Deep Convolutional Autoencoder Applied for Noise Reduction in Range-Doppler Maps of FMCW Radars

In this paper, we discuss the usage of deep Convolutional Autoencoders (CAE) for denoising Range-Doppler (RD) maps of an FMCW radar in a near-field situation with pedestrians and cyclists as moving objects. Traditional methods for noise reduction such as CFAR with Peak Detection (PD) have poor performance under highly noisy environments, especially when objects have low reflectivity, like pedestrians. We propose the use of Convolutional Autoencoders, in various configurations, to overcome those limitations. Due to its Artificial Neural Network nature, CAE can extract features, learn to identify patterns and recognize moving objects - such as humans and bicycles - while the traditional method acts passively. The results indicate that the usage of CAE overcame CFAR with PD by 76.8%, on average, for reconstructing objects in their correct positions. The results of this work may have many applications, for instance, detecting distant moving objects that are too faint for regular detection.

[1]  Nasser Kehtarnavaz,et al.  Convolutional Autoencoder-Based Multispectral Image Fusion , 2019, IEEE Access.

[2]  By Shehrzad Qureshi.  Embedded Image Processing on the Tms320c6000(tm) DSP: Examples in Code Composer Studio(tm) and MATLAB , 2006 .

[3]  S. Z. Gürbüz,et al.  Deep convolutional autoencoder for radar-based classification of similar aided and unaided human activities , 2018, IEEE Transactions on Aerospace and Electronic Systems.

[4]  Eanes Torres Pereira,et al.  Deep Convolutional Autoencoder for EEG Noise Filtering , 2018, 2018 IEEE International Conference on Bioinformatics and Biomedicine (BIBM).

[5]  Massimo Piccardi,et al.  Background subtraction techniques: a review , 2004, 2004 IEEE International Conference on Systems, Man and Cybernetics (IEEE Cat. No.04CH37583).

[6]  S. Blake.  OS-CFAR theory for multiple targets and nonuniform clutter , 1988 .

[7]  A Vengadarajan,et al.  Design of connected component analysis based clustering of CFAR image in pulse Doppler radars , 2017, 2017 IEEE Aerospace Conference.

[8]  Christian Waldschmidt,et al.  Deep Learning for Range-Doppler Map Single Frame Classifications of Cooking Processes , 2018, 2018 15th European Radar Conference (EuRAD).

[9]  Francesco Fioranelli,et al.  Classification of Unarmed/Armed Personnel Using the NetRAD Multistatic Radar for Micro-Doppler and Singular Value Decomposition Features , 2015, IEEE Geoscience and Remote Sensing Letters.

[10]  Qinghua Hu,et al.  Combining multiple neural networks for classification based on rough set reduction , 2003, International Conference on Neural Networks and Signal Processing, 2003. Proceedings of the 2003.

[11]  W. Eric L. Grimson,et al.  Adaptive background mixture models for real-time tracking , 1999, Proceedings. 1999 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (Cat. No PR00149).

[12]  Olivier Bachem,et al.  Recent Advances in Autoencoder-Based Representation Learning , 2018, ArXiv.

[13]  Gerhard Krieger,et al.  Range-Doppler Based CFAR Ship Detection with Automatic Training Data Selection , 2019, Remote. Sens..

[14]  Peter F. McGuire,et al.  Target detection in synthetic aperture radar imagery: a state-of-the-art survey , 2013, 1804.04719.

[15]  Marcin Korytkowski,et al.  Convolutional Neural Networks for Time Series Classification , 2017, ICAISC.

[16]  W. Brown,et al.  Range-Doppler Imaging with Motion through Resolution Cells , 1969, IEEE Transactions on Aerospace and Electronic Systems.

[17]  Mohammed Ghanbari,et al.  Investigating the PSNR calculation methods for video sequences with source and channel distortions , 2014, 2014 IEEE International Symposium on Broadband Multimedia Systems and Broadcasting.

[18]  Traffic Monitoring Object Detection and Tracking Reference Design Using Single-Chip mmWave Radar Sensor , 2017 .