位置强化与双路径边缘增强解码的盲道分割方法

doi:10.3778/j.issn.1673-9418.2409073

摘要/Abstract

摘要： 针对盲道分割任务中背景复杂、边界模糊以及盲道形状多样化导致的分割效果不佳等问题，提出了一种位置强化与双路径边缘增强解码的盲道分割方法。设计了位置强化特征提取模块，通过在降采样过程中融入MobileVitv3构成的主干网络，增强模型对盲道特征的感知能力，充分保留上下文信息；提出了融合通道位置强化注意力模块，分别在通道和空间维度上强化特征提取，提升模型在低对比度场景下有效区分盲道与背景信息的能力；采用双路径边缘增强解码方式，对盲道的区域与边界信息进行解码，并结合联合损失函数进一步优化边界细节的处理。此外，针对当前缺乏大规模公开盲道数据集的问题，自制了一个多场景盲道数据集（MSBD），为模型训练和实验验证提供了更丰富的数据支持。实验结果表明，该网络在MSBD数据集上的mIoU、Precision、Recall以及F1-score分别达到96.82%、96.84%、96.48%、96.66%，均优于SegFormer、Deeplabv3+等网络；在输入图片大小为512×512×3时，参数量和计算量分别为1.73×106和1.93 GFLOPs，且推理帧率可达86 FPS，综合性能优于所对比网络；同时该网络在公开盲道人行道数据集（BACD）和Cityscapes数据集上的综合指标也优于所对比网络。

关键词: 盲道分割, 位置强化, 注意力模块, 多场景, 边界

Abstract: In order to solve the problems of complex backgrounds, blurred boundaries, and diverse blind lane shapes in blind lane segmentation tasks, a blind lane segmentation method based on position enhancement and dual-path edge-enhanced decoding is proposed. Firstly, a position-enhanced feature extraction module is designed, which integrates Mobile- Vitv3 into the down-sampling process to improve the model??s ability to perceive blind lane features and fully preserve contextual information. Secondly, a fusion of channel and position-enhanced attention modules is proposed, enhancing feature extraction in both channel and spatial dimensions, improving the model??s ability to effectively distinguish blind lanes from background information in low-contrast scenes. Finally, the dual-path edge-enhanced decoding method decodes region and boundary information, further optimizing boundary details with a joint loss function. Additionally, in order to address the lack of large-scale public blind lane datasets, a multi-scenario blind track dataset (MSBD) is self-constructed to provide richer data support for model training and experimental validation. Experimental results show that the network achieves mIoU, Precision, Recall, and F1-score of 96.82%, 96.84%, 96.48%, and 96.66% on the MSBD dataset, outperforming SegFormer, Deeplabv3+, etc. With an input image size of 512×512×3, the parameter count and computation count are 1.73×106 and 1.93 GFLOPs , and the inference frame rate reaches 86 FPS. The network also shows superior performance on the blind road and pedestrian crosswalk dataset (BACD) and Cityscapes dataset.

Key words: blind lane segmentation, position enhancement, attention module, multi-scenario, boundary

翁静梁, 王龙业, 曾晓莉, 刘梦瑶, 易婷. 位置强化与双路径边缘增强解码的盲道分割方法[J]. 计算机科学与探索, 2025, 19(9): 2408-2418.

WENG Jingliang, WANG Longye, ZENG Xiaoli, LIU Mengyao, YI Ting. Blind Lane Segmentation Method Based on Position-Enhanced and Dual-Path Edge- Enhanced Decoding[J]. Journal of Frontiers of Computer Science and Technology, 2025, 19(9): 2408-2418.

参考文献

[1] BOURNE R, STEINMETZ J D, FLAXMAN S, et al. Trends in prevalence of blindness and distance and near vision impairment over 30 years: an analysis for the global burden of disease study[J]. The Lancet Global Health, 2021, 9(2): 130-143.
[2] ALVAREZ J M á, ?OPEZ A M. Road detection based on illuminant invariance[J]. IEEE Transactions on Intelligent Transportation Systems, 2011, 12(1): 184-193.
[3] ZHANG H, YE C. An indoor wayfinding system based on geometric features aided graph SLAM for the visually impaired[J]. IEEE Transactions on Neural Systems and Rehabilitation Engineering, 2017, 25(9): 1592-1604.
[4] BAI J Q, LIAN S G, LIU Z X, et al. Virtual-blind-road following-based wearable navigation device for blind people[J]. IEEE Transactions on Consumer Electronics, 2018, 64(1): 136-143.
[5] WEI Y L, LEE M. A guide-dog robot system research for the visually impaired[C]//Proceedings of the 2014 IEEE International Conference on Industrial Technology. Piscataway: IEEE, 2014: 800-805.
[6] 赵磊, 李振伟, 杨晓利, 等. 一种基于图像处理的提示盲道检测方法[J]. 计算机技术与发展, 2021, 31(2): 91-96.
ZHAO L, LI Z W, YANG X L, et al. A warning blind sidewalk detection method based on image processing[J]. Computer Technology and Development, 2021, 31(2): 91-96.
[7] 柯剑光, 赵群飞, 施鹏飞. 基于图像处理的盲道识别算法[J]. 计算机工程, 2009, 35(1): 189-191.
KE J G, ZHAO Q F, SHI P F. Blind way recognition algorithm based on image processing[J]. Computer Engineering, 2009, 35(1): 189-191.
[8] 彭玉青, 薛杰, 郭永芳. 基于颜色纹理信息的盲道识别算法[J]. 计算机应用, 2014, 34(12): 3585-3588.
PENG Y Q, XUE J, GUO Y F. Blind road recognition algorithm based on color and texture information[J]. Journal of Computer Applications, 2014, 34(12): 3585-3588.
[9] 王民, 肖磊, 杨放. 基于显著性检测和改进投影字典对的盲道分割[J]. 激光与光电子学进展, 2017, 54(4): 41001.
WANG M, XIAO L, YANG F. Blind road segmentation based on saliency detection and improved projective dictionary pair[J]. Laser & Optoelectronics Progress, 2017, 54(4): 41001.
[10] YU T, DUAN Y. Blind road deviation detection algorithm based on threshold segmentation[EB/OL]. [2024-06-12]. https://dx.doi.org/10.2139/ssrn.4740536.
[11] 魏彤, 周银鹤. 基于机器学习识别与标记分水岭分割的盲道图像定位[J]. 光学精密工程, 2019, 27(1): 201.
WEI T, ZHOU Y H. Blind sidewalk image location based on machine learning recognition and marked watershed segmentation[J]. Optics and Precision Engineering, 2019, 27(1): 201.
[12] 方超伟, 李雪, 李钟毓, 等. 基于双模型交互学习的半监督医学图像分割[J]. 自动化学报, 2023, 49(4): 805-819.
FANG C W, LI X, LI Z Y, et al. Interactive dual-model learning for semi-supervised medical image segmentation[J]. Acta Automatica Sinica, 2023, 49(4): 805-819.
[13] CAO Z C, XU X W, HU B, et al. Rapid detection of blind roads and crosswalks by using a lightweight semantic segmentation network[J]. IEEE Transactions on Intelligent Transportation Systems, 2021, 22(10): 6188-6197.
[14] CHEN J Z, BAI X Z. Atmospheric transmission and thermal inertia induced blind road segmentation with a large-scale dataset TBRSD[C]//Proceedings of the 2023 IEEE/CVF International Conference on Computer Vision. Piscataway: IEEE, 2023: 1053-1063.
[15] XIA Y Q, LI Y Q, YE Q Q, et al. Image segmentation for blind lanes based on improved SegNet model[J]. Journal of Electronic Imaging, 2023, 32(1): 013038.
[16] BADRINARAYANAN V, KENDALL A, CIPOLLA R. SegNet: a deep convolutional encoder-decoder architecture for image segmentation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(12): 2481-2495.
[17] MA Y N, XU Q, WANG Y, et al. EOS: an efficient obstacle segmentation for blind guiding[J]. Future Generation Computer Systems, 2023, 140: 117-128.
[18] WADEKAR S N, CHAURASIA A. MobileViTv3: mobile-friendly vision transformer with simple and effective fusion of local, global and input features[EB/OL]. [2024-06-12].https://arxiv.org/abs/2209.15159.
[19] CORDTS M, OMRAN M, RAMOS S, et al. The Cityscapes dataset for semantic urban scene understanding[C]//Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2016: 3213-3223.
[20] WAN Q, HUANG Z, LU J, et al. SeaFormer: squeeze-enhanced axial transformer for mobile semantic segmentation[EB/OL]. [2024-06-12]. https://arxiv.org/abs/2301.13156.
[21] ZHANG W Q, HUANG Z L, LUO G Z, et al. TopFormer: token pyramid transformer for mobile semantic segmentation[C]//Proceedings of the 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2022: 12073-12083.
[22] ZHAO H S, SHI J P, QI X J, et al. Pyramid scene parsing network[C]//Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2017: 6230-6239.
[23] 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.
[24] YU C Q, WANG J B, PENG C, et al. BiSeNet: bilateral segmentation network for real-time semantic segmentation[C]//Proceedings of the 15th European Conference on Computer Vision. Cham: Springer, 2018: 334-349.
[25] BAI H T, WANG P H, ZHANG R F, et al. SegFormer: a topic segmentation model with controllable range of attention[J]. Proceedings of the AAAI Conference on Artificial Intelligence, 2023, 37(11): 12545-12552.
[26] CHEN L C, ZHU Y K, PAPANDREOU G, et al. Encoder-decoder with atrous separable convolution for semantic image segmentation[C]//Proceedings of the 15th European Conference on Computer Vision. Cham: Springer, 2018: 833-851.
[27] XU J C, XIONG Z X, BHATTACHARYYA S P. PIDNet: a real-time semantic segmentation network inspired by PID controllers[C]//Proceedings of the 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2023: 19529-19539.
[28] YU C Q, GAO C X, WANG J B, et al. BiSeNet V2: bilateral network with guided aggregation for real-time semantic segmentation[J]. International Journal of Computer Vision, 2021, 129(11): 3051-3068.
[29] WOO S, PARK J, LEE J Y, et al. CBAM: convolutional block attention module[C]//Proceedings of the 15th European Conference on Computer Vision. Cham: Springer, 2018: 3-19.
[30] HU J, SHEN L, SUN G. Squeeze-and-excitation networks[C]//Proceedings of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2018: 7132-7141.
[31] WANG Q L, WU B G, ZHU P F, et al. ECA-Net: efficient channel attention for deep convolutional neural networks[C]//Proceedings of the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2020: 11534-11542.
[32] HOU Q B, ZHOU D Q, FENG J S. Coordinate attention for efficient mobile network design[C]//Proceedings of the 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2021: 13713-13722.