›› 2018, Vol. 3 ›› Issue (6): 6-12.

• 数学与计算机科学 • 上一篇    下一篇

洱海湿地昆虫智能识别与实时监测系统

  

  1. (1.大理大学数学与计算机学院,云南大理671003;大理大学昆虫生物医药开发研究院,
    云南大理671003)
  • 收稿日期:2018-03-23 出版日期:2018-06-15 发布日期:2018-06-15
  • 作者简介:罗桂兰,副教授,博士,主要从事物联网、智能生态研究.6
  • 基金资助:

    国家自然科学基金资助项目(61661001);云南省教育厅科学研究基金资助项目(2017ZDX018);云南省大学
    生创新创业训练计划项目(S-CXCY-2017-28)

Insect Intelligent Identification and Real-time Monitoring System in Erhai Wetland

  1. (1.College of Mathematics and Computer, Dali University, Dali, Yunnan 671003, China;2. Institute of Insect Biological and
    Pharmaceutical Development, Dali University, Dali, Yunnan 671003, China)
  • Received:2018-03-23 Online:2018-06-15 Published:2018-06-15

摘要:

为解决洱海湿地隐蔽性昆虫识别与监测问题,借助物联网技术,基于多特征昆虫识别算法设计了一种智能昆虫识别与
实时监测系统。该系统通过红外对射、超声波、声音、颜色和图像等传感器采集昆虫特征信息,在单片机控制下,通过网络将特
征数据传输至云服务器,再经过生命、声音和颜色等多特征数据分析,从而实现智能识别。性能测试表明:与图像边缘监测、相
关反馈法相比,在时间和空间复杂度上具有一定优越性,该方法是可行的。

关键词: 物联网, 湿地昆虫, 智能识别, 实时监测, 多特征提取

Abstract:

In order to identify and monitor the hidden insects in the Erhai wetland, this paper, with the help of the Internet designed
an intelligent insect recognition and real- time monitoring system based on the multifeature insect recognition algorithm. The
information of insect characteristics was collected by infrared radiation sensors, ultrasonic sensors, sound sensors, color sensors and
image sensors. Under the control of the single chip computer, the characteristic data was transmitted to the cloud server through the
network. Then the intelligent recognition was realized by analyzing the characteristics such as life, sound and color. Performance tests
show that: compared with image edge detection and correlation feedback method, it has some advantages in time and space complexity,
and the method is feasible.

Key words: internet of things, wetland insects, intelligent recognition, real-time monitoring, multi-feature extraction

中图分类号: