Animal intrusion detection based on convolutional neural network

Wenling Xue, Ting Jiang, Jiong Shi

2017 · 39 citations · 13 references

Concepts

Abstract

The conflict between humans and animals is seen across the country in a variety of forms, including monkey menace in the urban areas, crop raiding by wild pigs and so on. Providing effective solutions for human-animals conflict is now one of the most significant challenges all over the world. In this paper, a wireless sensor network based on UWB technology is used to deploy intrusion detection. By analyzing the characteristics of Ultra wide band (UWB) signals, convolutional neural network is used to learn the characteristics of UWB signals automatically. And finally the SVM or Softmax classifier is used to classify human beings from animals. Several experiments are tested in corn field and the experimental results show that the method proposed in this paper can detect human and animal intrusion very effectively and improve the accuracy of detection by nearly 16% compared to the traditional manual extraction.

References

13