Docstoc

A NOVEL NEURAL NETWORK BASED APPROACH FOR THE CLASSIFICATION OF BETEL LEAVES

Document Sample
A NOVEL NEURAL NETWORK BASED APPROACH FOR THE CLASSIFICATION OF BETEL LEAVES Powered By Docstoc
					    International Journal of Emerging Trends & Technology in Computer Science (IJETTCS)
       Web Site: www.ijettcs.org Email: editor@ijettcs.org, editorijettcs@gmail.com
Volume 1, Issue 2, July – August 2012                                          ISSN 2278-6856




         A NOVEL NEURAL NETWORK BASED
        APPROACH FOR THE CLASSIFICATION
                OF BETEL LEAVES
                                                  SANDEEP KUMAR.E
                                            Department of Telecommunication Engineering
                                            JNN College of Engineering, Shimoga-577204
                                                       Karnataka State, India

                                                                  leaves are beneficial in the treatment of nervous pains,
Abstract: Betel leaves has its dominant place in Indian           nervous exhaustion and debility. Betel leaf has analgesic
commercial market. This is because of its high medicinal          and cooling properties, instantly relieves constipation.
values as per the history of Ayurvedic medicinal system, also     Local application of the leaves is effective in treating sore
habit of people chewing betel leaves with areca nut. In           throat. The application of leaves smeared with oil is said
Karnataka state of India, there are two species of betel leaves   to promote secretion of milk when applied on the breasts
which are commercially recognized by local names Mysore           during lactation and many more. Hence there is a sudden
betel leaf and Ambadi betel leaf. These leaves are separated
                                                                  requirement to grow these plants and improve them
and packed manually. In this context, here is a new approach
                                                                  commercially. This research work is a step towards
using machine vision and machine intelligence blend; for
classification of these two species of betel leaves leading to
                                                                  automatic separation of the betel leaf species such as
less human effort. This paper uses image processing for           Mysore leaf and Ambadi leaf for further processing and
machine vision where the leaf features such as width, gray        packing.
histogram and color histogram is extracted. Machine
intelligence part uses Neural Networks with back propagation      The work consists of two parts: machine vision and
algorithm.                                                        machine intelligence. The machine vision part uses image
Keywords: Ayurvedic medicinal system, Mysore Betel leaf,          processing where the features such as width, color
Ambadi betel leaf, machine vision, machine intelligence, neural   histogram and gray histogram are extracted. These
network.                                                          parameters are fed to the machine intelligence part which
                                                                  uses artificial neural networks as the tool for the
1. INTRODUCTION                                                   classification.
The scientific name of betel leaf is Piper betle, belonging
to the family Piperaceae. These leaves are well known as
Vellya dele in Karnataka and Paan in other places of              2. RELATED WORK
India. The betel plant is a slender, aromatic creeper,            This is a new approach for the classification of leaves
rooting at the nodes. The branches of the plant are               belonging to the same botanical family. This in fact was a
swollen at the nodes. The plant has alternate, heart-             difficult work. Researchers have proposed methods to find
shaped, smooth, shining and long-stalked leaves, with             the area of betel leaves [4]. Many others have proposed
pointed apex. It has five to seven ribs arising from the          methods to classify leaves based on leaf features such as
base; minute flowers and one-seeded spherical small               shape, texture as such [3] [8] [9] [11]. But specifically for
berries. The use of betel leaf can be traced as far back as       betel leaf species has not been done. Hence this a new
two thousand years. It is described in the most ancient           approach for the classification of betel leaves using neural
historic book of Sri Lanka, Mahavasma, written in Pali.           networks.
Betel is a native of central and eastern Malaysia. It spread
at a very early date throughout tropical Asia and later to        3. PROPOSED METHODOLOGY
Madagascar and East Africa. In India, it is widely
cultivated in Tamil Nadu, Madhya Pradesh, West Bengal,            This section involves the various steps and techniques
Orissa, Maharashtra and Uttar Pradesh. Offering betel             used in the process of separation of the leaves.
morsel (pan-supari) to guests in Indian subcontinent is a
common courtesy.                                                     3.1 Image acquisition
                                                                  The image of betel leaves where taken from a 5 mega
 Betel leaves has many medicinal properties, some of              pixel camera, keeping 15 cms distance in between leaf
them are: It is useful in arresting secretion or bleeding         and camera. All the images where taken from top view
and is an aphrodisiac. Its leaf is used in several common         with white background.
household remedies, helps in easing urination. Betel
Volume 1, Issue 2 July-August 2012                                                                                   Page 10
   International Journal of Emerging Trends & Technology in Computer Science (IJETTCS)
       Web Site: www.ijettcs.org Email: editor@ijettcs.org, editorijettcs@gmail.com
Volume 1, Issue 2, July – August 2012                                          ISSN 2278-6856


  3.2 Image samples
In this work, seventy image samples were taken from
Mysore betel leaf species. From which, fifty were used for
the training neural network and nineteen samples were
taken as the test samples. Seventy image samples were
taken from Ambadi betel leaf species. From which, fifty
were used for training neural network and eighteen
samples were taken as test samples.
The leaf images of both species are shown in Figure-1.

   3.3 Devised Methodology
In this work two species of betel leaves: Mysore betel leaf
and Ambadi betel leaf is considered. The image is
captured using a 5MP camera. Later in the pre-processing
stage it is resized to 200 X 300 resolutions, which is as
per our requirement. Also image is processed for the
noise removal. Then the first information i.e., color
histogram is extracted. Then the color image is converted
to its equivalent grayscale image and gray histogram of
the leaf is calculated. This is the second parameter useful
to notify the leaf intensity. This grayscale image is
converted to its equivalent binary image and width of the
leaf is calculated. This is the third parameter. Once we
extract all these three parameters in the feature extraction
                                                                           Figure 2 System Block diagram
stage, we feed these parameters as the training parameters
for the classifier, which is an Artificial Neural Network
using Back propagation algorithm.
                                                               4. FEATURE EXTRACTION
The system block diagram is shown in Figure-2.                 This section deals with the three important features which
                                                               are extracted from the betel leaves which are used for
                                                               their classification.

                                                                 4.1 Width of leaf




                                                                              Figure 3 Width of the leaf

                                                               Width is one of the important parameters which give a
                                                               fine differentiation between the two leaf species. As one
Figure 1 (a) Ambadi betel leaf. (b) Mysore betel leaf.         can notice in Figure 1, Mysore betel leaf is broader and
                                                               width of the leaf is more compared with the Ambadi leaf.
                                                               Hence this section makes an attempt to extract this
                                                               information.




Volume 1, Issue 2 July-August 2012                                                                              Page 11
   International Journal of Emerging Trends & Technology in Computer Science (IJETTCS)
       Web Site: www.ijettcs.org Email: editor@ijettcs.org, editorijettcs@gmail.com
Volume 1, Issue 2, July – August 2012                                          ISSN 2278-6856


  Algorithm:                                                       Step 4: Add the three histograms values and find the
                                                                   average to get the overall image color histogram.
    Step 1: Start                                                  Step 5: Stop
    Step 2: Read the image
    Step 3: Convert the color image to black and white            4.3 Gray Histogram
    image                                                      Referring to Figure-4 one can notice that the there is
    Step 4: Starting from first row first pixel i.e., (1, 1)   minute intensity difference in the two images. i.e.,
    move column wise until we get a black pixel or till        Mysore betel leaf is darker compared to the Ambadi betel
    column end is reached.                                     leaf. Hence this section extracts this information from the
    Step 5: If black pixel is encountered, load the pixel      leaf. The gray histograms extracted can be seen referring
    position to variables say (h, b). This is co-ordinate 1;   to Figure-5 and Figure-8.
    Go to Step 7 else Go to Step 6.
    Step 6: If black pixel not encountered and we have           Algorithm:
    reached to the column end, increment row count and
    move to the next row and start moving column wise              Step 1: Start
    and go to Step 5.                                              Step 2: Acquire the leaf image.
    Step 7: Start from last row first pixel i.e., (200, 1)         Step 3: Convert the RGB image to its equivalent
    move column wise until we get a black pixel or till            grayscale image.
    row end is reached.                                            Step 4: Calculate the histogram value of this image.
    Step 8: If black pixel is encountered, load the pixel          Step 5: Stop
    position to variables say (h1, b1). This is co-
    ordinate; Go to Step 10 else Go to Step 9.
    Step 9: If black pixel not encountered and we have
    reached to the column end, increment row count and
    move to the next row and start moving column wise
    and go to Step 8.
    Step 10: Calculate the Euclidean distance. Let this
    be variable ‘Width’.
    Step 11: Stop.
    Euclidean distance (in this case width) is calculated
    by the formula:



   Euclidean distance (width) =
                          [(h1-h) ^2 + (b1-b) ^2] ^
                   0.5
Where,

(h, b)  Co-ordinate 1.
(h1, b1)  Co-ordinate 2.

   4.2 Color Histogram
Referring to Figure-1, the leaf color of Ambadi species is
light green and that of the Mysore species is dark green.
Hence this section of the work extracts this color variation
of both the leaves by calculating the entire image color        Figure 4 Gray scale image of (a) Mysore Betel leaf (b)
histogram. The color histogram extracted can be seen                             Ambadi betel leaf.
referring to Figure-6, Figure-7, Figure-9 and Figure-10.
                                                               Color histogram and gray histogram of the image cannot
  Algorithm:                                                   be fed directly to the neural network. Hence the histogram
                                                               is reduced to a single value by applying Euclidean
    Step 1: Start                                              distance formula. For the gray scale image, histogram is
    Step 2: Acquire the leaf image                             calculated and then normalized; the distance of the
    Step 3: Calculate the green histogram value, blue          histogram from the origin of histogram plot is calculated
    histogram value and red histogram value separately         this is fed to the neural network. Since the gray histogram
    of the image.                                              of Mysore betel leaf is nearer to the origin compared to
                                                               the Ambadi betel leaf. Similarly we apply for the color
Volume 1, Issue 2 July-August 2012                                                                              Page 12
   International Journal of Emerging Trends & Technology in Computer Science (IJETTCS)
       Web Site: www.ijettcs.org Email: editor@ijettcs.org, editorijettcs@gmail.com
Volume 1, Issue 2, July – August 2012                                          ISSN 2278-6856


image also. In this case, we calculate the histogram of the
red plane, blue plane and green plane separately then
normalize it, then distance of the histogram is calculated
separately for then three planes and average of the three
is calculated. This is fed as color histogram value for the
neural networks. One can notice that even the color
histogram of Mysore betel leaf is nearer to the origin of
the plot than the Ambadi betel leaf [3].
Hence the width, color histogram and the gray histogram
are the three important features that differentiates these
two betel leaf species and these parameters are used to
train the classifier.

  4.4 System algorithm

    Step 1: Start
    Step 2: Read leaf image
    Step 3: Calculate color histogram
    Step 4: Convert color image to gray scale image
    Step 5: Calculate gray histogram
    Step 6: Convert the gray image to black and white
    image
    Step 7: Calculate ‘width’ of the leaf
    Step 8: Input these values to the trained neural
    network
    Step 9: Compare the output, whether positive or
    negative
    Step 10: If positive display ‘Ambadi Betel leaf’, if
    negative display ‘Mysore Betel leaf’.
    Step 11: Repeat Step 2 to Step 10 for all test images
    Step 12: Stop




                                                              Figure 7 Red and Blue Histograms of Ambadi Betel Leaf




Volume 1, Issue 2 July-August 2012                                                                         Page 13
   International Journal of Emerging Trends & Technology in Computer Science (IJETTCS)
       Web Site: www.ijettcs.org Email: editor@ijettcs.org, editorijettcs@gmail.com
Volume 1, Issue 2, July – August 2012                                          ISSN 2278-6856


                                                    5. CLASSIFICATION OF LEAVES
                                                    As a classifier this rejuvenated work uses Artificial
                                                    Neural Networks. These Networks can be likened to
                                                    collections of identical mathematical models that emulate
                                                    some of the observed properties of biological nervous
                                                    systems and draw on the analogies of adaptive biological
                                                    learning. The key element of an Artificial Neural
                                                    Network is its structure. It is composed of a number of
                                                    interconnected processing elements tied together with
                                                    weighted connections, which take inspiration from
                                                    biological neurons. Learning like in a biological system
                                                    takes place through training, or exposure to a set of input
                                                    and output data where the training algorithm adjusts the
                                                    weights iteratively. Artificial Neural Networks are good
                                                    pattern recognition engines and robust classifiers, with
                                                    the ability to make decisions about imprecise input data.
                                                    What is needed is a set of examples that is representative
                                                    of all the variations in those two types of leaf species.

                                                    This neural network is trained using Back Propagation
                                                    Algorithm. Figure-11 shows the block diagram
                                                    representation of an Artificial Neural Network.




                                                       Back Propagation Algorithm:
                                                    Back propagation was created by generalizing the
                                                    Widrow-Hoff learning rule to multiple-layer networks and
                                                    nonlinear differentiable transfer functions. Input vectors
                                                    and the corresponding target vectors are used to train a
                                                    network until it can approximate a function, associate
                                                    input vectors with specific output vectors, or classify input
                                                    vectors in an appropriate way as defined by you.
                                                    Networks with biases, a sigmoid layer, and a linear output
                                                    layer are capable of approximating any function with a
                                                    finite number of discontinuities. The term back
                                                    propagation refers to the manner in which the gradient is
                                                    computed for nonlinear multilayer networks. There are a
                                                    number of variations on the basic algorithm that are
                                                    based on other standard optimization techniques, such as
                                                    conjugate gradient , Newton methods so on.

                                                    Properly trained back propagation networks tend to give
                                                    reasonable answers when presented with inputs that they
                                                    have never seen. Typically, a new input leads to an output
                                                    similar to the correct output for input vectors used in
                                                    training that are similar to the new input being presented.
Figure 9: Red and Blue Histograms of Mysore Betel   This generalization property makes it possible to train a
Leaf                                                network on a representative set of input/target pairs and


Volume 1, Issue 2 July-August 2012                                                                     Page 14
   International Journal of Emerging Trends & Technology in Computer Science (IJETTCS)
       Web Site: www.ijettcs.org Email: editor@ijettcs.org, editorijettcs@gmail.com
Volume 1, Issue 2, July – August 2012                                          ISSN 2278-6856


get good results without training the network on all                                  Test image 13          -0.6100
possible input/output pairs.                                                          Test image 14          0.5810
                                                                Mysore betel leaf     Test image 15          -0.7809
In the following scenario there is one input vector for the                           Test image 16          0.6252
feed forward network with three elements:                                             Test image 17          -0.9265
p= [width; color; gray];                                                              Test image 18          -0.9246
Where,                                                                                Test image 19          -0.2687
           width Width of leaf parameter.
            color Color histogram parameter.                 Out of 19 leaf images 14 were correctly identified as
            gray Gray histogram parameter.                   Mysore betel leaf. Hence the accuracy is 74%.

Inputs ‘color’ and ‘gray’ are in the range [0 1] and width         Table 2: Results obtained for Ambadi betel leaf
is in the range [1 200].Hidden layer has 25 neurons and
the output layer has a single neuron. The transfer function     Leaf species          Input test image       Output
in the hidden layer and the output layer are ‘tan-sigmoid’.                                                  obtained
The training function is ‘trainscg’- Scaled conjugate
                                                                                      Test image 1           0.1540
gradient algorithm.
                                                                                      Test image 2           0.2712
Training function, Scaled conjugate gradient algorithm                                Test image 3           0.7817
was developed by Moller and is designed to avoid the                                  Test image 4           0.5611
time-consuming line search. The basic idea is to combine                              Test image 5           0.5731
the model-trust region approach with the conjugate                                    Test image 6           0.9573
gradient approach.                                                                    Test image 7           0.4842
                                                                                      Test image 8           0.9667
                                                                Ambadi betel leaf     Test image 9           0.6060
The neural network is trained for +1 for Mysore betel leaf
species and -1 for Ambadi betel leaf species.                                         Test image 10          0.3399
                                                                                      Test image 11          -0.8439
                                                                                      Test image 12          -0.9913
6. RESULTS AND DISSCUSSIONS                                                           Test image 13          0.2946
                                                                                      Test image 14          -0.6981
50 samples of Mysore leaf species and 50 samples of                                   Test image 15          -0.9013
Ambadi leaf species were used to train the neural
                                                                                      Test image 16          0.8443
network. Epochs were set for 500. Learning rate was set
                                                                                      Test image 17          -0.8879
for 0.5. The neural network was trained to -1 for Mysore
                                                                                      Test image 18          0.6393
betel leaf species and +1 for Ambadi leaf species. 19
samples of Mysore leaf images were given as the test
                                                              Out of 18 leaf images 13 were correctly identified as
images and 18 samples of Ambadi leaf images were given
                                                              Ambadi betel leaf. Hence the accuracy is 72%. Table 1
as the test images. The output of the neural network is
                                                              and Table 2 depict the results obtained.
tresholded for negative value for Mysore betel leaf and
positive value for Ambadi betel leaf. The algorithm was
                                                              Here classification of leaves was a challenge, since they
coded and tested using MATLAB 2010.
                                                              belong to the same family Piperaceae. The border pattern,
                                                              the length, the vein pattern of both the leaf species
     Table 1: Results obtained for Mysore betel leaf
                                                              remains same. The only difference observed was with
                                                              respect to the width and the color. That too both the
  Leaf species         Input test image      Output
                                                              leaves are of green color. But one is light green and the
                                             obtained
                                                              other is little darker than the other. Also the difference in
                       Test image 1          -0.1080
                                                              width, color and gray histograms values was also very
                       Test image 2          -0.5167          less. With these limited ranged inputs successfully the
                       Test image 3          -0.5626          network has been trained for the accuracy of 74% and
                       Test image 4          0.3779           72%.
  Mysore betel leaf    Test image 5          0.1222           With still more different sets of samples for training, one
                       Test image 6          -0.4745          can get better accuracy.
                       Test image 7          -0.8955
                       Test image 8          -0.2863          7. CONCLUSION
                       Test image 9          -0.7414
                       Test image 10         0.5460           This is an innovative approach ever done for the
                                                              classification of betel leaves. The methodology uses a
                       Test image 11         -0.5958
                                                              blend of machine vision and machine intelligence for
                       Test image 12         -0.9530
Volume 1, Issue 2 July-August 2012                                                                               Page 15
   International Journal of Emerging Trends & Technology in Computer Science (IJETTCS)
       Web Site: www.ijettcs.org Email: editor@ijettcs.org, editorijettcs@gmail.com
Volume 1, Issue 2, July – August 2012                                          ISSN 2278-6856


precision agriculture. In machine vision part, image          [6] Chris Solomon and Toby Breckon, “Fundamentals of
processing was used where the leaf features were              Digital Image Processing”- A practical approach with
extracted. In machine intelligence part neural network        examples in Matlab.
working on back propagation algorithm was used.
         Hence the work was successful with the accuracy      [7] R.C. Gonzalez, R. E. Woods and S. L. Eddins,
of 74% and 72%. This is a small contribution towards          “Digital Image Processing”, Prentice Hall, 2004.
agriculture and growing this medicinally valued precious
plant species, which is moving towards extinction because     [8] Basavaraj        Anami,               J.D.     Pujari,
of negligence now a days.                                     Rajesh.Yakkundimath “Identification and Classification
                                                              of Normal an Affected Agriculture/horticulture Produce
                                                              Based on Combined Color and Texture Feature
             ACKOWLEDGEMENT                                   Extraction”, International Journal of Computer
                                                              Applications in Engineering Sciences, [VOL I, ISSUE III,
Author likes to thank his family for their valuable support
                                                              SEPTEMBER 2011]
through out his work. He likes to thank Dr. S.V Sathya
                                                              [VOL I, ISSUE III, SEPTEMBER 2011]
Naryana, Professor, Dept. of Electronics and
                                                              [9] Stephen Gang Wu, Forrest Sheng Bao, Eric You Xu,
Communication       Engineering,     JNN College of
                                                              Yu-Xuan Wang, Yi-Fan Chang and Qiao-Liang
Engineering, for his valuable guidelines in making this
                                                              Xiang4,“A Leaf Recognition Algorithm for Plant
paper a success. Author likes to thank Mr. Sashikiran S,
                                                              Classification Using Probabilistic Neural Network”
Asst.    Professor,    Dept.    of     Telecommunication
Engineering, JNN college of Engineering. Author is
                                                              [10] Abdul Kadir, Lukito Edi Nugroho, Adhi Susanto,
grateful to Mrs. Veena K.N, Asst. Professor, Dept. of
                                                              Paulus Insap Santosa “Leaf Classification Using Shape,
Telecommunication Engineering, JNN College of
                                                              Color, and Texture Features” International Journal of
Engineering and Mr. Pawan Kumar M.P, Lecturer, Dept.
                                                              Computer Trends and Technology- July to Aug Issue
of Information Science & Engineering JNN college of
                                                              2011.
Engineering, for their support and advice through out this
work.
                                                              [11] P. Pattanasethanon and B. Attachoo, “Thai botanical
Also Author is thankful to the principal, JNN college of
                                                              herbs and its characteristics: Using artificial neural
Engineering for his support and Co-operation through out
                                                              network” African Journal of Agricultural Research Vol.
this work.
                                                              7(2), pp. 344-351, 12 January, 2012.

                                                              [12] MATLAB Neural Network toolbox.
REFERENCES
                                                              AUTHOR
[1] Dr. R.C.Prajapati, I.F.S, APCCF & Member
Secretary, KARNATAKA BIODIVERSITY BOARD,                                          Sandeep Kumar. E completed his Bachelor
“BIODIVERSITY OF KARNATAKA”-At a Glance.                                          of Engineering in Telecommunication
                                                                                  Engineering from JNN college of
[2] P. Guha, “Betel Leaf: The Neglected Green Gold of                             Engineering (Affiliated to Vishveshvaraya
India”, J. Hum. Ecol., 19(2): 87-93 (2006).                                       Technological     University),     Shimoga,
                                                                                  Karnataka state, India. Presently he is
                                                                                  working as Lecturer in Department of
[3] Sandeep Kumar.E, “LEAF COLOR, AREA AND
                                                                                  Telecommunication       Engineering,   JNN
EDGE FEATURES BASED APPROACH FOR                              college of Engineering, Shimoga, Karnataka state, India with 3
IDENTIFICATION        OF     INDIAN       MEDICINAL           years of teaching experience. He has handled subjects: C, C++,
PLANTS”, Indian Journal of Computer Science and               Data Structures, Optical Communication & networking. His
Engineering (IJCSE), Vol. 3 No.3 Jun-Jul 2012.                area of Interest being machine vision and machine intelligence,
                                                              towards precision agriculture. Published 2 papers on machine
[4] Sanjay B Patil and Dr. Shrikant K Bodhe, “ Betel          vision in national journal/ conference. Received many merit
Leaf     Measurement    Using     Image     Processing”,      awards and scholarships during his carrier.
International Journal on Computer Science and
engineering (IJSCE), Vol. 3 No. 7 July 2011.

[5] Hamzah Ali, “Genetic algorithm based approach in
Artificial Neural Network for Pattern Recognition”-
M.tech Report, 2008-2009,Dept. of Information Science
& Engineering, M.S Ramaiah Institute of Technology,
Bangalore.


Volume 1, Issue 2 July-August 2012                                                                                 Page 16

				
DOCUMENT INFO
Description: International Journal of Emerging Trends & Technology in Computer Science (IJETTCS) is an online Journal in English published bimonthly for scientists, Engineers and Research Scholars involved in computer science, Information Technology and its applications to publish high quality and refereed papers. Papers reporting original research and innovative applications from all parts of the world are welcome. Papers for publication in the IJETTCS are selected through rigid peer review to ensure originality, timeliness, relevance and readability. The aim of IJETTCS is to publish peer reviewed research and review articles in rapidly developing field of computer science engineering and technology. This journal is an online journal having full access to the research and review paper. The journal also seeks clearly written survey and review articles from experts in the field, to promote intuitive understanding of the state-of-the-art and application trends. The journal aims to cover the latest outstanding developments in the field of Computer Science and engineering Technology.