Abstract
Novelty detection is concerned with detecting features that do not belong to any known class or are not well represented by existing models. Ergo, in autonomous navigation novelty detection determines whether an input camera frame contains certain entities of high interest which do not correspond to a known category. One of the key requirements for the future space exploration missions is the reduction of the information to be transferred back to Earth. Thus, novelty detection techniques have been developed to select the subset of acquired images with significant measurements that justify utilisation of the limited bandwidth from the available information link. Such methods are based on the identification of salient regions, which are then evaluated against a set of trained classifiers. We explore a novelty detection approach, based on the reasoning properties of Neural Networks, which follow the same guidelines while also being trainable in an end-to-end manner. This characteristic allows for the intertwined optimisation of the individual components leading to a closer estimation of a global solution. Our experiments reveal that the proposed novelty detection system achieves better performance, as compared to hand-crafted techniques, when the learning and testing examples refer to similar environments.
Highlights
Planetary Exploration Missions require advanced perception to overcome distance‐induced latency and bandwidth limitations
We explore a novelty detection approach, based on the reasoning properties of Neural Networks, which follow the same guidelines while being trainable in an end‐to‐end manner
Our experiments reveal that the proposed novelty detection system achieves better performance, as compared to hand‐ crafted techniques, when the learning and testing examples refer to similar environments
Summary
Planetary Exploration Missions require advanced perception to overcome distance‐induced latency and bandwidth limitations. A set of binary classifiers, trained on recognising known entities with no particular scientific interest, is applied over each of the RoIs. novelty detection rules are applied, characterising a salient region as an abnormality if the classification results are not conclusive. In order to identify an RoI within an image the objectness measurement produced by RPN can be used as a training method, instead of measuring the saliency of a particular region. On this account, we highlight the main differences between the mechanisms of “objectness” and “saliency”: Objectness is adaptive: For an entity to be detected through saliency, the respective region needs to obey a set of handcrafted rules. All CNN‐based classifiers require a vast amount of learning data and can be biased towards the class taking in the most GT examples
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.