Wheel detection integrating convolutional neural network and recurrent neural network

doi:10.13474/j.cnki.11-2246.2020.0266

Abstract

Abstract: Target detection is a key technology of target location based on vision. Aiming at the environmental sensitivity of the existing wheel detection methods,this paper proposes a wheel detection model integrating RNN and Faster R-CNN named FusionRNN. Leveraging RNN to deal with temporal information and CNN to extract the recessive characteristics in the spatial domain, it can improve the real-time performance, reduce the number of parameters, and make the model more expressive.At the same time, it has the ability to analyze the semantic relationship between serialization vectors and recognize the geometric features of wheels. The model can accurately detect the wheel in the three-dimensional point cloud scanned by the LiDAR, provide the vehicle position information for the intelligent manipulator used to grab the vehicle to the designated parking space, accurately obtain the vehicle's center of gravity, and realize the safety, rapid capture and storage of the vehicle.

Key words: intelligent garage, wheel detection, RNN, CNN, LiDAR

CLC Number:

P208

MA Chao. Wheel detection integrating convolutional neural network and recurrent neural network[J]. Bulletin of Surveying and Mapping, 2020, 0(8): 139-143.

References

[1] 李仔麒, 马慧彬, 李殿奎, 等. 改进区域生长法的肝部CT图像ROI提取[J]. 计算机技术与发展, 2019, 29(1):150-153.
[2] LOWE D G. Distinctive image features from scale-invariant keypoints[J]. International Journal of Computer Vision, 2004, 60(2):91-110.
[3] BAY H, ESS A, TUYTELAARS T, et al. Speeded-up robust features(SURF)[J]. Computer Vision and Image Understanding, 2008, 110(3):346-359.
[4] DALAL N, TRIGGS B. Histograms of oriented gradients for human detection[C]//Proceedings of 2005 IEEE Conference on Computer Vision and Pattern Recognition.San Diego, CA:IEEE Computer Society, 2005:886-893.
[5] EL KHADIRI I, KAS M, EL MERABET Y, et al. Repulsive- and-attractive local binary gradient contours:new and efficient feature descriptors for texture classification[J].Information Sciences, 2018(467):634-653.
[6] CORTES C, VAPNIK V. Support-vector networks[J]. Machine Learning, 1995(3):273-197.
[7] LODHA S K, FITZPATRICK D M, HELMBOLD D P. Aerial LiDAR data classification using AdaBoost[C]//Proceedings of the 6th International Conference on 3D Digital Imaging and Modeling. Montreal. QC:IEEE Computer Society, 2007:435-442.
[8] 杨必胜, 梁福逊, 黄荣刚. 三维激光扫描点云数据处理研究进展、挑战与趋势[J].测绘学报, 2017,46(10):311-318.
[9] 浦石, 李京伟, 郭四清. 融合语义特征与GPS位置的地面激光点云拼接方法[J].测绘学报,2014,47(6):545-550.
[10] SU H, MAJI S, KALOGERAKIS E, et al. Multi-view convolutional neural networks for 3d shape recognition[C]//Proceedings of 2015 IEEE International Conference on Computer Vision. Santiago:IEEE Computer Society, 2015:945-953.
[11] QI C R, SU H, NIESSNER M, et al. Volumetric and multi-view CNNs for object classification on 3D data[C]//Proceedings of 2016 IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas:IEEE Computer Society, 2016:5648-5656.
[12] WU B,WAN A. YUE X, et al. SqueezeSeg:convolutional neural nets with recurrent CRF for real-time road-object segmentation from 3D LiDAR point cloud[C]//Proceedings of 2018 IEEE International Conference on Robotics and Automation (ICRA). Brisbane, QLD:IEEE Computer Society, 2018:1887-1893.
[13] CHARLES R Q, SU H, KAICHUN M, et al. PointNet:deep learning on point sets for 3D classification and segmentation[C]//Proceedings of 2017 IEEE Conference on Computer Vision and Pattern Recognition. Honolulu, HI:IEEE Computer Society, 2017:77-85.
[14] REN S, HE K, GIRSHICK R, et al. Faster R-CNN:towards real-time object detection with region proposal networks[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2016,39(16):1137-1149.
[15] 杨嘉树, 梅天灿, 仲思东. 顾及局部特性的CNN在遥感影像分类的应用[J]. 计算机工程与应用, 2018, 57(7):188-195.