International Research Journal of Engineering and Technology (IRJET) Volume: 04 Issue: 03 | Mar -2017
www.irjet.net
e-ISSN: 2395 -0056 p-ISSN: 2395-0072
RECOGNITION OF SILVERLEAF WHITEFLY AND WESTERN FLOWER THRIPS VIA IMAGE PROCESSING AND ARTIFICIAL NEURAL NETWORK Deepa Nair, Ankita Parte, Yogita Pokharkar, Nikita Pande 1Student,Department
of Computer Engineering,D.Y.P.I.E.T Pimpri,Maharashtra,India of Computer Engineering,D.Y.P.I.E.T Pimpri,Maharashtra,India 3Student,Department of Computer Engineering,D.Y.P.I.E.T Pimpri,Maharashtra,India 4Student,Department of Computer Engineering,D.Y.P.I.E.T Pimpri,Maharashtra,India ---------------------------------------------------------------------***--------------------------------------------------------------------for complete disease detection phase. We will take an Abstract - IPM(Integrated Pest Management) is used to 2Student,Department
minimize or reduce the use of chemicals in greenhouse agriculture. IPM is basically depends upon early detection and continuous monitoring of pest populations which is very critical or time consuming task as it require continuous monitoring and it also dependent on human judgment due to this it has lots of error. To minimize this error, we propose a general approach for finding and monitoring of adult-stage whitefly and thrip in greenhouses which is based on the grouping of two process first is image-processing algorithm and second is artificial neural networks.For image processing a sticky trap paper is used from which image is taken by the process of image acquisition system. The detection of the objects from the images is perform through segmentation, and morphological and color property opinion performed by an image processing algorithm. Lastly the identification of the objects is performing through feed-forward multi-layer artificial neural network. Key Words: — IPM, Early pest detection, Insect identification, Image processing, Artificial neural network.
image of a defected plant leaves as a input and take out the features of leaves. In our project we will consider color as feature. By using this feature we will compare our defected plant leaves with the database present there. We are going to use Artificial Neural Network as our classifier for comparison of leaves. An artificial neural network (ANN), usually called neural network (NN). A neural network consists of an interrelated group of artificial neurons, and it process information using connectionist approach for computation.In the majority cases ANN is an adaptive system which changes its structure based on external or internal information that flows through the network during the learning phase. We have formed a database of diseased
1. INTRODUCTION
cotton leaf considering two different diseases they are identified as Silverleaf Whitefly and Western flower
In this project we are going to create a system called
thrips.We have to take out the separate H, S and V
Artificial Neural Network Technology to detect and
features and compare this features with the features
classify leaves diseases. On plants generally 80 to 90 %
that are extracted from the input test image. We have
of disease is on its leaves. So due to this reason our
done various preprocessing steps on the input test
study of interest is leaf of the tree rather than whole
image like gray scaling, thresholding, cropping for
plant. In the automated system now a days, which
detecting the boundary of the image. We have divided
normally consists of computer, digital camera and
the whole area of interest into blocks and then we have
application software, various kinds of algorithms are
compared features of each block with the features of
developed in the software application. We use image
images in the database.
processing and Artificial Neural Network Technology Š 2017, IRJET
|
Impact Factor value: 5.181
|
ISO 9001:2008 Certified Journal
|
Page 1773