Classification of Buried Objects from Ground Penetrating Radar Images by using Second Order Deep Learning Models
This work addresses the challenge of robust object detection in geophysical imaging for applications like landmine detection, though it appears incremental by combining existing CNN and SPD matrix techniques.
The paper tackles the problem of classifying buried objects from Ground Penetrating Radar images by proposing a new model based on covariance matrices derived from CNN outputs, which outperforms shallow networks and conventional CNNs, especially with limited or mislabeled training data and across different weather conditions.
In this paper, a new classification model based on covariance matrices is built in order to classify buried objects. The inputs of the proposed models are the hyperbola thumbnails obtained with a classical Ground Penetrating Radar (GPR) system. These thumbnails are then inputs to the first layers of a classical CNN, which then produces a covariance matrix using the outputs of the convolutional filters. Next, the covariance matrix is given to a network composed of specific layers to classify Symmetric Positive Definite (SPD) matrices. We show in a large database that our approach outperform shallow networks designed for GPR data and conventional CNNs typically used in computer vision applications, particularly when the number of training data decreases and in the presence of mislabeled data. We also illustrate the interest of our models when training data and test sets are obtained from different weather modes or considerations.