Deep-learning techniques have significantly improved object detection performance, especially with binocular images in 3D scenarios. To supervise the depth information in stereo 3D object detection, reconstructing the 3D dense depth of LiDAR point clouds causes higher computational costs and lower inference speed. After exploring the intrinsic relationship between the implicit depth information and semantic texture features of the binocular images, we propose an efficient and accurate 3D object detection algorithm, FCNet, in stereo images. First, we construct a multi-scale cost–volume containing implicit depth information using the normalized dot-product by generating multi-scale feature maps from the input stereo images. Secondly, the variant attention model enhances its global and local description, and the sparse region monitors the depth loss deep regression. Thirdly, for balancing the channel information preservation of the re-fused left–right feature maps and computational burden, a reweighting strategy is employed to enhance the feature correlation in merging the last-layer features of binocular images. Extensive experiment results on the challenging KITTI benchmark demonstrate that the proposed algorithm achieves better performance, including a lower computational cost and higher inference speed in 3D object detection.
Currently a number of research efforts have been paid to study Transverse Flux magnetic Motor (TFM) and Switched Reluctance Motor (SRM). Generally, TFM is energy efficient and has low reluctance but it is costly in manufacture. On the contrary, because of the simple structure, SRM is produced easily but it has high energy expenditure. In order to take advantages of TFM and SRM, we introduce a novel Transverse Flux Switched Magnetic Energy Motor (TFSMEM) and its experimental prototype. In this paper, firstly we present the internal structure, operation principle and technical parameters of designed prototype. Based on this, we conduct the theoretical analysis of magnetic field and formulate the model of prototype. Meanwhile, according to the analysis results above, we present the finite element simulation model through finite element analysis. Then, we finished the design of servo control system including the hardware and software parts. Finally, based on cSPACE simulation system and MTS-I motor testing system, our experiments validate that our prototype outperforms the TFM and SFM both in mechanical and load characteristics. In addition, the servo control system can control the prototype operation well.
1 January 2018 in Wireless Communications and Mobile Computing
Appropriate resource allocation has great significance to enhance the energy efficiency (EE) for cooperative communication system. The objective is to allocate the resource to maximize the energy efficiency in single‐cell multiuser cooperative communication system. We formulate this problem as subcarrier‐based resource allocation and solve it with path planning in graph theory. A two‐level neural network model is designed, in which the users and subcarrier are defined as network nodes. And then we propose an improved intelligent water drops algorithm combined with Genetic Algorithm; boundary condition and initialization rules of path soil quantity are put forward. The simulation results demonstrate that the proposed resource allocation scheme can effectively improve the energy efficiency and enhance QoS performance.