Lightweight Mask R-CNN for Long-Range Wireless Power Transfer Systems

Published on Oct 1, 2019
路 DOI :10.1109/WCSP.2019.8927856
Hao Li (Tongji University), Aozhou Wu (Tongji University)+ 4 AuthorsChen Wei26
Estimated H-index: 26
(THU: Tsinghua University)
Resonant Beam Charging (RBC) is a wireless charging technology which supports multi-watt power transfer over meter-level distance. The features of safety, mobility and simultaneous charging capability enable RBC to charge multiple mobile devices safely at the same time. To detect the devices that need to be charged, a Mask R-CNN based dection model is proposed in previous work. However, considering the constraints of the RBC system, it's not easy to apply Mask R-CNN in lightweight hardware-embedded devices because of its heavy model and huge computation. Thus, we propose a machine learning detection approach which provides a lighter and faster model based on traditional Mask R-CNN. The proposed approach makes the object detection much easier to be transplanted on mobile devices and reduce the burden of hardware computation. By adjusting the structure of the backbone and the head part of Mask R-CNN, we reduce the average detection time from 1.02s per image to 0.6132s, and reduce the model size from 245MB to 47.1MB. The improved model is much more suitable for the application in the RBC system.
  • References (17)
  • Citations (0)
馃摉 Papers frequently viewed together
2019ICCAE: International Conference on Computer and Automation Engineering
3 Authors (Guo Jianlong, ..., Wen Manhua)
8 Authors (Hongxiang Fan, ..., Wayne Luk)
1 Citations
78% of Scinapse members use related papers. After signing in, all features are FREE.
#1Khaled Ben LetaiefH-Index: 66
#2Chen WeiH-Index: 26
Last. Ying-Jun Angela ZhangH-Index: 2
view all 5 authors...
The recent upsurge of diversified mobile applications, especially those supported by AI, is spurring heated discussions on the future evolution of wireless communications. While 5G is being deployed around the world, efforts from industry and academia have started to look beyond 5G and conceptualize 6G. We envision 6G to undergo an unprecedented transformation that will make it substantially different from the previous generations of wireless cellular systems. In particular, 6G will go beyond mo...
24 CitationsSource
#1Tianyu Wang (NU: Nanjing University)H-Index: 2
#2Shaowei Wang (NU: Nanjing University)H-Index: 21
Last. Zhi-HuaZhou (NU: Nanjing University)H-Index: 83
view all 3 authors...
During the past few decades, mobile wireless communications have experienced four generations of technological revolution, namely from 1G to 4G, and the deployment of the latest 5G networks is expected to take place in 2019. One fundamental question is how we can push forward the development of mobile wireless communications while it has become an extremely complex and sophisticated system. We believe that the answer lies in the huge volumes of data produced by the network itself, and machine le...
1 CitationsSource
Dec 1, 2018 in GLOBECOM (Global Communications Conference)
#1Aozhou Wu (Tongji University)
#2Qingqing Zhang (Tongji University)H-Index: 4
Last. Pengfei Xia (Tongji University)H-Index: 19
view all 7 authors...
Resonant Beam Charging (RBC) is a promising multi-Watt and multi-meter wireless power transfer method with safety, mobility and simultaneously-charging capability. However, RBC system operation relies on information availability including power receiver location, class label and the receiver number. Since smartphone is the most widely-used mobile device, we propose a Mask R-CNN based smartphone detection model in the RBC system. Experiments illustrate that our model reduces the smartphone scanni...
#1Klaus David (University of Kassel)H-Index: 22
#2Hendrik Berndt (University of Kassel)H-Index: 1
To address the question in the subtitle of this article, we start by discussing earlier mobile communication systems, beginning with the first analog wireless cellular standards, then moving on to second generation (2G) [or Global System for Mobile Communications (GSM)], passing third generation (3G) and fourth generation (4G), and proceeding to fifth generation (5G). First, we present each generation's key achievements in terms of user services, each generation's technologyrelated factors of su...
37 CitationsSource
Oct 1, 2017 in ICCV (International Conference on Computer Vision)
#1Kaiming He (Facebook)H-Index: 57
#2Georgia Gkioxari (Facebook)H-Index: 20
Last. Ross Girshick (Facebook)H-Index: 65
view all 4 authors...
We present a conceptually simple, flexible, and general framework for object instance segmentation. Our approach efficiently detects objects in an image while simultaneously generating a high-quality segmentation mask for each instance. The method, called Mask R-CNN, extends Faster R-CNN by adding a branch for predicting an object mask in parallel with the existing branch for bounding box recognition. Mask R-CNN is simple to train and adds only a small overhead to Faster R-CNN, running at 5 fps....
1,349 CitationsSource
Jul 1, 2017 in CVPR (Computer Vision and Pattern Recognition)
#1Fran莽ois Chollet (Google)H-Index: 10
We present an interpretation of Inception modules in convolutional neural networks as being an intermediate step in-between regular convolution and the depthwise separable convolution operation (a depthwise convolution followed by a pointwise convolution). In this light, a depthwise separable convolution can be understood as an Inception module with a maximally large number of towers. This observation leads us to propose a novel deep convolutional neural network architecture inspired by Inceptio...
562 CitationsSource
We present a class of efficient models called MobileNets for mobile and embedded vision applications. MobileNets are based on a streamlined architecture that uses depth-wise separable convolutions to build light weight deep neural networks. We introduce two simple global hyper-parameters that efficiently trade off between latency and accuracy. These hyper-parameters allow the model builder to choose the right sized model for their application based on the constraints of the problem. We present e...
1,995 Citations
#1Qingwen LiuH-Index: 13
#2Jun Wu (Tongji University)H-Index: 1
Last. Lajos Hanzo (University of Southampton)H-Index: 75
view all 7 authors...
Increasing the battery-recharge period of smartphones is becoming a challenge since their power consumption is increased as a result of enhanced functions that require sophisticated multimedia signal processing. An attractive solution is constituted by wireless charging, which is capable of replenishing the battery over the ether. Given this motivation, we present the fundamental physics and the related system structure of a promising wireless charging technique, i.e., distributed laser charging...
33 CitationsSource
Oct 8, 2016 in ECCV (European Conference on Computer Vision)
#1Wei Liu (UNC: University of North Carolina at Chapel Hill)H-Index: 13
#2Dragomir AnguelovH-Index: 24
Last. Alexander C. Berg (UNC: University of North Carolina at Chapel Hill)H-Index: 44
view all 7 authors...
We present a method for detecting objects in images using a single deep neural network. Our approach, named SSD, discretizes the output space of bounding boxes into a set of default boxes over different aspect ratios and scales per feature map location. At prediction time, the network generates scores for the presence of each object category in each default box and produces adjustments to the box to better match the object shape. Additionally, the network combines predictions from multiple featu...
3,523 CitationsSource
Jun 1, 2016 in CVPR (Computer Vision and Pattern Recognition)
#1Kaiming He (Microsoft)H-Index: 57
#2Xiangyu Zhang (Xi'an Jiaotong University)H-Index: 25
Last. Jian Sun (Microsoft)H-Index: 88
view all 4 authors...
Deeper neural networks are more difficult to train. We present a residual learning framework to ease the training of networks that are substantially deeper than those used previously. We explicitly reformulate the layers as learning residual functions with reference to the layer inputs, instead of learning unreferenced functions. We provide comprehensive empirical evidence showing that these residual networks are easier to optimize, and can gain accuracy from considerably increased depth. On the...
14k CitationsSource
Cited By0