In the task of space target recognition, due to the limited conditions of image acquisition, the accumulated train dataset and test dataset have obvious differences in different poses, and the drastic changes of image representation caused by cross-view greatly restrict the performance of space target recognition. In this paper, cross-view space target recognition from multi-view stereo is proposed. The method is based on multi-view stereo to estimate the image depth information, which is fused with texture information of 2D image, and finally uses the 3D CNN to classify the image. The network learns and inferences depth information without the supervision of depth map truth value, which adds 3D information such as geometry of space objects to the classification task, and improves the performance of target recognition. The experimental results on the self-made space target data set prove that the proposed network MVS-STRNet is more efficient on the cross-view data set.
In actual urban scenes, the uncanonical facilities complicate the EM wave propagation environment, which makes traditional methods hard to work. To solve this problem, a novel method for Localization based on ray tracing is established. Firstly, the ray tracing technique is introduced to calculate the theoretical propagation path from the target, by which a dictionary of predicted multipath signals relative to different position parameters is constructed. Consequently, the position of the target can be estimated via matching Euclidean distance between the received multipath echoes and the predicted multipath dictionary. The simulation experiments show that the proposed method has higher localization accuracy especially in complicated scene.
Multipath exploitation radar (MER) integrates the prior environment to make use of the extra target information encoded in multipath signals, which is capable of localization for single target with a wide-beam antenna. Nevertheless, as the number of targets in the urban scene increases, the association between the target and its corresponding multipath’s time-of- arrival (ToA) faces the problem of combination explosion. Moreover, accounting for the measurement and extraction error attached with the ToAs, there may be multiple combinations with similar probability, which leads to a significant accuracy decline or even wrong location results. To solve the problem, this paper proposes a novel algorithm that jointly uses MER and time-frequency (TF) features for multi-target localization especially for pedestrians. Through the TF analysis of the micro-Doppler feature, the pedestrian characteristics such as pace and phase of periodic action can be obtained. Based on these characteristics, the multipath’s ToAs in multi-target scenario can be associated with each different target, hence the aforementioned multi-target location problem can be transferred to a series of single-target localization problems. The impacts of target number on localization accuracy are analyzed in detail. The effectiveness of the proposed method is validated through the simulation experiments. The results indicate that, compared with the traditional methods, the proposed method has higher localization accuracy in multi-target scene.
A planar-array multiple-input-multiple-output (MIMO) radar system possess the ability to gain a three-dimensional (3-D) image in single snapshot due to the wide-band signal and two-dimensional (2-D) virtual aperture. And the conventional inverse synthetic aperture radar (ISAR) obtains the cross-range resolution thanks to the relative rotational movement during the observation. Naturally, the planar-array MIMO radar 3-D images in multiple snapshots also include slow-time domain Doppler information. In order to take advantage of the Doppler shift along the slow-time domain for a better 3-D imaging result, we investigate the method of MIMO radar 3-D imaging via jointly utilizing the time-space observation. By coherent processing along the velocity direction, inverse aperture caused by target movement is incorporated into the 3-D image focusing and therefore the resolution can be increased. Simulation results validate the effectiveness of the proposed method. Comparing to ISAR, the longtime observation as well as the complicated motion compensation in the proposed 3-D imaging method is not necessary. Besides, comparing to the 3-D image in single snapshot, the proposed method can improve the resolution along the target trajectory efficiently.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.