Fine edge detection model based on RCF

doi:10.11772/j.issn.1001-9081.2019030462

Abstract

Abstract:

Aiming at the roughness and blur of edges generated by edge detection technology based on deep learning, an end-to-end fine edge detection model based on RCF (Richer Convolutional Features for edge detection) was proposed. In this model based on RCF model, attention mechanism was introduced in the backbone network, Squeeze-and-Excitation (SE) module was used to extract image edge features. In order to avoid excessive loss of detail information, two subsampling in the backbone network were removed. In order to increase the receptive field of the model, dilation convolution was used in the backbone. A residual module was used to fuse the edge images in different scales. The model was trained on the Berkeley Segmentation Data Set (BSDS500)and PASCAL VOC Context dataset by a multi-step training approach and was tested on the BSDS500. The experimental results show that the model improves the ODS (Optimal Dataset Scale) and OIS (Optimal Image Scale) to 0.817 and 0.838 respectively, and it not only generates finer edges without affecting real-time performance but also has better robustness.

Key words: edge detection, Richer Convolutional Features for edge detection (RCF), deep learning, dilation convolution, attention mechanism

摘要：

针对目前基于深度学习的边缘检测技术生成的边缘粗糙及模糊等问题，提出一种基于更丰富特征的边缘检测（RCF）模型的端到端的精细边缘检测模型。该模型以RCF模型为基础，在主干网络中引入"注意力"机制，采用SE模块提取图像边缘特征，并且去掉主干网络部分下采样，避免细节信息过度丢失，使用扩张卷积技术增大模型感受野，并利用残差结构将不同尺度的边缘图进行融合。对伯克利分割数据集（BSDS500）进行增强，使用一种多步骤的训练方式在BSDS500和PASCAL VOC Context数据集上进行训练，并用BSDS500进行测试实验。实验结果表明，该模型将全局最佳（ODS）和单图最佳（OIS）指标分别提高到了0.817和0.838，在不影响实时性的前提下可以输出更精细的边缘，同时还具有较好的鲁棒性。

关键词: 边缘检测, 更丰富的卷积特征检测, 深度学习, 扩张卷积, 注意力机制

CLC Number:

TP391.41

JING Nianzhao, YANG Wei. Fine edge detection model based on RCF[J]. Journal of Computer Applications, 2019, 39(9): 2535-2540.

景年昭, 杨维. 基于RCF的精细边缘检测模型[J]. 计算机应用, 2019, 39(9): 2535-2540.

References

[1] REN X F, BO L F. Discriminatively trained sparse code gradients for contour detection[C]//NIPS 2012:Proceedings of the 25th International Conference on Neural Information Processing Systems. North Miami Beach, FL, USA:Curran Associates, 2012, 1:584-592.
[2] 张广燕, 王俊平, 邢润森,等.PSLIP新模型及在边缘检测和图像增强中的应用[J].电子学报,2015,43(2):377-382.(ZHANG G Y, WANG J P, XING R S, et al. A new PSLIP model and its application in edge detection and image enhancement[J]. Acta Electronica Sinica, 2015, 43(2):377-382.)
[3] KOHLI P, LADICKY L, TORR P H S. Robust higher order potentials for enforcing label consistency[J]. International Journal of Computer Vision, 2009, 82(3):302-324.
[4] 石美红,李青,赵雪青,等.一种基于保角相位的图像边缘检测新方法[J].电子与信息学报,2015,37(11):2594-2600.(SHI M H, LI Q, ZHAO X Q, et al. A new approach for image edge detection based on conformal phase[J]. Journal of Electronics and Information Technology, 2015, 37(11):2594-2600.)
[5] PANTOFARU C, SCHMID C, HERBERT M. Object recognition by integrating multiple image segmentations[C]//ECCV 2008:Proceedings of the 10th European Conference on Computer Vision, LNCS 5304. Berlin:Springer, 2008:481-494.
[6] FELDMAN J A, FELDMAN G M, FALK G, et al. The Stanford hand-eye project[C]//IJCAI'69:Proceedings of the 1st International Joint Conference on Artificial Intelligence. San Francisco, CA:Morgan Kaufmann, 1969:521-526.
[7] CANNY J. A computational approach to edge detection[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1986, 8(6):679-698.
[8] KONISHI S, YUILLE A L, COUGHLAN J M, et al. Statistical edge detection:learning and evaluating edge cues[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2003, 25(1):57-74.
[9] MARTIN D R, FOWLKES C C, MALIK J. Learning to detect natural image boundaries using local brightness, color, and texture cues[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2004, 26(5):530-549.
[10] GANIN Y, LEMPITSKY V. N⁴-Fields:neural network nearest neighbor fields for image transforms[C]//Proceedings of the 2014 Asian Conference on Computer Vision, LNCS 9004. Berlin:Springer, 2014:536-551.
[11] SHEN W, WANG X G, WANG Y, et al. DeepContour:a deep convolutional feature learned by positive-sharing loss for contour detection[C]//Proceedings of the 2015 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway, NJ:IEEE, 2015:3982-3991.
[12] BERTASIUS G, SHI J, TORRESANI L. DeepEdge:a multi-scale bifurcated deep network for top-down contour detection[C]//Proceedings of the 2015 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway, NJ:IEEE, 2015:4380-4389.
[13] XIE S, TU Z. Holistically-nested edge detection[J]. International Journal of Computer Vision, 2017, 125(1/2/3):3-18.
[14] SHELHAMER E, LONG J, DARRELL T. Fully convolutional networks for semantic segmentation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(4):640-651.
[15] LEE C-Y, XIE S, GALLAGHER P, et al. Deeply-supervised nets[EB/OL].[2019-01-02]. https://arxiv.org/pdf/1409.5185.pdf.
[16] LIU Y, CHENG M, HU X, et al. Richer convolutional features for edge detection[C]//Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway, NJ:IEEE, 2017:5872-5881.
[17] SIMONYAN K, ZISSERMAN A. Very deep convolutional networks for large-scale image recognition[EB/OL].[2018-08-12]. https://arxiv.org/pdf/1409.1556.pdf.
[18] HU J, SHEN L, ALBANIE S, et al. Squeeze-and-excitation networks[EB/OL].[2018-08-12]. https://arxiv.org/pdf/1709.01507.pdf.
[19] YU F, KOLTUN V. Multi-scale context aggregation by dilated convolutions[EB/OL].[2018-08-12]. https://arxiv.org/pdf/1511.07122.pdf.
[20] HE K M, ZHANG X Y, REN S Q, et al. Deep residual learning for image recognition[C]//Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway, NJ:IEEE, 2016:770-778.
[21] NAIR V, HINTON G E. Rectified linear units improve restricted Boltzmann machines[C]//ICML'10:Proceedings of the 27th International Conference on Machine Learning. Madison, WI:Omnipress, 2010:807-814.
[22] MARTIN D R, FOWLKES C C, TAL D, et al. A database of human segmented natural images and its application to evaluating segmentation algorithms and measuring ecological statistics[C]//ICCV 2001:Proceedings of the 8th IEEE International Conference on Computer Vision. Washington DC:IEEE Computer Society, 2001, 2:416-423.
[23] MOTTAGHI R, CHEN X, LIU X, et al. The role of context for object detection and semantic segmentation in the wild[C]//Proceedings of the 2014 IEEE Conference on Computer Vision and Pattern Recognition. Washington DC:IEEE Computer Society, 2014:891-898.
[24] FARABET C, COUPRIE C, NAJMAN L, et al. Learning hierarchical features for scene labeling[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2013, 35(8):1915-1929.
[25] 刘胜男,宁纪锋. 基于超像素的点互信息边界检测算法[J]. 计算机应用, 2016, 36(8):2296-2300. (LIU S N, NING J F. Super-pixel based pointwise mutual information boundary detection algorithm[J]. Journal of Computer Applications, 2016, 36(8):2296-2300.)
[26] DOLLÁR P, ZITNICK C L. Fast edge detection using structured forests[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2015, 37(8):1558-1570.
[27] ZITNICK C L, DOLLÁR P. Edge boxes:locating object proposals from edges[C]//Proceedings of the 2014 European Conference on Computer Vision, LNCS 8693. Berlin:Springer, 2014:391-405.