Detection of Bacterial Blight of Punica Granatum L using Digital Image Processing Techniques with Real time Alerting System

DOI : 10.17577/IJERTCONV3IS05005

Download Full-Text PDF Cite this Publication

Text Only Version

Detection of Bacterial Blight of Punica Granatum L using Digital Image Processing Techniques with Real time Alerting System

1Amal Jose, 2Akash P, 3Benaka Santhosh. S

1Dept.of ECE,CIT,Ponnampet,Karnataka,

2 Dept.of ECE,CIT,Ponnampet,Karnataka,

3Assistant Professor, Dept.of ECE,CIT,Ponnampet,Karnataka,

Abstract Pomegranate (punica granatum L) occupies an important position in the agricultural economies of south Indian farmers. It is the major economic crop of south India. Pomegranate agriculture is posed to many diseases since the plants are very sensitive, since the expenditure of pomegranate agriculture is too high, the diseases must be controlled at their early stage. The aim of the paper is to detect the diseases at their early stage. Nowadays the techniques of sophisticated digital cameras and digital image processing are extensively applied to agricultural science and it has great perspective especially in the plant protection field, which ultimately leads to crop management. The paper proposes software for disease detection on the infected images of pomegranate leaf. Images of the infected leaf are captured by digital camera and processed using image growing, segmentation techniques to detect infected parts of the particular plants. Then detected part is been processed for further feature extraction which gives general idea about parts. The paper proposes automatic detection and estimating the area of diseased leaves of the bacterial blight of pomegranate at early stages with real time intimation to farmer.

Keywords Bacterial blight, digital camera, feature extraction, segmentation, GSM


    Pomegranate occupies an important position in the agricultural economies of south Indian farmers. It is the major economic crop of south India. Pomegranate agriculture is posed to many diseases since the plants are very sensitive, since

    the expenditure of pomegranate agriculture is too high,The losses caused to farmers is too high because of these diseases.

    One of the major diseases infecting the pomegranate plantation is Bacterial blight of pomegranate. It was first reported in India in Delhi in 1952 and was of minor economic importance until 1998.Presently the disease occurs widely and out breaks has been recorded in all major pomegranate growing states including Maharashtra, Karnataka and Andhrapradesh. Losses caused by bacterial blight were recorded in Hanumangarh district of Rajasthan in 2009.Until recently, the disease was prevalent only in India. But it was also reported from South Africa in 2010.Bacterial blight of pomegranate affects leaves, twigs and fruits. Infected fruits and twigs are potential sources of primary inoculums. The secondary spread of bacterium is mainly through rain and spray splaster, irrigation water,humans and insect vectors.Entry is through wounds and natural openings.The first water soaked lesions develop within 2-3 days and appear as dark red spots.Bacterial cells are capable of surviving in soil for more than 120 days and also survive in fallen leaves during the off season.High temperatures and low humidity or both favor disease development. In this paper the arm is to detect the disease as early as possible. First the plants are regularly observed. Diseased images acquired using cameras on scanners. Then the acquired image has to be processed to interpret the image contents by digital image processing methods. The aim of paper is the interpretation of image for disease detection.


    Digital image processing is a technique to enhance raw images received from cameras/sensors. Image processing techniques are becoming popular due to easy availability of powerful personal computer, large size memory devices, graphic software etc.

    Digital image processing refers to processing digital images by means of a digital computer. The digital image is composed of a finite number of elements, each of which has a particular location and value.

    1. Image Acquisition

      Image acquisition is the first step in digital image processing. Acquisition could be as simple as being given an image that is

      already in digital form. Generally, the image acquisition stage involves preprocessing, such as scaling.

      The images can be captured using a regular digital camera with minimum six megapixels of resolution. For better quality, maintaining an equal distance, equal angle and equal illumination to the object with uniform background. All the images should be saved in the same format such as JPEG, TIF, BMP, PNG etc. For this study, Bacterial blight of Pomegranate was chosen because this requires early detection and treatment to prevent durable infection. Samples were manually cut and scanned directly in the green house as shown in figure 1.Once the image is acquired and scanned the next step is to implement image processing technique in order to get the information about the pest.

      The DIP techniques involve treating the image as a two dimensional signal and applying standard signal processing technique to it. In any image processing application, the important input is image it is basically a matrix (array) of square pixels for the purpose of early automatic detection of pests on scanned leaves the algorithm has to be followed. It is shown in figure 2.


      Analysis of image

      Extraction of

      Figure 1.Leaf of Punica Granatum L

      Region Feature Extracti on


      Local Feature Extracti on

      Local Feature Extracti on

      Image segmen atation


      Image Subtra ction

      Image Subtra ction

      Image Filtering

      Color Feature Extracti on









      Color Feature Extracti on









      Figure.2 Image processing operation work flow Diagram

      Region Based Segmenta tion

      Edge Based Segmentati on

      Gaussian Filtering

      Laplacian Filtering

      Watershed Segmentation

      Mean shift Segmentation

      Watershed Segmentation

      Mean shift Segmentation

      Object extraction is followed by feature extraction.Object extraction itself decomposes into a sequence background subtraction, the filtering and finally segmentation. Since background subtraction appears on the top and corresponds to a concrete program to execute, the system involves it. This program automatically extracts a leaf from its background image .The second sub operator, filtering may be performed in several different ways (Gaussian, Linear, LP, HP, Median or Laplacian filtering

      ).The corresponding denoising program will be executed. The next operation i.e. image segmentation operator, also corresponds to a choice between two alternative sub operators. Region based and edge based. Similarly, once the objects are extracted, the second step of image analysis is feature extraction .Computes the attributes corresponding to each region according to the domain feature concept (ex: color, shape and size descriptors) and to the operator graph. The runs up to the last programming the decomposition in our example it appears to be shape feature extractors .Finally through this we get the information about pests and its features which is useful data for its prventive measures that has to be done.

    2. Image Preprocessing

      Image pre-processing creates an enhanced image that is more useful or pleasing to a human observer pre processing uses various techniques like image resize, filtering, segmentation, morphological operations etc…In most of the image processing applications the initial captured image are resized to a fixed resolution to utilize the storage capacity or to reduce the computational burden .Since images may be captured from the fields it will be unavoidable .That some due drops, Insects excretes and dust might appear on the captured image . In DIP technique we are applying these are treated as image noises. They must be removed or weakened before any further image analysis Filter like Gaussian , Median, Linear, Laplacian filters etc can be used to remove the image noise once the image has been enhanced ,the next process is to extract region of interest in the image i.e. , diseased portion of image .This can be achieved through image segmentation.

    3. Image post processing

      Once the image has been enhanced and segmented, there may be some stabs, empty holes etc remained in the images. Hence to remove these noises morphological operations, region filling can be applied. Further, the interested part can be extracted and its features can be analysed.For an image, a feature can be defined as the interest part in an image. The name features is often used in the pattern recognition literature to denote a descriptor. The desirable property for a feature detector is repeatability i.e. Whether or not the same feature will be detected in different images. Features play a fundamental role in classification. In image processing image features usually include color, shape and texture features

    4. Disease classification and Grading

    Once the features are extracted, next step is to find to which disease. Class the query image belongs to and in which stage the disease is. Once the disease stage is identified, appropriate treatment advisory can be provided by seeking the help from agricultural experts so that the disease can be prevented from further spreading. For this purpose we can make use of the different machine learning techniques


  1. Image subtraction

    The image subtraction i.e. background subtraction is a type of image segmentation whose goal is to separate the parts image that are invariant over time (background) from the objects. That are moving or changing (foreground). The simplest technique use frame differencing and more advanced techniques require using stastical methods. As an example, we can consider a moving leave of a tree would be considered foreground using simple frame differencing but with a proper statistical method it can be considered background as the leave is always these, moving periodically

  2. Image Filtering

    A digital image has to be & filtered for the purpose of smoothing, sharpening, removing noise, edge detection. An image filter is denoted by a kernel, which is a small array applied to each pixel and its neighbors within an image. The process used to apply filters to an image is known as convolution. The filtering process of a digital image is carried out in spatial domain. In linear spatial filtering the response of a filter is the sum of the products of filtering co-efficient and the corresponding image pixels. Within the spatial domain the first part of convolution process multiplies the elements of the kernel by matching pixel values. When the kernel is centered over a pixel. The elements of the resulting array of the same size as that of the kernel are averaged, and the original pixel value is replaced with this results. The convolution function performs this convolution process for an entire image

  3. Image Segmentation

    Image segmentation includes the division or separation of the image into regions of similar attribute. The basic attribute for segmentation is image amplitude luminance for a monochrome image and color components for a color image. Image edges and texture are also useful attributes for segmentation. The result of image segmentation is a set of regions that collectively cover the entire image, on a set of contour extracted from the image

    Image segmentation does not involve classifying each segment of image. The segmentation only subdivides an image. It does not attempt to recognize the individual segments or their relationships to one another. There is no theory of image segmentation. As a consequence, no signal standard method of image segmentation has emerged. Rather there are a collection of adhoc methods that have received some degree of popularity. Because the methods are adhoc, it would be useful to have some means of accessing their performance. The goal of segmentation is to simplify and to change the representation of an image into something that is more meaningful and easier to analyze. Image segmentation is typically used to locate objects and boundaries in images.

    1. Watershed Segmentation

      A watershed is a basin-like landform defined by highpoints and ridgelines that descend into lower elevations and stream valleys. Watershed of a relief corresponds to the limits of the adjacent catchment basins of the drops of water. In graphs, watershed lines may be defined on the nodes, on the edges, or hybrid lines on both nodes and edges. Watershed algorithm is used in image processing primarily for segmentation Purposes.

    2. Mean shift Segmentation

The mean shift segmentation is a local homogenization technique that is very useful for damping, shading or tonality differences in localized objects. It is a non- parametric feature-space analysis technique for locating the maxima of a density function, also called mode-seeking algorithm.

D) Feature Extraction

A feature of image is a boundary region of points that satisfy a set of pre defined criteria. The criteria can be based on any quantities such as shape, time, similarity, orientation and spatial distribution. Transforming the input data into the set of features still describing the data with sufficient accuracy. In pattern recognition and image processing feature extraction is a special form of dimensionality reduction. It is used when the input data to an algorithm is too large to be processed and it is suspected to be redundant (much data, but not much information).


The percentage of infection in each leaf can be calculated by using image analysis.The given output is in the form of pixels.So the infected area in percentage can be calculated by using simple formula


In the developing countries like India, Agriculture plays an important role in the life of people. The disease caused to plants if they are detected at early stages only, then lot of loss to the farmer will be prevented. Along with the above result, in future we wish to add up the extra things like a protocol to decide the intensity of infection, depending on the decision made, an automatic decision making algorithm to spray the required pesticide for the plantation including the on time information to the farmer through SMS.


  1. Sanjeev S Sannakk,Vijay S Rajpurchit,V B Nargund,Arun kumar R,Prema S Yallur,Leaf Disease Grading by Machine Vision and Fuzzy Logic,Int J.Comp,Tech,Appl.Vol 2(5),1709- 1716,2011.

  2. Laakronen,J Koskela,M.oja.E,Self organizing maps for content based image retrieval,Internatinal Joint Conference on Neural Networks,Vol.4.PP 2470-2473,1999

  3. J-K Park, E-J Hwang, and Y. Nam, A vention based leaf image classification scheme, Alliance of Information and Referral Systems, 2006, pp. 416-428.

  4. A. Kadir, L.E. Nugroho, A. Susanto, and P.I. Santosa, A comparative experiment of several shape methods in recognizing plants, International Journal of Computer Science & Information Technology (IJCSIT), vol. 3, no. 3, 2011, pp. 256-263

  5. N. Valliammal, and S. N. Geethalakshmi, Hybrid image segmentation algorithm for leaf recognition and characterization, International Conference on Process Automation, Control and Computing (PACC), 2011, pp. 1-6.

  6. G. Cerutti, L. Tougne, J. Mille, A. Vacavant, and D. Coquin, Guiding active contours for tree leaf segmentation and identification, Cross-Language Evaluation Forum (CLEF), Amsterdam, Netherlands, 2011.

  7. A. J. Perez, F. Lopez, J. V. Benlloch, and S. Christensen, Color and shape analysis techniques for weed detection in cereal fields, Computers and Electronics in Agriculture, vol. 25, 2000, pp. 197- 212.

  8. J. M. Timmermans, and A. A. Hulzebosch, Computer vison system for on-line sorting of pot plants using an artificial neural network classifier, Computers and Electronics in Agriculture, vol. 15, 1996, pp. 41-55.

  9. M.Seetha, I.V.muralikrishna, B.L. Deekshatulu, B.L.malleswari,

    Percentage of infection =


    Nagaratna, P.Hegde a Artificial neural networks and other methods of image classification, Journal of Theoretical and Applied Information Technology, © 2005 – 2008 JATIT. All rights

    From the reult we can calculate the total infection on leaf which in turn gives us information about the intensity of infection on leaf.

    Calculating the area of infection in each leaf gives the idea about in which stage the disease ins in whether it is in mature stage or is in pre mature stage.The given output is in the form of pixels.

    Table.1 Result of Bacterial Blight of Pomogranate

    reserved. [] Heymans, B.C., Onema, J.P., Kuti, J.O.: A neural network for opuntia leaf-form recognition.In: Proceedings of IEEE International Joint Conference on Neural Networks (1991)

  10. Ye, Y., Chen, C., Li, C.T., Fu, H., Chi, Z.: A computerized plant species recognition system. In: Proceedings of 2004 International Symposium on Intelligent Multimedia, Video and Speech Processing, Hong Kong (October 2004)

  11. T. Brendel, J. Schwanke, P. Jensch, and R. Megnet, Knowledgebased object recognition for different morphological classes of plants, Proceedings of SPIE, vol. 2345, 1995.

  12. Alan, K.K., Philip, A., Fay, et al.: Rainfall Variability, Carbon Cycling, and Plant Species Diversity.

Samle of leaf


Area of infection

Percentage of infection























Leave a Reply