RSWI:a rescue system with WiFi sensing and image recognition
Rescue robots play an important role in the disaster,which are often used to search trapped people with the help of image technology.However,image-based robots have some limitations such as high energy consumption.To address this issue,this paper introduces a rescue system combining WiFi sensing and image recognition with two intelligent robots,which is called RSWI.This rescue system consists of two module:WiFi-based module and image-based mod-ule.WiFi-based module can fast detect the trapped people based on channel state information with a low energy consumption.Once the WiFi-based module detects someone in target area,image-based module will open immediately and execute the visibility calculation method which can determine whether the current environment is suitable for using image recognition technology.If the environment permits,this system will use image recognition technology based on Faster R-CNN to confirm someone existing information.Finally,we evaluate our design in indoor and outdoor environment.The results show that this rescue system is characterized by low energy consumption,reduction of false alarm rate,and adaptation to many scenes.
Rescue robots WiFi Sensing Channel state information image recognition
Jikun Guo Huihui Li
Heilongjiang Institute Of Science And Technology Haerbin,China Heilongjiang Institute of Construction Technology Haerbin,China
国际会议
2019国图灵大会(ACM Turing Celebration conference-China 2019 )
成都
英文
299-302
2019-05-17（万方平台首次上网日期，不代表论文的发表时间）