• 中国计算机学会会刊
  • 中国科技核心期刊
  • 中文核心期刊

J4 ›› 2015, Vol. 37 ›› Issue (08): 1566-1572.

• 论文 • Previous Articles     Next Articles

Aerator state detection based on
corner optical flow and SVM algorithm  

HE Jinhui1,XUE Yueju1,MAO Liang1,LI Hongsheng2,LIN Huankai1,ZHANG Xiao1   

  1. (1.College of Information,South China Agriculture University,Guangzhou 510642;
    2.College of Engineering,South China Agriculture University,Guangzhou 510642,China)
  • Received:2014-05-04 Revised:2014-08-22 Online:2015-08-25 Published:2015-08-25

Abstract:

Realtime detection of the aerator state is an important job in aquiculture. We propose a method to detect the state of the aerator based on corner optical flow and support vector machine (SVM ) model. We firstly collect the videos of stopping and running states of the aerator through camera. Then, we extract two adjacent frames in sequence of the video frames, use the Harris algorithm to detect the interest points of spray in the former frame and calculate its optical flow by the pyramid Lucaskanade algorithm according to the latter frame, thus the average displacement of the interest points between the two frames is obtained. Thirdly, we train the SVM model by the average displacement data of the interest points between the two frames at the learning phase, and utilize the trained SVM model to predict the status of the aerator accordingly at the detecting phase. In addition, we introduce a method to eliminate the frames with abnormal average displacement of the interest points to improve the detection accuracy. Experimental results show that our method is robust and can be adapted to monitor the state of the aerator in real time under conditions such as various illumination, different shooting angles and distances with higher detection accuracy than that of the histogram thresholding method.

Key words: corner detection;optical flow method;image pyramid;support vector machine