SAR Target Recognition with Limited Training Samples in Open Set Conditions
It is difficult to collect training samples for all types of synthetic aperture radar (SAR) targets. A realistic problem comes when unseen categories exist that are not included in training and benchmark data at the time of recognition, which is defined as open set recognition (OSR). Without the aid of side-information, generalized OSR methods used on ordinary optical images are usually not suitable for SAR images. In addition, OSR methods that require a large number of samples to participate in training are also not suitable for SAR images with the realistic situation of collection difficulty. In this regard, a task-oriented OSR method for SAR is proposed by distribution construction and relation measures to recognize targets of seen and unseen categories with limited training samples, and without any other simulation information. The method can judge category similarity to explain the unseen category. Distribution construction is realized by the graph convolutional network. The experimental results on the MSTAR dataset show that this method has a good recognition effect for the targets of both seen and unseen categories and excellent interpretation ability for unseen targets. Specifically, while recognition accuracy for seen targets remains above 95%, the recognition accuracy for unseen targets reaches 67% for the three-type classification problem, and 53% for the five-type classification problem.
| Medienart: |
E-Artikel |
|---|
| Erscheinungsjahr: |
2023 |
|---|---|
| Erschienen: |
2023 |
| Enthalten in: |
Zur Gesamtaufnahme - volume:23 |
|---|---|
| Enthalten in: |
Sensors (Basel, Switzerland) - 23(2023), 3 vom: 02. Feb. |
| Sprache: |
Englisch |
|---|
| Beteiligte Personen: |
Zhou, Xiangyu [VerfasserIn] |
|---|
| Links: |
|---|
| Themen: |
Graph convolutional network (GNN) |
|---|
| Anmerkungen: |
Date Completed 13.02.2023 Date Revised 14.02.2023 published: Electronic Citation Status PubMed-not-MEDLINE |
|---|
| doi: |
10.3390/s23031668 |
|---|
| funding: |
|
|---|---|
| Förderinstitution / Projekttitel: |
|
| PPN (Katalog-ID): |
NLM35282414X |
|---|
| LEADER | 01000naa a22002652 4500 | ||
|---|---|---|---|
| 001 | NLM35282414X | ||
| 003 | DE-627 | ||
| 005 | 20231226054614.0 | ||
| 007 | cr uuu---uuuuu | ||
| 008 | 231226s2023 xx |||||o 00| ||eng c | ||
| 024 | 7 | |a 10.3390/s23031668 |2 doi | |
| 028 | 5 | 2 | |a pubmed24n1176.xml |
| 035 | |a (DE-627)NLM35282414X | ||
| 035 | |a (NLM)36772708 | ||
| 035 | |a (PII)1668 | ||
| 040 | |a DE-627 |b ger |c DE-627 |e rakwb | ||
| 041 | |a eng | ||
| 100 | 1 | |a Zhou, Xiangyu |e verfasserin |4 aut | |
| 245 | 1 | 0 | |a SAR Target Recognition with Limited Training Samples in Open Set Conditions |
| 264 | 1 | |c 2023 | |
| 336 | |a Text |b txt |2 rdacontent | ||
| 337 | |a ƒaComputermedien |b c |2 rdamedia | ||
| 338 | |a ƒa Online-Ressource |b cr |2 rdacarrier | ||
| 500 | |a Date Completed 13.02.2023 | ||
| 500 | |a Date Revised 14.02.2023 | ||
| 500 | |a published: Electronic | ||
| 500 | |a Citation Status PubMed-not-MEDLINE | ||
| 520 | |a It is difficult to collect training samples for all types of synthetic aperture radar (SAR) targets. A realistic problem comes when unseen categories exist that are not included in training and benchmark data at the time of recognition, which is defined as open set recognition (OSR). Without the aid of side-information, generalized OSR methods used on ordinary optical images are usually not suitable for SAR images. In addition, OSR methods that require a large number of samples to participate in training are also not suitable for SAR images with the realistic situation of collection difficulty. In this regard, a task-oriented OSR method for SAR is proposed by distribution construction and relation measures to recognize targets of seen and unseen categories with limited training samples, and without any other simulation information. The method can judge category similarity to explain the unseen category. Distribution construction is realized by the graph convolutional network. The experimental results on the MSTAR dataset show that this method has a good recognition effect for the targets of both seen and unseen categories and excellent interpretation ability for unseen targets. Specifically, while recognition accuracy for seen targets remains above 95%, the recognition accuracy for unseen targets reaches 67% for the three-type classification problem, and 53% for the five-type classification problem | ||
| 650 | 4 | |a Journal Article | |
| 650 | 4 | |a graph convolutional network (GNN) | |
| 650 | 4 | |a limited training samples | |
| 650 | 4 | |a open set recognition (OSR) | |
| 650 | 4 | |a relation measure | |
| 650 | 4 | |a synthetic aperture radar (SAR) | |
| 700 | 1 | |a Zhang, Yifan |e verfasserin |4 aut | |
| 700 | 1 | |a Liu, Di |e verfasserin |4 aut | |
| 700 | 1 | |a Wei, Qianru |e verfasserin |4 aut | |
| 773 | 0 | 8 | |i Enthalten in |t Sensors (Basel, Switzerland) |d 2007 |g 23(2023), 3 vom: 02. Feb. |w (DE-627)NLM187985170 |x 1424-8220 |7 nnns |
| 773 | 1 | 8 | |g volume:23 |g year:2023 |g number:3 |g day:02 |g month:02 |
| 856 | 4 | 0 | |u http://dx.doi.org/10.3390/s23031668 |3 Volltext |
| 912 | |a GBV_USEFLAG_A | ||
| 912 | |a GBV_NLM | ||
| 951 | |a AR | ||
| 952 | |d 23 |j 2023 |e 3 |b 02 |c 02 | ||