Visual Objects Detection Based Robust Ridge Regression Indoor Localization Method

Haowei XU; Baowang LIAN; Xiaojun ZOU; Zhe YUE; Peng WU

doi:10.11999/JEIT170876

Volume 40 Issue 10

Sep. 2018

Turn off MathJax

Article Contents

Article Navigation > Journal of Electronics & Information Technology > 2018 > 40(10): 2453-2460

Haowei XU, Baowang LIAN, Xiaojun ZOU, Zhe YUE, Peng WU. Visual Objects Detection Based Robust Ridge Regression Indoor Localization Method[J]. Journal of Electronics & Information Technology, 2018, 40(10): 2453-2460. doi: 10.11999/JEIT170876

Citation:

Haowei XU, Baowang LIAN, Xiaojun ZOU, Zhe YUE, Peng WU. Visual Objects Detection Based Robust Ridge Regression Indoor Localization Method[J]. Journal of Electronics & Information Technology, 2018, 40(10): 2453-2460. doi: 10.11999/JEIT170876

Citation:

PDF( 2350 KB)

Visual Objects Detection Based Robust Ridge Regression Indoor Localization Method

doi: 10.11999/JEIT170876

School of Electronics and Information, Northwestern Polytechnical University, Xi’an 710072, China

Funds: The National Natural Science Foundation of China (61473308, 61771393)

Received Date: 2017-09-18
Rev Recd Date: 2018-07-23

Available Online: 2018-07-26

Publish Date: 2018-10-01

Abstract

Abstract

The indoor vision positioning algorithm based on object detection is a novel indoor positioning solution, which determines the position of the user through the process of objects detection, position matching, location equation calculation, etc. However, limited by the field-of-view of monocular camera and objects detection accuracy, the localization equation, which is constructed according to the detected objects range information, is seriously ill conditioned. Therefore, this paper proposes a novel localization method based on an improved robust ridge regression estimation, which reduces the influence of the lower accurate observations by iterative weight selection. The experimental results show that compared with Ordinary Least Square (OLS), Levenberg-Marquardt (LM) and Ridge Regression (RR) algorithms, the proposed improved robust ridge regression estimation algorithm can effectively improve the positioning success rate and positioning accuracy of the object detection based indoor navigation method.
- Indoor localization,
- Ridge regression,
- Deep learning

FullText(HTML)

References(16)

References

KOPPANYI Z and TOTH C K. Indoor ultra-wide band network adjustment using maximum likelihood estimation[J]. ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2014, 2(1): 31–35 doi: 10.5194/isprsannals-II-1-31-2014

陈增强, 国峰, 张青. 基于模糊神经网络建模的RFID室内定位算法[J]. 系统科学与数学, 2014, 34(12): 1438–1450

CHEN Zengguo, GUO Feng, and ZHANG Qing. An RFID indoor location algorithm based on fuzzy neural network model[J]. Journal of Systems Science&Mathematical Sciences, 2014, 34(12): 1438–1450

周牧, 唐云霞, 田增山, 等. 基于流形插值数据库构建的WLAN室内定位算法[J]. 电子与信息学报, 2017, 39(8): 1826–1834 doi: 10.11999/JEIT161269

ZHOU Mu, TANG Yunxia, TIAN Zengshan, et al. WLAN indoor localization algorithm based on manifold interpolation database construction[J]. Journal of Electronics&Information Technology, 2017, 39(8): 1826–1834 doi: 10.11999/JEIT161269

CHAE H, PARK C, and SONG J B. Sliding window based landmark extraction for indoor 2D SLAM using forward-looking monocular camera[C]. Control, Automation and Systems (ICCAS), Busan, Korea, 2015: 336–341.

SHIM J H and CHO Y I. A Mobile robot localization using external surveillance cameras at indoor[J]. Procedia Computer Science, 2015, 56: 502–507 doi: 10.1016/j.procs.2015.07.242

LU Guoyu, YAN Yan, SEBE N, et al. Indoor localization via multi-view images and videos[J]. Computer Vision and Image Understanding, 2017, 32: 121–132 doi: 10.1016/j.cviu.2017.05.003

XU Haowei, KOPPANYI Z, TOTH C K, et al. Indoor localization using region-based convolutional neural network[C]. Proceedings of the 2017 International Technical Meeting of ION, Monterey, USA, 2017: 314–323.

TOTH C K, KOPPANYI Z, XU Haowei, et al. Localization using region-based convolution neural network: A comparison study[C]. Proceedings of the 10th International Conference of Mobile Mapping Technology, Cairo, Egypt, 2017: 134–142.

UEDA K and YAMASHITA N. On a global complexity bound of the Levenberg-Marquardt method[J]. Journal of Optimization Theory and Applications, 2010, 147(3): 443–453.

CHEN B H, DA R, and TEKIN I. Location determination using weighted ridge regression[P]. USA, Patent, US6587692, 2003-7-1.

GIRSHICK R, DONAHUE J, DARRELL T, et al. Rich feature hierarchies for accurate object detection and semantic segmentation[C]. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Columbus, USA, 2014: 580–587.

GIRSHICK R. Fast R-CNN[C]. Proceedings of the IEEE International Conference on Computer Vision, Santiago, Chile, 2015: 1440–1448.

REN Shaoqing, HE Kaiming, 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, 2017, 39(6): 1137–1149 doi: 10.1109/TPAMI.2016.2577031

HOERL A E and KENNARD R W. Ridge regression: Biased estimation for nonorthogonal problems[J]. Technometrics, 1970, 12(1): 55–67.

MORÉ J J. The Levenberg-Marquardt Algorithm: Implementation and Theory[M]. Berlin, Heidelberg: Springer, 1978: 105–116.

KIBRIA B M G. Performance of some new ridge regression estimators[J]. Communications in Statistics-Simulation and Computation, 2003, 32(2): 419–435 doi: 10.1081/SAC-120017499

Relative Articles

Supplements(0)

Cited By

Proportional views