Design And Implementation Of The Automatic Identification Of Sogatella Furcifera System

As the most important food crop,rice plays decisive part in agricultural production and food security.Sogatella furcifera is one of the major rice pests threatening high and stable production of rice.Therefore,we must monitor and predict the species and number of Sogatella furcifera accurately.We designed the system of automatic identification of Sogatella furcifera based on image processing technology according to the opinion of plant experts and entomologist to quantify the color,texture and geometric shape of the black of Sogatella furcifera.A automatic acquisition device had been designed and made to obtain the clear image of Sogatella furcifera.The acquistion device is composed of the mechanical platform of collecting device and automatic shooting control system.The mechanical platform of collecting device is composed of table,collection table and transmission system.The table supports the whole experiment platform and is a 100cm X 100cm framework shape.The collection table placing white curtain provides place for Sogatella furcifera stay and the size is 120cm X 90cm.The transmission system is responsible for driving the camera movement along two dimensional.The automatic shooting control system is composed of shooting system and motion control system.The shooting system is composed of camera,telecentric lens,the image acquisition card and RI ring light and takes clear image of Sogatella furcifera.The motion control system is composed of servo motor,servo drive and PLC.The servo driver receives plus which is emitted by PLC and drives servo motor to move camera along two dimensional of X-axis and Y-axis.At the same time,PLC tigers camera to shoot screen orderly.Finally,we got the Sogatella furcifera image under the natural state.One hundred and thirty-one images of the Sogatella furcifera were captured in the field by using a automatic acquisition device which was developed by ourself.After filtering,these images were segmented according to the threshold value(Blue component.B=130)and the banalized images were obtained.Then the banalized and the the gray images of back regional of the individual insects were extracted.After the insects which is obviously different from the Sogatella furcifera in the size and color were removed from the images by the statistics analysis of the size and color of Sogatella furcifera images,the moment invariant and 2-d Fourier spectrum data were used to extract eighty-eight characteristic parameters related to the geometry,color and texture of Sogatella furcifera back images.The seven parameters and thelxl(l=1,2,...,9)2-d logarithm spectrum data which extracted from the Sogatella furcifera back images were combined into a feature vector.Finally,based on the feature vector,a classification model of Sogatella furcifera was developed using the support vector technology.The experimental results showed that the correct identification rate of Sogatella furcifera sample is 95%,showing the proposed method can be used to identify Sogatella furcifera automatically.