is not only related to its Classical Receptive Field(CRF),but also modulated by its non-Classical Receptive Field(nCRF). This interaction is useful for eliminating background textures. Based on this perception mechanism and multi-resolution analysis,a multi-scale contour fusion method is proposed. At first,based on the physiological characteristics of the non-Classical Receptive Field,multi-scale contours with surround inhibition were obtained. Based on the minimum scale contour binary map,the current pixels with the optimal direction on both sides were used to determine whether there are pixels on other contour maps. After that,the Gaussian function was used to weight the relevant information and obtain the weight graph of the image. Finally,the weight graph was subjected to non-maximum suppression and binarization to obtain the final results. Comparied with original method,the proposed model can improve the performance effectively in contour detection.
In computer vision,contour detection from complex scene is one of the most important tasks. Research suggests that the response of neurons in primary visual cortex(V1 area) is not only related to its Classical Receptive Field(CRF),but also modulated by its non-Classical Receptive Field(nCRF). This interaction is useful for eliminating background textures. Based on this perception mechanism and multi-resolution analysis,a multi-scale contour fusion method is proposed. At first,based on the physiological characteristics of the non-Classical Receptive Field,multi-scale contours with surround inhibition were obtained. Based on the minimum scale contour binary map,the current pixels with the optimal direction on both sides were used to determine whether there are pixels on other contour maps. After that,the Gaussian function was used to weight the relevant information and obtain the weight graph of the image. Finally,the weight graph was subjected to non-maximum suppression and binarization to obtain the final results. Comparied with original method,the proposed model can improve the performance effectively in contour detection.
引文
[1] J Canny.A Computational Approach to Edge Detection[J].IEEE Transactions on Pattern Analysis & Machine Intelligence,1986,8(6):679.
[2] D H Hubel,T N Weisel.T N Wiesel.Receptive fields,binocular interaction and functional architecture in cat's visual cortex[J].J.Physiol.(London) 160,106-154.1962,160.
[3] N Petkov,M A Westenberg.Suppression of contour perception by band-limited noise and its relation to nonclassical receptive field inhibition[J].Biological Cybernetics,2003,88(3):236-246.
[4] C Grigorescu,N Petkov,M A Westenberg.Contour detection based on nonclassical receptive field inhibition[M].IEEE Press,2003.
[5] Q Tang,N Sang,T Zhang.Extraction of salient contours from cluttered scenes[J].Pattern Recognition,2007,40(11):3100-3109.
[6] 杜晓凤,李翠华,李晶.基于复合感受野的轮廓检测方法[J].电子与信息学报,2009,31(7):1630-1634.
[7] C Zeng,et al.Contour detection based on a non-classical receptive field model with butterfly-shaped inhibition subregions[J].Neurocomputing,2011,74(10):1527-1534.
[8] G Papari,N Petkov.An improved model for surround suppression by steerable filters and multilevel inhibition with application to contour detection[J].Pattern Recognition,2011,44(9):1999-2007.
[9] 林川,李亚,曹以隽.考虑微动机制与感受野特性的轮廓检测模型[J].计算机工程与应用,2016,52(24):210-216.
[10] K F Yang,C Y Li,Y J Li.Multifeature-based surround inhibition improves contour detection in natural images[J].IEEE Transactions on Image Processing,2014,23(12):5020-5032.
[11] S G Mallat.A Theory for Multiresolution Signal Decomposition:The Wavelet Representation[M].IEEE Computer Society,1989.
[12] R L D Valois,D G Albrecht,L G Thorell.Spatial frequency selectivity of cells in macaque visual cortex[J].Vision Research,1982,22(5):545-559.
[13] T Lindeberg.Edge detection and ridge detection with automatic scale selection[J].International Journal of Computer Vision,1998,30(2):117-156.
[14] X Ren.Multi-scale Improves Boundary Detection in Natural Images[C].European Conference on Computer Vision.Springer-Verlag,2008:533-545.
[15] 寿天德.视觉信息处理的脑机制[M].合肥:中国科学技术大学出版社,2010:152-158.
[16] 刘曙,罗予频,杨士元.基于多尺度的轮廓匹配方法[J].计算机工程,2008,34(1):201-203.
[2] D H Hubel,T N Weisel.T N Wiesel.Receptive fields,binocular interaction and functional architecture in cat's visual cortex[J].J.Physiol.(London) 160,106-154.1962,160.
[3] N Petkov,M A Westenberg.Suppression of contour perception by band-limited noise and its relation to nonclassical receptive field inhibition[J].Biological Cybernetics,2003,88(3):236-246.
[4] C Grigorescu,N Petkov,M A Westenberg.Contour detection based on nonclassical receptive field inhibition[M].IEEE Press,2003.
[5] Q Tang,N Sang,T Zhang.Extraction of salient contours from cluttered scenes[J].Pattern Recognition,2007,40(11):3100-3109.
[6] 杜晓凤,李翠华,李晶.基于复合感受野的轮廓检测方法[J].电子与信息学报,2009,31(7):1630-1634.
[7] C Zeng,et al.Contour detection based on a non-classical receptive field model with butterfly-shaped inhibition subregions[J].Neurocomputing,2011,74(10):1527-1534.
[8] G Papari,N Petkov.An improved model for surround suppression by steerable filters and multilevel inhibition with application to contour detection[J].Pattern Recognition,2011,44(9):1999-2007.
[9] 林川,李亚,曹以隽.考虑微动机制与感受野特性的轮廓检测模型[J].计算机工程与应用,2016,52(24):210-216.
[10] K F Yang,C Y Li,Y J Li.Multifeature-based surround inhibition improves contour detection in natural images[J].IEEE Transactions on Image Processing,2014,23(12):5020-5032.
[11] S G Mallat.A Theory for Multiresolution Signal Decomposition:The Wavelet Representation[M].IEEE Computer Society,1989.
[12] R L D Valois,D G Albrecht,L G Thorell.Spatial frequency selectivity of cells in macaque visual cortex[J].Vision Research,1982,22(5):545-559.
[13] T Lindeberg.Edge detection and ridge detection with automatic scale selection[J].International Journal of Computer Vision,1998,30(2):117-156.
[14] X Ren.Multi-scale Improves Boundary Detection in Natural Images[C].European Conference on Computer Vision.Springer-Verlag,2008:533-545.
[15] 寿天德.视觉信息处理的脑机制[M].合肥:中国科学技术大学出版社,2010:152-158.
[16] 刘曙,罗予频,杨士元.基于多尺度的轮廓匹配方法[J].计算机工程,2008,34(1):201-203.