浏览全部资源
扫码关注微信
Published： 16 December 2024 ，
移动端阅览
杜仕荣， 范影乐， 蔡哲飞， 房涛. 2024. 结合双视通路与尺度信息融合的轮廓检测方法. 中国图象图形学报， 29(12):3657-3669
Du Shirong， Fan Yingle， Cai Zhefei， Fang Tao. 2024. A contour detection method combing dual visual pathway and scale information fusion. Journal of Image and Graphics， 29(12):3657-3669
目的
2
考虑到图像信息在视觉通路中的表征是多尺度的，为了实现自然场景下多对比度分布图像的轮廓检测任务，提出了一种基于双视通路尺度信息融合的轮廓检测新方法。
方法
2
首先构建对亮度信息敏感和对颜色信息敏感的大细胞（M）、小细胞（P）并行通路，构建不同尺度的感受野模拟神经节细胞对刺激的模糊和精细感知，使用亮度对比度和色差信息指导不同尺度感受野响应的自适应融合，使其能够充分提取亮度轮廓和颜色轮廓。其次结合外膝体（lateral geniculate nucleus，LGN）多尺度方向差异编码与多尺度朝向选择性抑制方法，构建显著轮廓提取模型，实现轮廓区域的增强以及背景纹理的抑制。最后将加工后的亮度轮廓和颜色轮廓前馈至初级视皮层（V1）区，构建双通道响应权重调节模型整合M、P通路所得信息，进一步丰富轮廓。
结果
2
本文使用BSDS500（berkeley segmentation data set）图像库和NYUD（New York University-depth）图像库对提出的算法进行验证，其中在BSDS500图像库的最优平均准确率（average precision，AP）指标为0.74，相对于SCSI（subfield-based center-surround inhibition）、BAR（bilateral asymmetric receptive）和SED（surround-modulated edge detection）等基于生物视觉机制的检测方法有4%～13%的提升，所得结果轮廓图也更为连续、准确。
结论
2
本文利用M、P双通路机制以及亮度信息和颜色信息在前端视觉通路中的编码过程实现轮廓信息的加工与提取，可以有效实现自然图像的轮廓检测，尤其是对于图像中的细微轮廓边缘的检测，也为研究更高级皮层中视觉信息机制提供新的思路。
Objective
2
The extraction and utilization of contour information， as a low-level visual feature of the target subject， contribute to the efficient execution of advanced visual tasks， such as object detection and image segmentation. When processing complex images， contour detection based on biological vision mechanisms can quickly extract object contour information. However， the perception of primary contour information is currently based on a single scale receptive field template or a simple fusion of multiple scale receptive field templates， which ignores the dynamic characteristics of receptive field scales and makes it difficult to accurately extract contours in complex scenes. Considering the serial parallel transmission and integration mechanism of visual information in the magnocellular （M） and parvocellular （P） dual vision pathways， we propose a new contour detection method based on the fusion of dual vision pathway scale information.
Method
2
First， we introduce Lab， a color system that is close to human visual physiological characteristics， to extract color difference and brightness information from an image. Compared with conventional RGB color systems， Lab is more in line with the way the human eye perceives visual information. Considering that the scale of the receptive field of ganglion cells varies with the size of local stimuli to adapt to different visual task requirements across various scenes， a smaller scale of the receptive field corresponds to a more refined perception of detailed information. We then simulate the fuzzy and fine perception of the stimuli by ganglion cells using two different scale receptive fields， and we use color difference and brightness contrast information to guide the adaptive fusion of large- and small-scale receptive field responses and highlight the contour details. Second， considering the differences in the perception of orientation information among receptive fields at different scales of the lateral geniculate body， we introduce the standard deviation of the optimal orientation obtained from perception at multiple scales as the encoding weight for the direction difference， thereby achieving a modulation of texture region suppression weight information. We also combine local contrast information to guide the lateral inhibition intensity of non-classical receptive fields based on the difference between the central and peripheral directions. Through the collaborative integration of these two， we successfully enhance the contour regions and suppress the background textures. Finally， to simulate the complementary fusion mechanism of color and brightness information in the primary visual cortex （V1） region， we propose a weight association model integrating contrast information. Based on the fusion weight coefficients obtained from the local color contrast and brightness contrast， we achieve a complementary fusion of information flows in the M and P paths， thereby enriching the contour details.
Result
2
We compared our model with three biological-vision-based mechanisms （SCSI， SED， and BAR） and one deep-learning-based model （PiDiNet）. On the BSDS500 dataset， we used several quantitative evaluation indicators， including optimal dataset scale （ODS）， optimal image scale （OIS）， average precision （AP） indicators， and precision-recall （PR） curves， and selected five images to compare the detection performance of each method. Experimental results show that our model has a better overall performance than the other models. Compared with SCSI， SED， and BAR， our model obtains 4.45%， 2.94%， and 4.45% higher ODS index， 2.82%， 5.80%， and 8.96% higher OIS index， and 7.25%， 4.23%， and 5.71% higher AP index， respectively. While the PiDiNet model based on deep learning has some shortcomings compared with various indicators， this model does not require a pre-training of data， has biological interpretability， and has a small computational power requirement. We further extracted four images from the NYUD dataset to visually compared the false detection rate， missed detection rate， and overall performance of the models. We also conducted a series of ablation experiments to demonstrate the contribution of each module in the model to its overall performance.
Conclusion
2
In this paper， we use the M and P dual-path mechanism and the encoding process of luminance and color information in the front-end visual path to realize contour information processing and extraction. Our proposed approach can effectively realize a contour detection of natural images， especially for subtle contour edge detection in images， and provide novel insights for studying visual information mechanism in the higher-level cortex.
轮廓检测双视通路多尺度自适应融合方向差异
countour detectiondual visual pathwaymulti-scaleadaptive fusiondirectional difference
Akbarinia A and Parraga C A. 2018. Feedback and surround modulated boundary detection. International Journal of Computer Vision， 126（12）： 1367-1380 ［DOI： 10.1007/s11263-017-1035-5http://dx.doi.org/10.1007/s11263-017-1035-5］
Cui J Y， Jia H C， Wang H P and Xu F. 2022. A fast threshold neural network for ship detection in large-scene SAR images. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing， 15： 6016-6032 ［DOI： 10.1109/JSTARS.2022.3192455http://dx.doi.org/10.1109/JSTARS.2022.3192455］
Fang T， Fan Y L and Wu W. 2020. Salient contour detection on the basis of the mechanism of bilateral asymmetric receptive fields. Signal， Image and Video Processing， 14（7）： 1461-1469 ［DOI： 10.1007/s11760-020-01689-1http://dx.doi.org/10.1007/s11760-020-01689-1］
Grigorescu C， Petkov N and Westenberg M A. 2003. Contour detection based on nonclassical receptive field inhibition. IEEE Transactions on Image Processing， 12（7）： 729-739 ［DOI： 10.1109/TIP.2003.814250http://dx.doi.org/10.1109/TIP.2003.814250］
Guo X R， Nakamura S， Fujii Y， Seno T and Palmisano S. 2021. Effects of luminance contrast， averaged luminance and spatial frequency on vection. Experimental Brain Research， 239（12）： 3507-3525 ［DOI： 10.1007/s00221-021-06214-5http://dx.doi.org/10.1007/s00221-021-06214-5］
Huang S， Zhang H J， Sun H， Yu L and Chen L W. 2017. Frequency estimation of multiple sinusoids with three sub-Nyquist channels. Signal Processing， 139： 96-101 ［DOI： 10.1016/j.sigpro.2017.04.013http://dx.doi.org/10.1016/j.sigpro.2017.04.013］
Jing J F， Liu S J， Wang G， Zhang W C and Sun C M. 2022. Recent advances on image edge detection： a comprehensive review. Neurocomputing， 503： 259-271 ［DOI： 10.1016/j.neucom.2022.06.083http://dx.doi.org/10.1016/j.neucom.2022.06.083］
Jones H E， Grieve K L， Wang W and Sillito A M. 2001. Surround suppression in primate V1. Journal of Neurophysiology， 86（4）： 2011-2028 ［DOI： 10.1152/jn.2001.86.4.2011http://dx.doi.org/10.1152/jn.2001.86.4.2011］
Jones J P and Palmer L A. 1987. An evaluation of the two-dimensional Gabor filter model of simple receptive fields in cat striate cortex. Journal of Neurophysiology， 58（6）： 1233-1258 ［DOI： 10.1152/jn.1987.58.6.1233http://dx.doi.org/10.1152/jn.1987.58.6.1233］
Lee B B and Swanson W H. 2022. Detection and discrimination of achromatic contrast： a ganglion cell perspective. Journal of Vision， 22（8）： #11 ［DOI： 10.1167/jov.22.8.11http://dx.doi.org/10.1167/jov.22.8.11］
Li C Y and Li W. 1994. Extensive integration field beyond the classical receptive field of cat's striate cortical neurons—classification and tuning properties. Vision research， 34（18）： 2337-2355 ［DOI： 10.1016/0042-6989（94）90280-1http://dx.doi.org/10.1016/0042-6989（94）90280-1］
Li F Z， Lin C， Zhang Q and Wang R X. 2020. A biologically inspired contour detection model based on multiple visual channels and multi-hierarchical visual information. IEEE Access， 8： 15410-15422 ［DOI： 10.1109/ACCESS.2020.2966916http://dx.doi.org/10.1109/ACCESS.2020.2966916］
Li W X， Zhou R G and Yu H. 2021. Quantum image edge detection based on multi-directions gray-scale morphology. International Journal of Theoretical Physics， 60（11）： 4162-4176 ［DOI： 10.1007/s10773-021-04966-yhttp://dx.doi.org/10.1007/s10773-021-04966-y］
Lin C， Wen Z Q， Xu G L， Cao Y J and Pan Y C. 2022. A bio-inspired contour detection model using multiple cues inhibition in primary visual cortex. Multimedia Tools and Applications， 81（8）： 11027-11048 ［DOI： 10.1007/s11042-022-12356-7http://dx.doi.org/10.1007/s11042-022-12356-7］
Manookin M B， Patterson S S and Linehan C M. 2018. Neural mechanisms mediating motion sensitivity in parasol ganglion cells of the primate retina. Neuron， 97（6）： 1327-1340 ［DOI： 10.1016/j.neuron.2018.02.006http://dx.doi.org/10.1016/j.neuron.2018.02.006］
Morrow W M， Paranjape R B， Rangayyan R M and Desautels J E L. 1992. Region-based contrast enhancement of mammograms. IEEE transactions on Medical Imaging， 11（3）： 392-406 ［DOI： 10.1109/42.158944http://dx.doi.org/10.1109/42.158944］
Shen W， Wang X G， Wang Y， Bai X and Zhang Z J. 2015. DeepContour： a deep convolutional feature learned by positive-sharing loss for contour detection//Proceedings of 2015 IEEE Conference on Computer Vision and Pattern Recognition （CVPR）. Boston， USA： IEEE： 3982-3991 ［DOI： 10.1109/CVPR.2015.7299024http://dx.doi.org/10.1109/CVPR.2015.7299024］
Solomon S G. 2021. Retinal ganglion cells and the magnocellular， parvocellular， and koniocellular subcortical visual pathways from the eye to the brain. Handbook of Clinical Neurology， 178： 31-50 ［DOI： 10.1016/B978-0-12-821377-3.00018-0http://dx.doi.org/10.1016/B978-0-12-821377-3.00018-0］
Solomon S G， White A J R and Martin P R. 2002. Extraclassical receptive field properties of parvocellular， magnocellular， and koniocellular cells in the primate lateral geniculate nucleus. The Journal of Neuroscience， 22（1）： 338-349 ［DOI： 10.1523/JNEUROSCI.22-01-00338.2002http://dx.doi.org/10.1523/JNEUROSCI.22-01-00338.2002］
Su Z， Zhang J H， Wang L G， Zhang H， Liu Z， Pietikäinen M and Liu L. 2023. Lightweight pixel difference networks for efficient visual representation learning. IEEE Transactions on Pattern Analysis and Machine Intelligence， 45（12）： 14956-14974 ［DOI： 10.1109/TPAMI.2023.3300513http://dx.doi.org/10.1109/TPAMI.2023.3300513］
Vanni S， Hokkanen H， Werner F and Angelucci A. 2020. Anatomy and physiology of macaque visual cortical areas V1， V2， and V5/MT： bases for biologically realistic models. Cerebral Cortex， 30（6）： 3483-3517 ［DOI： 10.1093/cercor/bhz322http://dx.doi.org/10.1093/cercor/bhz322］
Wei H， Lang B and Zuo Q S. 2013. Contour detection model with multi-scale integration based on non-classical receptive field. Neurocomputing， 103： 247-262 ［DOI： 10.1016/j.neucom.2012.09.027http://dx.doi.org/10.1016/j.neucom.2012.09.027］
Wu Y Y and Li Q. 2022. The algorithm of watershed color image segmentation based on morphological gradient. Sensors， 22（21）： #8202 ［DOI： 10.3390/s22218202http://dx.doi.org/10.3390/s22218202］
Yang K F， Gao S B， Guo C F， Li C Y and Li Y J. 2015. Boundary detection using double-opponency and spatial sparseness constraint. IEEE Transactions on Image Processing， 24（8）： 2565-2578 ［DOI： 10.1109/TIP.2015.2425538http://dx.doi.org/10.1109/TIP.2015.2425538］
Yang K F， Li C Y and Li Y J. 2014. Multifeature-based surround inhibition improves contour detection in natural images. IEEE Transactions on Image Processing， 23（12）： 5020-5032 ［DOI： 10.1109/TIP.2014.2361210http://dx.doi.org/10.1109/TIP.2014.2361210］
Young R A and Lesperance R M. 2001. The Gaussian derivative model for spatial-temporal vision： II. Cortical data. Spatial Vision， 14（3/4）： 321-389 ［DOI： 10.1163/156856801753253591http://dx.doi.org/10.1163/156856801753253591］
Yu Y， Yang Y and Liu K Z. 2021. Edge-aware superpixel segmentation with unsupervised convolutional neural networks//Proceedings of 2021 IEEE International Conference on Image Processing. Anchorage， USA： IEEE： 1504-1508 ［DOI： 10.1109/ICIP42928.2021.9506289http://dx.doi.org/10.1109/ICIP42928.2021.9506289］
Zhang Q， Lin C and Li F Z. 2021. Application of binocular disparity and receptive field dynamics： a biologically-inspired model for contour detection. Pattern Recognition， 110： #107657 ［DOI： 10.1016/j.patcog.2020.107657http://dx.doi.org/10.1016/j.patcog.2020.107657］
Zhou W J， Lv Y， Lei J S and Yu L. 2021. Global and local-contrast guides content-aware fusion for RGB-D saliency prediction. IEEE Transactions on Systems， Man， and Cybernetics： Systems， 51（6）： 3641-3649 ［DOI： 10.1109/TSMC.2019.2957386http://dx.doi.org/10.1109/TSMC.2019.2957386］
相关文章
相关作者
相关机构
京公网安备11010802024621