Analysing the Ability of SLAM Algorithms Fused with CNN to Acquire Deep Information

Di Wu

doi:10.54097/d34b9s42

Authors

Di Wu

DOI:

https://doi.org/10.54097/d34b9s42

Keywords:

Convolutional Neural Network, Simultaneous localization and mapping, Deep Learning, texture features.

Abstract

Given that SLAM has benefited from the development of vision technology in the last decade, it has been widely popularised into people's daily life, such as drones and UAVs using sensors to acquire environmental data and construct maps to avoid obstacles and achieve path planning. However, since the extent of SLAM technology is mostly based on the texture features of the input image, SLAM is prone to misjudge the depth information when the features of the input image are not obvious. Combined with the current development of Deep Learning, the paper can consider integrating the Convolutional Neural Network (CNN), which has a good processing effect on image data, into the SLAM algorithm, so as to improve the effectiveness and accuracy of the SLAM algorithm in obtaining depth data. This paper will analyse the basic principles of SLAM and CNN, and discuss the current existence of CNN and SLAM fusion related technology.

Downloads

Download data is not yet available.

References

[1] Chang Q L, Hung K W, Jiang J M. Deep learning based image Super-resolution for nonlinear lens distortions. Neurocomputing, 2018, 275: 969-982.

[2] Yuan Z K, Cheng K, Tang J H, et al. RGB-D DSO: Direct Sparse Odometry With RGB-D Cameras for Indoor Scenes. Ieee Transactions on Multimedia, 2022, 24: 4092-4101.

[3] Li L, Yang M. Point Cloud Registration Based on Direct Deep Features With Applications in Intelligent Vehicles. Ieee Transactions on Intelligent Transportation Systems, 2022, 23: 13346-13357.

[4] Chen M Q, Yu L J, Zhi C, et al. Improved faster R-CNN for fabric defect detection based on Gabor filter with Genetic Algorithm optimization. Computers in Industry, 2022, 134.

[5] He K M, Gkioxari G, Dollar P, et al. Mask R-CNN. Ieee Transactions on Pattern Analysis and Machine Intelligence, 2020, 42: 386-397.

[6] Keisuke T, Federico T, Iro L, Nassir N. CNN-SLAM: Real-time dense monocular SLAM with learned depth prediction. Computer Vision and Pattern Recognition, 2017.

[7] Egrioglu E. Median-Pi artificial neural network for forecasting. Neural computing & applications, 2019, 31.

[8] Pang B, Nijkamp E, Wu Y N. Deep Learning with TensorFlow: A Review. Journal of Educational and Behavioral Statistics, 2020, 45: 227-248.

[9] Shiozaki T, Dissanayake G. Eliminating Scale Drift in Monocular SLAM Using Depth from Defocus. Ieee Robotics and Automation Letters, 2018, 3: 581-587.

[10] Zhou D F, Dai Y C, Li H D. Ground-Plane-Based Absolute Scale Estimation for Monocular Visual Odometry. Ieee Transactions on Intelligent Transportation Systems, 2020, 21: 791-802.

[11] Zhang Z, Huang Y H, Bao S, et al. Panoramic annular image restoration algorithm by prediction based on the lens design characteristics. Applied Optics, 2023, 62: 518-527.

[12] Engel J, Schps T, and Cremers D. LSD-SLAM: LargeScale Direct Monocular SLAM. In European Conference on Computer Vision (ECCV), 2014.

[13] Zhang Y J, Zhao J X, Han H Y. A 3D Machine Vision-Enabled Intelligent Robot Architecture. Mobile Information Systems, 2021.

[14] Zhou Y, Yan F H, Zhou Z. Handling pure camera rotation in semi-dense monocular SLAM. Visual Computer, 2019, 35: 123-132.

[15] Niu Z Y, Zhong G Q, Yu H. A review on the attention mechanism of deep learning. Neurocomputing, 2021, 452: 48-62.

[16] Lu J, Wang Z, Hua B W, et al. Automatic point cloud registration algorithm based on the feature histogram of local surface. Plos One, 2020, 15.

[17] Cui Y, Yu Z K, Han J C, et al. Dual-Triple Attention Network for Hyperspectral Image Classification Using Limited Training Samples. Ieee Geoscience and Remote Sensing Letters, 2022, 19.

[18] Chen S Y, Zhang Z X, Zhong R F, et al. A Dense Feature Pyramid Network-Based Deep Learning Model for Road Marking Instance Segmentation Using MLS Point Clouds. Ieee Transactions on Geoscience and Remote Sensing, 2021, 59: 784-800.

[19] Sakpere W, Adeyeye-oshin M, Mlitwa N B W. A state-of-the-art survey of indoor positioning and navigation systems and technologies. South African Computer Journal, 2017, 29: 145-197.

[20] Wang P, Shen X, Lin Z, Cohen S, Price B, and Yuille A L. Towards unified depth and semantic prediction from a single image. In Proc. Conf. Computer Vision and Pattern Recognition (CVPR), pages 2800–2809, 2015.