Economy highly depends onagricultural productivity.Increasing amount the growth of crops need automatic monitoring disease. Detectingdisease from the images of the rice crops is one of the interesting researchareas.
This survey presents different image processing techniques used indetection of rice crops images of different techniques but also discussconcepts of image processing applied to rice plant disease detection andclassification. Its include size of image dataset no. of diseases, segmentationtechniques, pre-processing, accuracy of classifiers etc. We utilize our surveyto design our work on detection and classification of rice plant diseases.Keywords—Image processing, classification; clustering, disease classification, diseasedetection. I.
INTRODUCTIONMost of the indian peoples income is depend onfarming. Rice is the most cultivated food all over the world. The losses ofcrops it brings indian economy decrease on agricultural field because 70% ofthe indian population depends on producing crops. Rice disease destroys 10 to15% of production in Asia.
Fungus, bacteria and viruses are responsible fordisease in the plant, so monitoring of disease on plant an important role inthe successful cultivation. Different disease that occur on rice plants areleaf blast, brown spot, sheath blight and leaf scald. This surveyfocuses on how image processing is utilized in detection of diseases in riceplants. Disease identification. The rice crop diseases are discussed in detail.
1- Leaf Blast Disease: A region varying fromsmall round, dark spot to oval spots with narrow reddish-brown margins and grayor white centre. 2- Brown Spots Disease: Round to oval shape with dark brown lesions. its occurs on leaves of the rice plant. 3- Bacterial Blight Disease: Lesions consist of elongated lesions near theleaf tip. turn white to yellow and then gray due to saprophytic fungi. 4- Sheath Blight Disease: Lesions consist of alternating wide band ofwhite, reddish-brown or brown. Fungal survival structures called sclerotia mayfrom on the leaf surface. under favourable conditions, bird nest area of deadtissue may form.
5- Sheath Rot Disease: General reddish-brown discoloration of flag leafsheath, panicles emerging poorly; white frosting of conidia on inside of leafsheath, florets discolored a uniform reddish-brown or dark brown. 6- Node Blast Disease: Clum mode turns black and gray as plants approachmaturity; nodes turn dark to blue-gray. Leaf blast Brown Leaf Spot Bacterial Blight Sheath Blight Sheath Rot Node Blast Fig. 1. Different Typesof Rice Leaf Diseases II.CONTRIBUTIONBY PREVIOUS RESEARCHERHere we describe different works thatalready done by researchers in different fields such as leaf diseaseclassification, classification and segmentation of rice leaf. so aim of thesurvey used all these control methods for a good harvesting rice plantation.Suraksha I.
S., et al,” Disease Prediction ofPaddy Crops Using Data Mining and Image Processing Techniques,” IJAREEIE,Vol. 5, Issue 5, ISSN: 2320-3765 (2011).First, the input is digital a colour image of paddydisease leaf. Then a method of mathematics morphology is used to segment theseimages. Erosion method has been used to removes small-scale details from abinary images but simultaneously reduces the size of regions of interest. Thedilation is one of the basic operations in mathematical morphology.
Thedilation operation usually uses a mesh for expanding the shapes contained inthe input image. Santanu Phadikar, et al, ” Rice diseasesclassification using feature selection and rule generation techniques,”Computers and Electronics in Agriculture (2012).Proposed a method classifying diseases ofthe rice plant. In their approach, fermi energy based region extraction methodis applied to overcome the limitation of selecting the proper threshold value.
To identify the shape of the infected region, GA is applied that bestapproximates the structure of the region. The position of infection isdetermined by partitioning the spot into different blocks and arranged as aquadtree at different lables. The binary representation of each block reducescomputational complexity reasonably. Using rough set concept features areselected by generating all reduces which minimize loss of information. From thereduced dataset a set of classification rules is derived using a novelclassification rule mining technique.
The advantage of the proposed method isthat it does not require any gain calculation of the rules and so involveslesser computational complexity. S. Phadikar, et al,”Classification of Rice Leaf Diseases Based on Morphological Changes,”International Journal of Information and Electronics Engineering (2012). Proposed a method classifythe leaf brown spot and leaf blast diseases based on the morphological changescaused by diseases. Here used SVM classifier. SVM and Bayes’ is applied bestapproximates of the region. Colour distortion of leafs occur in messclassification. Bayes’ classifier used in time complexity.
Radhika Deshmukh, et al,” Detection of Paddy Leaf Diseases,” International Conferenceon Advances in Science and Technology (2015). Proposed aprototype for detection of paddy leaf disease using K-means clustering basedsegmentation and neural network classifier to detect as well as classify thedisease affected leaf of paddy crop. The main purpose is the accurate and fastdetection of leaf disease.
They test programs such as paddy blast, brown spotpaddy spot, and normal paddy, The proposed approach is image processing based.They use a set of paddy leaf images as a dataset. Due to this experiment, thepaddy disease can be identified at the initial stage.
This eliminates thesubjectivity of traditional methods and human induced errors. Nikita Rishi, et al,” An Overview onDetection and Classification of Plant Diseases in Image Processing,” International Journal of Scientific Engineeringand Research (2015) Here discussed variousmethod and techniques; images cropping, compression, Otsu method; to detect thediseases in the heterogeneous plant. They make use of neural networksclassifiers such as BPNN, RBF, GRNN and PNNs to diagnose wheat diseases.
Cannyfilters and feature extraction applied to recognize the diseases on cotton andrice leaf. Bindushree H.B,”Detection Of Plant Leaf Disease Using Image Processing Technique,”International Journal of Technology Enhancement and Emerging EngineeringResearch (2015) Proposed the method toanalyse leaf diseases using different Image Processing technique. First, usingk-Means clustering to easily detection the disease. second, using GLCM forfeature extraction which is more efficient to extract the features.
Third, theresult using SVM used for machine learning technique used for classification.It was implemented for linear separation. Result from the SVM able to predictthe images accrately. K. Jagan Mohan, etal,” Detection and Recognition of Diseases from Paddy Plant Leaf Images,”International Journal of Computer Application (2016) Proposed Scale InvariantFeature Transform(SIFT) is used to get features from the disease affected images.Then these features are taken to recognize the image using Support VectorMachine (SVM) and K-Nearest Neighbours. This work mainly concentrates on threemain diseases of paddy plant namely brown spot, leaf blast and bacterialblight.
It is useful to farmers. Experimental result showed that SVM and K-NNis capable of predicting disease accuracy of 91.10% in SVM and 93.
33% in K-NN. U.Gayathri et al,” Effective Disease Detection for Plants”International Journal of Advanced Research Methodology in Engineering & Technology(2017). This survey concentrate on the image processingtechniques used to enhance the quality of the image and neural networktechnique to classify the banana disease. The methodology involves imageacquisition, pre-processing and segmentation, analysis and classification ofthe disease. All the banana sample will be passing through the RGB calculationbefore it proceed to the binary conversion. If the range of normal Banana RGB,then it is automatically, classify as type 4 which is Normal. Then, all thesegmented Banana disease sample will be convert into the binary data forclassification training and testing.
Consequently, by employing the neuralnetwork technique, the Banana diseases are recognized about 92.5 percentaccuracy rates. This prototype has a very great potential to be furtherimproved in the future to detect the plant related issues in the field ofagricultural analysis. This survey gradually decrease the effects of disease inplants and the plants can be easily monitored via camera at less expenditure. Hence,more plants were saved by the advent of the project 12.