Although they achieve satisfactory results in natural images, direct transfer to SAR images for ship detection tasks is difficult. In the field of object recognition, the detection methods based on CNN can be classified into two categories (1): single-stage detectors such as SDD, Yolov1/v2v3/v4, and RetinaNet and (2) two-stage detectors such as R-CNN, Fast R-CNN, Faster R-CNN, and Mask-RCNN. The development and success of artificial intelligence have contributed to the tremendous application of convolutional neural networks (CNNs), such as object recognition, image classification, and automatic object clustering. Moreover, the setting of the protection and background window sizes not only affects the detection result but also takes a long time. However, applying a sliding window when processing SAR images by CFAR is necessary. In the previous studies, constant false alarm rate (CFAR), as a classical target detection method, has been usually used for ship detection. Experimental results show that the performance of the proposed method for small-sized ship detection can achieve 97.78% accuracy and 0.96 F1-score with Sentinel-1 images in complex nearshore areas. Finally, a lightweight CNN embedded with active learning scheme is used to classify the ships using only a small labeled training set. Then, the strategy of boundary box optimization is defined and applied to reduce the noise and false alarms caused by island and land targets as well as by sidelobe interference. The candidate target detection results are obtained by global threshold segmentation. In this study, a novel method based on candidate target detection, boundary box optimization, and convolutional neural network (CNN) embedded with active learning strategy is proposed to improve the accuracy and efficiency of ship detection in nearshore areas. In addition, the land and island with high backscattering often cause high false alarms for ship detection in the nearshore area.
TRANSDATA TEST SET 2100 MANUAL
Manual labeling each datum not only takes too much time but also requires a high degree of professional knowledge. Another difficulty is that a huge amount of data needs to be manually labeled to obtain a reliable CNN model.
Although deep learning can achieve a much better detection performance than traditional methods, it is difficult to achieve satisfying performance for small-sized ships nearshore due to the weak scattering caused by their material and simple structure.
The rapid development of big data and artificial intelligence technology has facilitated the wide use of deep learning methods in SAR imagery for ship detection.
TRANSDATA TEST SET 2100 MAC
Microsoft Windows 98SE, NT, 2000, XP, Vista™ or Windows 7, 8, 8.1, 10, Mac OS, NetWare, UNIX or LinuxInternet Explorer 11, Firefox 12.0, Chrome 20.0, Safari 4.Marine ship detection by synthetic aperture radar (SAR) is an important remote sensing technology. Parental Controls, Network Address Translation (NAT), StaticRouting, RIP v1/v2 (optional), DNS Relay, DDNS, IGMP V1/V2/V3Ģ.4 GHz Guest Network, 5 GHz Guest NetworkĭSL Modem RouterExternal SplitterRJ11 DSL CableRJ45 Ethernet CablePower AdapterQuick Installation Guide Virtual Server, Port Triggering, DMZ, ALG, UPnPĪTM Forum UNI3.1/4.0, PPP over ATM (RFC 2364), PPP overEthernet (RFC2516), IPoA (RFC1577/2225), MER\IPoE (RFC 1483 Routed), Bridge (RFC1483 Bridge), PVC – Up to 8 PVCs Server, Client, DHCP Client List, Address Reservation, DHCP Relay
TRANSDATA TEST SET 2100 UPDATE
Web-Based Configuration (HTTP / HTTPS ), Remote management,Command Line Interface, SSL for TR-069, SNMP v1/2c, Firmware Upgrade, Diagnostic Tools, Free Online Firmware Update VDSL/ADSL Modem Router, 3G/4G Router, Wireless Router NAT Firewall, SPI Firewall, Access Control, Service Filtering, Denial of Service (DoS), SYN Flooding, Ping of Death, IP and MAC Address Binding Support 2.4 GHz and 5 GHz Wireless Schedule IEEE 802.11ac/n/a 5 GHz, IEEE 802.11b/g/n 2.4 GHzĮnable/Disable Wireless Radio, Wireless StatisticsĦ4/128-bit WEP, WPA/WPA2, WPA-PSK/WPA-PSK2encryptions WPS Button, Reset Button, Power On/Off Button, Wi-Fi Button, LEDOn/Off Button 3 Gigabit LAN Ports, 1 Gigabit LAN/WAN Port, 1 RJ11 Port,1 USB Port
0 kommentar(er)
