Name
Abstract | ||
|---|---|---|
In this article, we propose the false-alarm-rate-controllable (FAC) classification of sea clutter recurrence plots (RPs) based on convolutional neural networks (CNNs), which is shortened as RPs-CNN. Sea clutter data are a nonlinear and recursive time series, and RPs provide qualitative analysis of nonlinear and recursive dynamic systems. Thus, we construct the RPs' datasets to extract the recursive feature of sea clutter. In the RPs' datasets, RPs' parameters, embedding delay tau and embedding dimension in, are obtained by the average mutual information (AMI) and false nearest neighbor (FNN) algorithms, respectively. In addition, to extract the local features of RPs, CNN is applied. CNN makes full use of local features of the datasets, has the advantages of translation invariance and strong generalization ability, and shares the available weights to simplify network structure. We implement a proper LeNet-5 CNN training on the constructed RPs datasets and verity it using intelligent PIXel (IPIX) processing radar measured datasets. The experimental results demonstrate that CNN successfully classifies the RPs of targets and clutter. Moreover, the feasibility of RPs-CNN is verified by seven aspects, i.e., accuracy, precision, false alarm, miss rate, recall, F1-measure, and Kappa. In addition, six parameters that may affect the classification performance are also analyzed, including time series length, embedding delay, embedding dimension, convolutional kernel size, kernel depth, and optimization function. The results indicate that the proposed RPs-CNN method can reach a 92.05% F1-measure and 87.19% Kappa which performs better than other classification methods. Meanwhile, the false alarm rate (FAR) of RPs-CNN is testified by FAC classification. Experimental results demonstrate that the proposed RPs-CNN significantly improves the detection probability over other classification detectors in low FAR cases. |
Year | DOI | Venue |
|---|---|---|
2022 | 10.1109/TGRS.2022.3192986 | IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING |
Keywords | DocType | Volume |
Classification, convolutional neural network (CNN), false-alarm-rate-controllable (FAC), recurrence plots (RPs), sea clutter | Journal | 60 |
ISSN | Citations | PageRank |
0196-2892 | 0 | 0.34 |
References | Authors | |
0 | 4 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Yanling Shi | 1 | 5 | 3.88 |
| Yaxing Guo | 2 | 0 | 0.34 |
| Tingting Yao | 3 | 0 | 0.34 |
| Zipeng Liu | 4 | 0 | 0.34 |