Robust Visual Terrain Classification with Recurrent Neural Networks

A novel approach for robust visual terrain classification by generating feature sequences on repeatedly mutated image patches is pre- sented. These sequences providing the feature vector progress under a cer- tain image operation are learned with Recurrent Neural Networks (RNNs). The approach is studied for image patch based terrain classification for wheeled robots. Thereby, various RNN architectures, namely, standard RNNs, Long Short Term Memory networks (LSTMs), Dynamic Cortex Memory networks (DCMs) as well as bidirectional variants of the men- tioned architecture are investigated and compared to recently used state- of-the-art methods for real-time terrain classification. The results show that the presented approach outperforms previous methods significantly.

[1]  Jürgen Schmidhuber,et al.  Learning to Forget: Continual Prediction with LSTM , 2000, Neural Computation.

[2]  Jürgen Schmidhuber,et al.  Long Short-Term Memory , 1997, Neural Computation.

[3]  Xiaoyang Tan,et al.  Enhanced Local Texture Feature Sets for Face Recognition Under Difficult Lighting Conditions , 2007, IEEE Transactions on Image Processing.

[4]  Matti Pietikäinen,et al.  A comparative study of texture measures with classification based on featured distributions , 1996, Pattern Recognit..

[5]  Thomas M. Breuel,et al.  Using recurrent networks for non-temporal classification tasks , 2014, 2014 International Joint Conference on Neural Networks (IJCNN).

[6]  Sebastian Otte,et al.  JANNLab Neural Network Framework for Java , 2013, MLDM Posters.

[7]  Kuldip K. Paliwal,et al.  Bidirectional recurrent neural networks , 1997, IEEE Trans. Signal Process..

[8]  Andreas Zell,et al.  3D LIDAR- and Camera-Based Terrain Classification Under Different Lighting Conditions , 2012, AMS.

[9]  Andreas Zell,et al.  Terrain classification with conditional random fields on fused 3D LIDAR and camera data , 2013, 2013 European Conference on Mobile Robots.

[10]  Andreas Zell,et al.  Visual terrain classification by flying robots , 2012, 2012 IEEE International Conference on Robotics and Automation.

[11]  Jürgen Schmidhuber,et al.  Learning Precise Timing with LSTM Recurrent Networks , 2003, J. Mach. Learn. Res..

[12]  Sebastian Otte,et al.  Local Feature Based Online Mode Detection with Recurrent Neural Networks , 2012, 2012 International Conference on Frontiers in Handwriting Recognition.

[13]  Andreas Zell,et al.  Dynamic Cortex Memory: Enhancing Recurrent Neural Networks for Gradient-Based Sequence Learning , 2014, ICANN.

[14]  Sebastian Otte,et al.  OCT A-Scan based lung tumor tissue classification with Bidirectional Long Short Term Memory networks , 2013, 2013 IEEE International Workshop on Machine Learning for Signal Processing (MLSP).

[15]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[16]  Jürgen Schmidhuber,et al.  Framewise phoneme classification with bidirectional LSTM and other neural network architectures , 2005, Neural Networks.