时间：2021年4月23日下午2：30

地点：理科楼B311

标题：《Integrating Weighted Feature Fusion and the Spatial AttentionModule with Convolutional Neural Networks for AutomaticAircraft Detection from SAR Images》

汇报人：王杰岚

摘要：

The automatic detection of aircrafts from SAR images is widely applied in both militaryand civil fields, but there are still considerable challenges. To address the high variety of aircraftsizes and complex background information in SAR images, a new fast detection framework based onconvolution neural networks is proposed, which achieves automatic and rapid detection of aircraftwith high accuracy. First, the airport runway areas are detected to generate the airport runwaymask and rectangular contour of the whole airport are generated. Then, a new deep neural networkproposed in this paper, named Efficient Weighted Feature Fusion and Attention Network (EWFAN), isused to detect aircrafts. EWFAN integrates the weighted feature fusion module, the spatial attentionmechanism, and the CIF loss function. EWFAN can effectively reduce the interference of negativesamples and enhance feature extraction, thereby significantly improving the detection accuracy.Finally, the airport runway mask is applied to the detected results to reduce false alarms and producethe final aircraft detection results. To evaluate the performance of the proposed framework, largescale Gaofen-3 SAR images with 1 m resolution are utilized in the experiment. The detection rateand false alarm rate of our EWFAN algorithm are 95.4% and 3.3%, respectively, which outperformsEfficientdet and YOLOv4. In addition, the average test time with the proposed framework is only15.40 s, indicating satisfying efficiency of automatic aircraft detection..

录取期刊：remote sensing

标题：《HOTSPOT: A UAV-Assisted Dynamic Mobility-Aware Offloading for Mobile Edge Computing in 3D Space》

汇报人：马银宝

摘要：

For massive access to the Internet of Things, edge computing servers are installed on cellular ground base stations (GBS) with fixed geographical locations, which easily suffer from traffic overload of the end-user (EU) with high density and mobility. To provide reliable and flexible offloading service, unmanned aerial vehicles (UAV) are explored to assist edge computing, which relieves the computation offloading pressure of both EUs and GBS. However, most existing UAV researches focus on trajectory design to reduce offloading delay, which ignoring the variability of user distribution and the energy limitation of UAV. This paper proposes a novel UAV-assisted edge computing framework, named as HOTSPOT, which locates the UAV in 3D space according to the time-varying hot spot of user distribution and provides the corresponding edge computing offloading assistance. By formulating the UAV positioning problem into a maximum clique problem, a light-weighted deterministic algorithm is proposed based on stochastic gradient descent to search the optimal location of UAV. With the elaborate UAV position, HOTSPOT further gives an opportunistic offloading balanced scheme to reach low latency. Simulation results show that when the GBS load is 75%, HOTSPOT reduces the average offloading delay by 33%. When the GBS load reaches 90%, the average delay reduction is up to 80%..

录取期刊：IEEE Internet of Things Journal

标题：《Lightweight Feedback Convolution NeuralNetwork for Remote Sensing ImagesSuper-Resolution》

汇报人：吴一鸣

摘要：

There are lots of image data in the eld of remote sensing, most of which have low-resolutiondue to the limited image sensor. The super-resolution method can effectively restore the low-resolutionimage to the high-resolution image. However, the existing super-resolution method has both heavy computingburden and number of parameters. For saving costs, we propose the feedback ghost residual densenetwork (FGRDN), which considers the feedback mechanism as the framework to attain lower featuresthrough high-level rening. Further, for feature extraction, we replace the convolution of the residual denseblocks (RDBs) with ghost modules (GMs), which can remove the redundant channels and avoid the increaseof parameters along with the network depth. Finally, the spatial and channel attention module (SCM) isemployed in the end of the RDB to learn more useful information from features. Compared to other SOTAlightweight algorithms, our proposed algorithm can reach convergences more rapidly with fewer parameters,and the performance of the network can be markedly enhanced on the image texture and object contourreconstruction with better peak signal-to-noise ratio (PSNR) and structural similarity (SSIM).

录取期刊：IEEEACCESS