Outdoor Landmark Detection for Real-World Localization using Faster R-CNN

This paper presents a method for outdoor localization using deep learning-based landmark detection. The proposed localization method relies on the Faster Regional Convolutional Neural Network (Faster R-CNN) landmark detector and the feedforward neural network (FFNN) trained with GPS data from geotags in images, retrieve location coordinates and compass orientation of the implemented device based on detected landmarks in the image. Results of the proposed localization method are illustrated with errors from the comparisons between results of the localization and geotags data within the images. The experiment results pointed the proposed method to be the promising alternative to conventional ways of outdoor localization.

[1]  Yazhe Tang,et al.  Vision-Aided Multi-UAV Autonomous Flocking in GPS-Denied Environment , 2019, IEEE Transactions on Industrial Electronics.

[2]  Nestor Michael C. Tiglao,et al.  RTKLIB-based GPS localization for multipath mitigation in ITS applications , 2016, 2016 Eighth International Conference on Ubiquitous and Future Networks (ICUFN).

[3]  Hui Zhang,et al.  Pedestrian Detection Method Based on Faster R-CNN , 2017, 2017 13th International Conference on Computational Intelligence and Security (CIS).

[4]  Kaiming He,et al.  Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks , 2015, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[5]  C. V. Jawahar,et al.  Accurate localization by fusing images and GPS signals , 2015, 2015 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW).

[6]  Hyeung-Sik Choi,et al.  Development of GPS-aided localization algorithm of Autonomous Underwater Vehicle , 2017, 2017 IEEE Underwater Technology (UT).

[7]  Eyuphan Bulut,et al.  Clustered Crowd GPS for Privacy Valuing Active Localization , 2018, IEEE Access.

[8]  Saeid Nahavandi,et al.  Cyclist detection in LIDAR scans using faster R-CNN and synthetic depth images , 2017, 2017 IEEE 20th International Conference on Intelligent Transportation Systems (ITSC).

[9]  Eric Sax,et al.  GPS-independent localization for off-road vehicles using ultra-wideband (UWB) , 2017, 2017 IEEE 20th International Conference on Intelligent Transportation Systems (ITSC).

[10]  Bin Guo,et al.  The emergence of visual-based localization and navigation using smartphone sensing , 2017, 2017 IEEE SmartWorld, Ubiquitous Intelligence & Computing, Advanced & Trusted Computed, Scalable Computing & Communications, Cloud & Big Data Computing, Internet of People and Smart City Innovation (SmartWorld/SCALCOM/UIC/ATC/CBDCom/IOP/SCI).

[11]  Ross B. Girshick,et al.  Fast R-CNN , 2015, 1504.08083.

[12]  Muhammad Bilal Kadri,et al.  Sensor fusion of INS, odometer and GPS for robot localization , 2016, 2016 IEEE Conference on Systems, Process and Control (ICSPC).

[13]  Yun Fu,et al.  Deep Geo-Constrained Auto-Encoder for Non-Landmark GPS Estimation , 2019, IEEE Transactions on Big Data.

[14]  Sever Pasca,et al.  Fall detection system for elderly with GSM communication and GPS localization , 2013, 2013 8TH INTERNATIONAL SYMPOSIUM ON ADVANCED TOPICS IN ELECTRICAL ENGINEERING (ATEE).

[15]  Ren C. Luo,et al.  Dynamic Wireless Indoor Localization Incorporating With an Autonomous Mobile Robot Based on an Adaptive Signal Model Fingerprinting Approach , 2019, IEEE Transactions on Industrial Electronics.

[16]  Colin Greatwood,et al.  Visual Localisation and Individual Identification of Holstein Friesian Cattle via Deep Learning , 2017, 2017 IEEE International Conference on Computer Vision Workshops (ICCVW).