Plant Disease Detection using Image Processing

DOI : 10.17577/IJERTCONV8IS13036
Download Full-Text PDF Cite this Publication
Text Only Version


Plant Disease Detection using Image Processing

Darshini DN#1, Nandini k#2, Pooja BR#3,ShilpaKH#4,Ravi L S#5 5Assistant professor, Dept. of ECE, RIT, Hassan, Karnataka, India 1,2,3,4Student, Dept.of ECE, RIT, Hassan, Karnataka, India

Abstract – Theoretical Identification of the plant infection is the way to forestalling the misfortunes in the yield and amount of the farming item. The investigations of the plant infection mean investigation of outwardly discernible examples seen on the plant. Wellbeing checking and infection location on plant is exceptionally basic for supportable agribusiness. It is exceptionally hard to screen the plant infections physically. It requires enormous measure of work, mastery in the plant infections, and furthermore require the inordinate preparing time. Thus, picture preparing is utilized for the identification of plant illnesses. Illness recognition includes the means like picture procurement, picture pre-preparing, picture division, highlight extraction and arrangement. This undertaking examines the techniques utilized for the location of plant infections utilizing their leaf pictures. This task likewise examines some division and highlight extraction calculation utilized in the plant malady discovery.

Catchphrases – Image handling, Detection, Identification of plant leaf infections, Convolution neural system.


    India is quick creating nation and agribusiness is the spine for the nations improvement in the beginning phases. Because of industrialization and globalization ideas the field is confronting obstacles. On head of that the mindfulness and the need of the development should be introduced in the psyches of the more youthful age. Presently a day’s innovation assumes indispensable job in all the fields however till today we are utilizing some old strategies in farming. Recognizing plant ailment wrongly prompts enormous loss of yield, time, cash and nature of item. Distinguishing the state of plant assumes a significant job for fruitful development.


    In times past recognizable proof is done physically by the accomplished individuals however due to such a significant number of natural changes the forecast was getting intense. So we can utilize picture preparing procedures for distinguishing proof of plant malady.


Picture is caught and it will be given as the contribution to the Mat lab. Stacking of the picture will happens as per the portrayal of the specific highlights. Following stage is to upgrade contrast. The improve differentiate picture will portray the obsessive highlights of the sick leaf. The third step is to take out the division picture. This progression which is made as per the program includes the affirmation of the temporary finding of that ailment of the plant which is attempted by the assistance of differential highlights of upgrade differentiate picture as appeared in fig 1.

Fig1: Working principle


    Image Acquisition

    The underlying procedure is to gather the information from the open vault. It accepts the picture as contribution for additional preparing. We have taken most mainstream picture areas with the goal that we can take any arrangements like

    .bmp, .jpg, .gif as contribution to our procedure.

    Image Pre-processing

    As the pictures are procured from the genuine field it might contain residue, spores and water spots as clamour. The motivation behind information pre-handling is to dispense with the commotion in the picture, in order to change the pixel esteems. It improves the nature of the picture.

    • Image segmentation

    Picture division is the third step in our proposed technique. The divided pictures are grouped into various segments utilizing Otsu classifier and k-mean bunching calculation. Prior to bunching the pictures, the RGB shading model is changed into Lab shading model. The appearance of Lab shading model is to handily bunch the sectioned pictures.

    Feature extraction

    Highlight extraction is the significant part to effortlessly foresee the tainted area. Here shape and textural include extraction is finished. The shape arranged component extraction like Area, Color pivot length, capriciousness, strength and edge are determined. Thus the texture situated component extraction like differentiation, relationship, vitality, homogeneity and mean. Leaf picture is caught and prepared to decide the soundness of each plant.


    At the point when we run the program, the methodolized information sources will show as per the reason to be served, for example;

    1. Load image
    2. Enhance contrast
    3. Segment image
    4. Classification result
    5. Affected region in %
    6. Accuracy in %

    Other extra useful highlights can be refreshed by the program.

    Figure 2: Output window

    The illness bacterial curse which is brought about by the bacterial pathogen Xanthomonas campestrip will offers ascend to pathogenic appearance of a leaf which is assigned on the heap picture as appeared in fig 3.

    Figure 3: Output window

    The customized obsessive highlights of the ailing leaf is appeared on the upgrade differentiate picture as appeared in fig 4.

    Figure 4: Output image

    When we press the start camera, the camera will be activated. If we press the capture image the image will be capture. That image will be preprocessing using MATLAB. The image will be represents in a RGB form. The RGB image

    is converted into HSV, but in HSV also not correctly identifies the intensity and chrominance.

    Figure 5: Output window

    On the off chance that we press the section picture the picture will be segment. Image division is the third step in our proposed technique. The fragmented pictures are grouped into various segments utilizing Otsu classifier and k-mean bunching calculation. Prior to bunching the pictures, the RGB shading model is changed into Lab shading model. The approach of Lab shading model is to effectively group the divided pictures as appeared in fig 5.

    Disease present:

    Figure 6: Disease present

    At the point when we press the heap picture, the picture will be stacked. In the event that we press the improve differentiate, picture will be upgraded. Again in the event that we press division, that picture will be divided. Also, on the off chance that we press arrangement result, it is speak to a sickness picture with influenced locale in rate alongside precision as appeared in fig 6.

    Figure 7: Disease not present

    At the point when we press the heap picture, the picture will be stacked. On the off chance that we press the upgrade differentiate, picture will be improved. Again on the off chance that we press division, that picture will be portioned. Furthermore, on the off chance that we press characterization result, it speaks to a solid picture with precision as appeared in fig 7


    The precise discovery and grouping of the plant sickness is significant for the effective development of yield and this should be possible utilizing picture preparing. This undertaking examined different procedures to portion the infection part of the leaf. This task additionally examined some Feature extraction and arrangement methods to extricate the highlights of tainted leaf and the order of plant maladies. The employments of SVM for arrangement of illness in plants, for example, self-sorting out element map. It very well may e effectively utilized. From these strategies, we can precisely recognize and order different plant sickness utilizing picture preparing techniques. It gives the exactness of90%.


Later on, the proposed approach can be incorporated with other yet to be created, techniques for infection recognizable proof and characterization utilizing shading and surface investigation to build up a specialist framework for early soya plant foliar malady cautioning and organization, where the illness type can be distinguished by shading and surface examination and the seriousness level estimation by our proposed strategy.


  1. Wenjiang Huang, Qingsong Guan, JuhuaLuo, Jingcheng Zhang,Jinling Zhao, Dong Liang, Linsheng Huang, and DongyanZhang, “New Optimized Spectral Indices for Identifying and Monitoring Winter Wheat Diseases”, IEEE diary of selected topics in applied earth perception and remote sensing, Vol. 7,No. 6, June 2014.
  2. Dr.K.Thangadurai, K.Padmavathi, “PC Vision image Enhancement For Plant Leaves Disease Detection”, 2014 World Congress on Computing and Communication Technologies.
  3. Monica Jhuria, Ashwani Kumar, and Rushikesh Borse, “Image Processing For Smart Farming: Detection Of Disease And Fruit Grading”, Proceedings of the 2013 IEEE Second International Conference on Image Information Processing (ICIIP-2013).
  4. Zulkifli Bin Husin, Abdul Hallis Bin Abdul Aziz, Ali Yeon Bin Md Shakaff Rohani Binti S Mohamed Farook, “Feasibility Study on Plant Chili Disease Detection Using Image Processing Techniques”, 2012 Third International Conference on Intelligent Systems Modeling and Simulation.
  5. Mrunalini R. Badnakhe, Prashant R. Deshmukh, “Contaminated Leaf Analysis and Comparison by Otsu Threshold and k-Means Clustering”, International Journal of Advanced Research in Computer Science and Software Engineering, Volume 2, Issue3, March 2012.
  6. H. Al-Hiary, S. Bani-Ahmad, M. Reyalat, M. Braik and Z.ALRahamneh, “Quick and Accurate Detection and Classification of Plant Diseases”, International Journal of Computer Applications (0975 8887)Volume 17No.1, March 2011.
  7. Chunxia Zhang, Xiuqing Wang, Xudong Li, “Structure ofMonitoring and Control Plant Disease System Based on DSP&FPGA”, 2010 Second International Conference on Networks Security, Wireless Communications and Trusted Computing.
  8. A. Meunkaewjinda, P. Kumsawat, K. Attakitmongcol and A.Srikaew, “Grape leaf infection identification from shading symbolism usinghybrid wise framework”, Proceedings of ECTI-CON 2008.
  9. Santanu Phadikar and Jaya Sil, “Rice Disease Identification using Pattern Recognition”, Proceedings of eleventh International Conference on Computer and Information Technology (ICCIT2008) 25-27 December, 2008, Khulna, Bangladesh.771.

Leave a Reply