Applications of Pattern Recognition Algorithms in Agriculture: A Review

Int. J. Advanced Networking and Applications
Volume: 6 Issue: 5 Pages: 2495-2502 (2015) ISSN: 0975-0290
2495
Applications of Pattern Recognition
Algorithms in Agriculture: A Review
M. P. Raj
Research Scholar, Rai University, Ahmedabad, Gujarat, India
Email: rajmayur2005@gmail.com
P. R. Swaminarayan
Professor ISTAR, Anand, Gujarat, India.
Email: swaminarayan.priya@yahoo.com
J. R. Saini
Associate Professor,
Narmada College of Computer Application, Zadeshwar, Bharuch, Gujarat, India
Email: saini_expert@yahoo.com
D. K. Parmar
Research Scholar, Rai University, Ahmedabad, Gujarat, India
Email: dig.matlab@gmail.com
--------------------------------------------------------------------ABSTRACT-------------------------------------------------------------
Pattern recognition has its roots in artificial intelligence and is a branch of machine learning that focuses on the
recognition of patterns and regularities in data. Data can be in the form of image, text, video or any other format.
Under normal scenario, pattern recognition is implemented by first formalizing a problem, explain and at last
visualize the pattern. In contrast to pattern matching, pattern recognition algorithms generally provide a fair
result for all possible inputs by considering statistical variations. Probabilistic classifiers have supported
Agricultural statistical inference for decades. Potential applications of this technique in agriculture are numerous
like pattern recognition from satellite imagery, identifying the type of disease from leaf image, weed detectionetc.
This paper explores employment of pattern recognition in an agricultural domain.
Keywords: Classification, Feature Extraction, Feature Selection, Pattern Recognition, Pattern Recognition
Models, Agriculture.
------------------------------------------------------------------------------------------------------------------------------------------------
Date of Submission: February 02, 2015 Date of Acceptance: March 04, 2015
-----------------------------------------------------------------------------------------------------------------------------------------------
I. INTRODUCTION
Agriculture industry reflects a large portion of
economic output. Together with the breeding industry,
researchers try to identify, improve, and breed key traits
to satisfy the growing demands, increase resistance to
parasites and diseases, reduce environmental impact
(less water, less fertilizer), always striving for a more
sustained agriculture.
These can be satisfied if precision farming is
implemented. Looking at the scientific literature on
precision farming, it appears that, most efforts so far
were focused on the development and deployment of
sensor technologies rather than on methods for data
analysis tailored to agricultural measurements. In other
words, up to now, contributions to computational
intelligence in agriculture mainly applied off-the shelf
techniques available in software packages or libraries
but did not develop specific frameworks or algorithms.
Yet, efforts in this direction are noticeably increasing
and in this paper we review recent work on pattern
recognition in agriculture.
Pattern recognition is a multi-disciplinary subject
covering the fields of statistics, engineering, artificial
intelligence, computer science, psychology, physiology,
etc. [[1][2][3]]. The field of pattern recognition is
concerned with the automatic discovery of regularities
in data through the use of computer algorithms and with
the use of these regularities to take actions such as
classifying the data into different categories [4]. In
simple terms, one can say that Pattern recognition
prominently used for classification or clustering. From
centuries, human beings are solving a number of
problems by analogical reasoning.
However, computer-based automated pattern
recognition systems are required when the human
senses fail to recognize patterns, or if there is a need for
automating and speeding up the recognition
process[5].Pattern reasoning employs same paradigm in
solving problems in different domains by scrutinizing
relevant patterns. Its main notion is to elicit patterns
from the study area and to bifurcate the study area in to
classes. Application of Pattern recognition systems can
be trained or untrained. Trained methods are known as
supervised learning and untrained is categorized as
unsupervised learning.Solutions provided by pattern
recognition can be found everywhere. For example it
can be used in disease categorization, prediction of
survival rates for patients withspecific disease[6],
fingerprint verification[7], face recognition[[8][9]], iris

2496
discrimination[10], chromosome shape
discrimination[11], optical character recognition[12],
texture discrimination[13]and many more.Pattern
recognitionsystem implements three major steps viz.
preprocessing, feature extraction, and classification for
solving any problem on hand.
Pattern recognition system should consider the
application domain for selecting pattern. Same pattern
recognition system cannot be employed for all domains.
For most practical applications, the original input
variables are typically preprocessed to transform them
into some new space of variables where, it is desired.
For instance, in the digit recognition problem, the
images of the digits are typically translated and scaled
so that each digit is contained within a box of a fixed
size. This greatly reduces the variability within each
digit class, because the location and scale of all the
digits are now the same, which makes it much easier for
a subsequent pattern recognition algorithm to
distinguish between the different classes.
II. PATTERN RECOGNITION PROCESS
Pattern
There are various definitions of the term pattern:
According to Wordweb dictionary; pattern is a
perceptual feature, a customary way of operation or
behavior, a decorative or artistic work, something
regarded as a normative example, a model considered
worthy of imitation, something intended as a guide for
making something else or graphical representation of
the spatial distribution of radiation from an antenna as a
function of an angle.
“A pattern is essentially an arrangement. It is
characterized by the order of the elements of which it is
made, rather than by the intrinsic nature of these
elements,” is a definition given by Norbert Wiener [14].
Watanabe [15] defines a pattern as “opposite of a chaos;
it is an entity, vaguely defined, that could be given a
name”. “It can also be defined by the common
denominator among the multiple instances of an entity.
For e.g., commonality in all fingerprint images defines
the fingerprint pattern; thus, a pattern could be a
fingerprint image, a handwritten cursive word, a human
face, a speech signal, a bar code, or a web page on the
Internet” [16].
Pattern recognition system
figure -1: General steps required for supervised and
unsupervised classification
First process of any pattern recognition system is
dimensionality reduction. Dimensionality reduction
process deals with the removal of noise (i.e. irrelevant)
and redundant features.
The design model of a pattern recognition system
essentially involves the following steps [[17][18]]:
1) Preprocessing: Here the data from the surrounding
environment is taken as an input. The raw data is then
processed by either removing noise from the data or
extracting pattern of interest from the background so as
to make the input readable by the pattern recognition
system.
Next two steps viz. feature extraction and feature
selection are capable of improving learning
performance, lowering computational complexity,
building better generalized models, and decreasing
required storage[19].
2) Feature extraction: Feature is the measurable or
observable data corresponding to the pattern. Feature
extraction eliminates redundant data and retrieves
characteristic information about the pattern. Elimination
of redundant information is of vital importance for the
reduction of processing time in the recognition process
[5].Form processed data identical and relevant features
are extracted. These relevant features collectively form
identity of an object to be recognized or classified.
Many methods of feature extraction exist like Fourier
transform, Radon transform, Gabor Wavelets transform,
Fuzzy invariant transform, principal component
analysis, Semidefinite embedding, Multifactor
dimensionality reduction, Multilinear subspace
learning, Nonlinear dimensionality reduction, Isomap,
Kernel PCA, Multilinear PCA, Latent semantic
analysis, Partial least squares, Independent component
analysis, Autoencoder etc.
3) Feature selection: The objective of variable
selection is three-fold: improving the prediction
performance of the predictors, providing faster and
more cost-effective predictors, and providing a
betterunderstanding of the underlying process that
generated the data[20].
List of feature extracted from the feature extraction step
are passed through a one more filtering process to
obtain more discriminative or representative subset of

2497
feature vector. During this process, filtering is done
without any transformation and maintains the physical
meaning of the original features. Feature vector/subset
available at the end of this step is also known as
training data set. Feature selection allows us to better
understand the domain and cost cutting can be achieved
by reducing set of predictors.
These properties of feature selection ultimately
help in improving performance of classification
algorithms. This process aims not only to increase
dimension reduction rate but also to prevent the effect
of curse of dimensionality [21][2]. Feature selection is
different from dimensionality reduction. Both methods
seek to reduce the number of attributes in the dataset,
but a dimensionality reduction method do so by creating
new combinations of attributes, whereas feature
selection methods include and exclude attributes present
in the data without changing them [22]. Feature
selection techniques at top level are bifurcated in to
wrappers, filters and embedded.
Wrapper methods use a predictive model to score
feature subsets. Each new subset is used to train a
model, which is tested on a hold-out set. Counting the
number of mistakes made on that hold-out set (the error
rate of the model) gives the score for that subset. As
wrapper methods train a new model for each subset,
they are very computationally intensive, but usually
provide the best performing feature set for that
particular type of model.
Filter methods use a proxy measure to score a feature
subset. This measure is chosen to be fast to compute.
Common measures include the mutual information,
[20] the point wise mutual information,[23] Pearson
product-moment correlation coefficient, inter/intra class
distance or the scores of significance tests for each
class/feature combinations.[23][24] Filters are usually
less computationally intensive than wrappers, but they
produce a feature set which is not tuned. Filters can be
uses are pre-processing part of wrapper methods.
Embedded methods perform variable selection in the
process of training and are usually specific to given
learning machines.
Feature selection techniques include methods like
Information Gain, Relief, Fisher Score and Lasso.
figure -2: General feature selection structure [25]
4) Classification: Classification is the problem of
identifying, i.e. to which set of categories
(subpopulations) a new observation belongs, on the
basis of a training set of data containing observations
(or instances) whose category membership is known
[19]. Performance of pattern recognition algorithm is
dependent on this step, so it is one of the crucial process
in pattern recognition systems. Inputs of this process are
resultant refined feature vector/set obtained at the end
of feature selection processes and classification dataset
which is to be classified based on former feature vector
or in some scenario it can be only classification dataset
only. In case if classification algorithm accepts refined
feature set from step 3 as input then it is known as
supervised classification algorithms and in its absence it
is known as unsupervised classification algorithms.
Supervised and unsupervised algorithms are enlisted in
Table-1.Sometimes unsupervised is also meansgrouping
the input data into clusters based on some implicit
similarity measure, rather than assigning each input
instance into one of a set of predefined classes [26]. So
in case of clustering or unsupervised classification
algorithm feature extraction and feature selection
processes are not mandatory. Figure-3 displays flow of
unsupervised and supervised classification algorithms.
Applications in which the training data along with
target data are employedare known as supervised
learning problems. The problems in which each input
vector is assigned to one of a finite number of discrete
categories, are called classification problems [4].
Regression is in which the desired output consists of
one or more continuous variables.
In other category, the training data consists of a set of
input vectors without target values. The motto in such
unsupervised learning problems is to identify groups of
similar sets within the data.This is called clustering, or
density estimation (to determine the distribution of data
within the input space), or visualization (to project the
data from a high-dimensional space down to two or
three dimensions)[4].
Unsupervised
Methods
Supervised Methods
1. Categorical
mixture models
2. Deep learning
methods
3. Hierarchical
clustering
4. K-means
5. Clustering
6. Correlation
clustering
7. Kernel PCA
1. Linear discriminant
analysis
2. Quadratic discriminant
analysis
3. Maximum entropy
classifier
4. Decision trees, decision
lists
5. Kernel estimation and
K-nearestneighbor
6. Naive Bayes classifier
7. Neural networks
(multilayer perceptrons)
8. Perceptrons
9. Support vector
machines
10. Gene expression
programming
Table 1– List of supervised and unsupervised methods

2498
figure-3: Unsupervised and supervised classification [27]
5) Decision making: Input of this process is classified
data. In any case i.e. (supervised or unsupervised
classification) this step is preceded by post processing
which help in better inferring and decisiveness [28].
III. PATTERN RECOGNITION MODELS
Pattern recognition models are bifurcated in to four
major categories viz. statistical approach, syntactic
approach,template matching & neural network[29][18].
Statistical Model
Statistical pattern recognition systems are extensively
used in today’s world because of its simplicity. It is
based on statistics and probabilities. In these systems,
traits are recoded in form of numbers and these
numbers vectors are used to create a pattern. Thus, each
pattern can be represented by specific multidimensional
vector, which in turn is used for pattern
recognition.Approximately about 80% of agricultural
research is supported by this approach.
Syntactic Model
Syntactic approach is widely used in theory of
computation. It is also known as a structural pattern
recognition model or rule based pattern recognition. In
this approach, patterns are represented by definite
structures like sentences belonging to language. In this
model, the patterns to be recognized are called
primitives and the complex patterns are represented by
the inter-relationship formed between these primitives
and the grammatical rules associated with this
relationship[29]. The patterns are the sentences
belonging to a language, primitives are the alphabet of
the language, and using these primitives, the sentences
are generated according to the grammar. Thus, the very
complex patterns can be described by a small number of
primitives and grammatical rules [30].
Template matching
Template matching is extensively used in image
processing domain. In this model pattern can be
recognized by clusters of pixel or curves to localize and
identify shapes in image. Thus patterns are in form of
templates. So from this it can be stated that supervised
classification algorithm will be mostly used. Scenario in
which pre-defined pattern are not known unsupervised
classification algorithm will be engaged.
Neural Network
Neural networks were originally inspired as being
models of the human nervous system. They have shown
many intelligent abilities, such as learning,
generalization and abstraction.Neural networks are
large networks of simple processing elements or node
which process information dynamically in response to
external inputs. The nodes are simplified models of
neurons or processing elements (PE). The knowledge in
a neural network is distributed throughout the network
in the form of internode connectionsand weighted links
(or synapse) which form the inputs to the nodes. The
link weights server to enhance or inhibit the input
stimuli values which are then added together at the
nodes. If the sum of all the inputs to a node exceeds
some threshold value T, the node executes and produces
an output which is passed on to other nodes or is used to
produce some output response.
IV. APPLICATIONS OF PATTERN RECOGNITION IN
AGRICULTURE
Pattern recognition is used in many area of science and
engineering that studies the structure of observations. It
is now frequently used in many applications in
manufacturing industry, health care and
military[16].Image processing based on morphology,
color and textural features of grains is necessary for
different applications in the grain industry including
assessing grain quality and variety classification. In
grain classification process, several techniques such
as statistical, artificial neural networks and fuzzy
logic have been used. Below listed is the some of the
contribution of pattern recognition in agriculture
domain:
Ankur M Vyas [31]surveyed different techniques
used to identify fruits based on colour. According to
them “In the automated fruit grading system the most
important feature is its colour. So for any automated
fruit grading system one should have the idea of
colour space and segmentation needs to be
performed. This paper provides a review of various
colour feature extraction techniques in detail.”
S. Arivazhagan et al.[32] proposed system as a
software solution for automatic detection and
classification of plant leaf diseases. The proposed
algorithm’s efficiency can successfully detect and
classify the examined diseases with an accuracy of
94%. Experimental results on a database of about 500
plant leaves confirm the robustness of the proposed
approach.

2499
J. Rajendra Prasad et al. [33] describe the DM
Framework development, description, components
used for crop prediction; planting strategist test
results are very much useful to the farmers to
understand market needs and planting strategies.
Victor Rodriguez-Galiano et al. [34] assessed
groundwater vulnerability to nitrate pollution using
Random Forest algorithm. Showed method of a
feature selection approach to reduce the number of
explicative variables. Developed predictive modeling
of nitrate concentrations at or above the quality
threshold of 50mg/L.
Christian Bauckhage Kristian and Kersting [35]
surveyed recent work on computational intelligence
in precision farming. From the point of view of
pattern recognition and data mining, the major
challenges in agricultural applications appear to be
the following:
1. The widespread deployment and ease of use
of modern, (mobile) sensor technologies
leads to exploding amounts of data. This
poses problems of BIG DATA and high-
troughput computation. Algorithms and
frameworks for data management and
analysis need to be developed that can easily
cope with TeraBytes of data.
2. Since agriculture is a truly interdisciplinary
venture whose practitioners are not
necessarily trained statisticians or data
scientists, techniques for data analysis need
to deliver interpretable and understandable
results.
3. Mobile computing for applications “out in
the fields” has to cope with resource
constraints such as restricted battery life,
low computational power, or limited
bandwidths for data transfer. Algorithms
intended for mobile signal processing and
analysis need to address these constraints.
They opted an approach based on a distributional
view of hyper-spectral signatures which they used for
Baysian prediction of the development of drought
stress levels. They also presented a cascade of simple
image processing and analysis steps of low
computational costs that allows for reliably
distinguishing different fungal leaf spots in natural,
unconstrained images of leaves of beet plants, that
allows farmers in the field to take pictures of plants
they suspect to be infected and have them analyzed
in real time.
Dr. D. Ashok Kumar & N. Kannathasan [36]
surveyed utility of data minning and pattern
recognition techniques for soil data minning and its
allied areas. The recommendations arising from this
research survey are:
A comparison of different data mining techniques
could produce an efficient algorithm for soil
classification for multiple classes. The benefits of a
greater understanding of soils could improve
productivity in farming, maintain biodiversity,
reduce reliance on fertilizers and create a better
integrated soil management system for both the
private and public sectors.
Farah Khan & Dr. Divakar Singh [37] endeavour to
provide an overview of some previous researches and
studies done in the direction of applying data mining
and specifically, association rule mining techniques
in the agricultural domain. They have also tried to
evaluate the current status and possible future trends
in this area. The theories behind data mining and
association rules are presented at the beginning and a
survey of different techniques applied is provided as
part of the evolution.
Amina Khatra [38] showed that using color based
image segmentation it is possible to extract the
yellow rust from the wheat crop images. Further, the
segmented yellow rust images can be exposed to
measurement algorithm where the actual penetration
of the yellow rust may be estimated in the yield. This
kind of image segmentation may be used for mapping
the changes in land use land cover taken over
temporal period in general but not in particular. The
success of the segmentation and actual penetration of
yellow rust mainly depend upon the positioning of
the cameras installed in order to acquire the images
from the field.
Archana A. Chaugule and Dr. Suresh Mali[39] in their
research Shape-n-Color feature set outperformed in
almost all the instances of classificationfour Paddy
(Rice) grains, viz. Karjat-6, Ratnagiri-2, Ratnagiri-4 and
Ratnagiri-24. They used Pattern classification was done
using a Two-layer (i.e. one-hidden-layer) back
propagation supervised neural networks with a single
hidden layer of 20 neurons with LM training
functions.The fifty-three features were used as inputs to
a neural network and the type of the seed as target.
Abirami et al. [40] used canny edge detection,
thersolding and scaled conjugate gradient training
with 9 neurons in hidden layer for grading basmati
rice granules. Scaled Conjugate Gradient Training
based Neural Network was able to classify granules
with the accuracy of 98.7%.
Various grading systems have been developed [[41],
[42],[43],[44]] which use different morphological
features for the classification of different cereal
grains.

2500
Utku, 2000[45] developed a system to identify 31
bread wheat and 14 durum wheat cultivars usingCCD
video camera.
Majumdar and Jayas [46][47][48][49] used digital
image processing and discriminate analysis to do
identification of different grain species. They used
morphological, color, textural and combination of
these features to describe physical properties of the
kernels.
Computer vision system offers an objective and
quantitative method for estimation of morphological
parameters and quality of agricultural products to
obtain quick and more reliable results [[50][51][52]].
Visen, 2004[53]has compared classification
performances of different neural network topology
by using morphological features of Canada Western
Amber Durum (CWAD) wheat, Canada Western Red
Spring (CWRS) wheat, oats, rye and barley.
Algorithms were developed to acquire and process
color imagesof bulk grain samples of five grain types,
namely barley, oats, rye,wheat, and durum wheat by
[54]. The developed algorithms were used toextract
over 150 color and textural features. A back
propagation neuralnetwork-based classifier was
developed to identify the unknown graintypes. The
color and textural features were presented to the
neuralnetwork for training purposes. The trained
network was then used toidentify the unknown grain
types. Classification accuracies of over98% were
obtained for all grain types.
R. D. Tillett [55] in his review highlighted multiple
areas of agriculture domain in which image
processing and different methods of pattern
recognition was implemented, viz. Harvesting of
oranges, tomatoes, mushrooms, apples, cucumbers,
Plant growth monitoring and grading of oranges,
potatoes, apples, carrots, green peppers, tomatoes,
peaches.
V. CONCLUSION
This paper is an attempt to provide an overview of some
previous research and studies done in the direction of
applying pattern recognition techniques in the
agricultural domain. A unique and proper combination
of pre-processing, feature extraction, feature selection
and classification process is required for each domain or
problem in order to optimize accuracy, speed and
reduce cost by minimizing feature set used for training
and classification. The theories behind pattern
recognition are presented at the beginning and a review
of different techniques applied in grading, remote
sensing, diseases detection etc.is provided as part of the
evolution.
REFERENCES
[1] R. O. Duda, P. Hart and D. Stork, Pattern
Recognition, USA: John Wiley & Sons, 2001.
[2] S. Theodoridis and K. Koutroumbas, Pattern
Recognition, USA: Academic Press, 2003.
[3] A. Webb, Statistical Pattern Recognition, England:
John Wiley & Sons Ltd., 2002.
[1] R. O. Duda, P. Hart and D. Stork, Pattern
Recognition, USA: John Wiley & Sons, 2001.
[2] S. Theodoridis and K. Koutroumbas, Pattern
Recognition, USA: Academic Press, 2003.
[3] A. Webb, Statistical Pattern Recognition, England:
John Wiley & Sons Ltd., 2002.
[4] C. M. Bishop, Pattern Recognition and Machine
Learning, Singapore: Springer Science+Business
Media, LLC, 2006 .
[5] D. S. Gunal, "AUTOMATED
CATEGORIZATION SCHEME FOR DIGITAL
LIBRARIES IN DISTANCE LEARNING:A
Pattern Recognition Approach," Turkish Online
Journal of Distance Education-TOJDE, vol. 9, p.
Number:4 Article 1, Octomber 2008.
[6] M. Steenweg, A. Vanderver, S. Blaser, T. d. K. A
Bizzi, G. Mancini and B. F. W. N. v. d. K. M. van
Wieringen WN, "Magnetic resonance imaging
pattern recognition in hypomyelinating disorders.,"
p. 136(Pt 9):2923, Sep 2013.
[7] A. A. Aburas and S. A. Rehiel, "Fingerprint
Patterns Recognition System Using Huffman
Coding," Proceedings of the World Congress on
Engineering, vol. III, 2008.
[8] W. Hwang, X. Huang, S. Z. Li and J. Kim, "Face
recognition using Extended Curvature Gabor
classifier bunch," Pattern Recognition, vol. 48, no.
4, p. 1243–1256, November 2014.
[9] S. Elaiwat, M. B. F. Boussaid and A. El-Sallam,
"A Curvelet-based approach for textured 3D face
recognition," Pattern Recognition, p. 1231–1242,
October 2014.
[10] J. Daugman, "The importance ofbeing random:
statistical principles ofiris recognition," Pattern
Recognition, vol. 36, p. 279 – 291, 2003.
[11] L. Zhang, W. Xu and C. Chang, "Genetic
algorithm for affine point pattern matching,"
Pattern Recognition Letters, vol. 24, pp. 9-19,
2003.
[12] F. Mohammad, J. Anarase, M. Shingote and P.
Ghanwat, "Optical Character Recognition
Implementation Using Pattern Matching," (IJCSIT)
International Journal of Computer Science and
Information Technologies, vol. 5, pp. 2088-2090,
2014.
[13] X. Liu and D. Wang, "A spectral histogram model

2501
for texton modeling and texture discrimination,"
Vision Research, vol. 42, no. 23, p. 2617–2634,
October 2002.
[14] R. C. Gonzalez , "“Object Recognition”, in Digital
image processing,," 3rd ed. Pearson, pp. 861-909,
2008.
[15] S. Watanabe, “Pattern Recognition: Human and
Mechanical”, New York: Wiley, 1985.
[16] A. K. Jain and P. D. Robert, “Introduction to
pattern recognition”, The Oxford Companion to the
Mind, Second Edition, Oxford, UK: Oxford
University Press, 2004.
[17] M. Parasher, S. Sharma, A. K. Sharma and J. P.
Gupta, "“Anatomy On Pattern Recognition","
Indian Journal of Computer Science and
Engineering (IJCSE), Vols. vol. 2, no. 3, Jun-Jul
2011.
[18] S. Asht and R. Dass, "“Pattern Recognition
Techniques: A Review”," International Journal of
Computer Science and Telecommunications, vol. 2,
no. 8, August 2012.
[19] J. Tang, S. Alelyani and H. Liu, Feature Selection
for Classification: A Review.
[20] I. Guyon and A. Elisseeff, "An Introduction to
Variable and Feature Selection," Andr´e Elisseeff,
vol. 3, pp. 1157-1182, March 2003.
[21] A. Jain and D. Zongker, "“Feature selection:
evaluation, application, and small sample
performance”," IEEE Trans. on Pattern Analysis
and Machine Intelligence, vol. 19, no. 2, p. 153–
158, 1997.
[22] J. Brownlee, "An Introduction to Feature
Selection," 06 Octomber 2014. [Online]. Available:
http://machinelearningmastery.com/an-
introduction-to-feature-selection/. [Accessed 29
December 2014].
[23] Y. Yang and J. O. Pedersen, "A comparative study
on feature selection in text categorization," ICML,
1997.
[24] G. Forman, "An extensive empirical study of
feature selection metrics for text classification,"
Journal of Machine Learning Research, vol. 3, p.
1289–1305, 2003.
[25] J. NOVAKOVIĆ, P. STRBAC and D.
BULATOVIĆ, "TOWARD OPTIMAL FEATURE
SELECTION USING RANKING METHODS
AND CLASSIFICATION ALGORITHMS,"
Yugoslav Journal of Operations Research 21, pp.
119-135, March 2011.
[26] "Pattern Recognition," [Online]. Available:
http://en.wikipedia.org/wiki/Pattern_recognition.
[Accessed 18 Dec 2014].
[27] "Lesson 1: Image Processing and Interpretation -
Morro Bay, California," [Online]. Available:
http://wgbis.ces.iisc.ernet.in/envisrs/?q=node/26/.
[Accessed 29 December 2014].
[28] P. Sharma and M. Kaur, "Classification in Pattern
Recognition: A Review," International Journal of
Advanced Research in Computer Science and
Software Engineering, vol. 3, no. 4, April 2013.
[29] J. Liu, J. Sun and S. Wang, "Pattern Recognition:
An overview," IJCSNS International Journal of
Computer Science and Network Security, vol. 6,
no. 6, June 2006.
[30] A. K. Jain, R. P. Duin and J. Mao, "Statistical
pattern recognition-A review," IEEE transactions
on Pattern Analysis and Machine Intelligence, vol.
22, no. 1, January 2000.
[31] A. M. Vyas, B. Talati and S. Naik, "Colour Feature
Extraction Techniques of Fruits: A Survey,"
International Journal of Computer Applications,
vol. 83(15), pp. 15-22, December 2013.
[32] S. Arivazhagan, R. N. Shebiah, S. Ananthi and S.
V. Varthini, "Detection of unhealthy region of
plant leaves and classification of plant leaf diseases
using texture features," Agric Eng Int: CIGR
Journal, vol. 15, no. 1, pp. 211-217, March 2013.
[33] J. R. Prasad, P. R. Prakash and S. S. Kumar,
"Identification of Agricultural Production Areas in
Andhra Pradesh," International Journal of
Engineering and Innovative Technology (IJEIT),
vol. 2, no. 2, pp. 137-140, August 2012.
[34] V. Rodriguez-Galiano, M. P. Mendes, M. J.
Garcia-Soldado, M. Chica-Olmo and L. Ribeiro,
"Predictive modeling of groundwater nitrate
pollution using Random Forest and multisource
variables related to intrinsic and specific
vulnerability: A case study in an agricultural
setting (Southern Spain)," Science of the Total
Environment 476–477, Elsevier, p. 189–206, 2014.
[35] C. Bauckhage and K. Kersting, "Data Mining and
Pattern Recognition in Agriculture," Springer-
Verlag Berlin Heidelberg, 2013.
[36] D. D. A. Kumar and N. Kannathasan, "A Survey
on Data Mining and Pattern Recognition
Techniques for Soil Data Mining," IJCSI
International Journal of Computer Science Issues,
vol. 8, no. 3, pp. 422-428, May 2011.
[37] F. Khan and D. D. Singh, "Association Rule
Mining in the field of Agriculture : A survey,"
International Journal of Scientific and Research
Publications, vol. 4, no. 7, July 2014.
[38] A. Khatra, "Yellow Rust Extraction in Wheat Crop
based on Color Segmentation Techniques," IOSR
Journal of Engineering (IOSRJEN), vol. 3, no. 12,
pp. 56-58, December. 2013.
[39] A. Chaugule and S. N. Mali, "Evaluation of Shape
and Color Features for Classification of Four
Paddy Varieties," I.J. Image, Graphics and Signal

2502
Processing, vol. 12, pp. 32-38, 2014.
[40] S. Abirami, P. Neelamegam and H. Kala, "Analysis
of Rice Granules using Image Processing and
Neural Network Pattern Recognition Tool,"
International Journal of Computer Applications,
vol. 96, no. 7, pp. 20-24, June 2014.
[41] S. P. Shouche, R. Rastogi, S. G. Bhagwat and K. S.
Jayashree, "Shape analysis of grains of Indian
wheat," Elsevier, Computers and Electronics in
Agriculture 33, p. 55–76, 2001.
[42] B. P. Dubey, S. G. Bhagwat, S. P. Shouche and J.
K. Sainis, " Potential of artificial neural networks
in varietal identification using morphometry of
wheat grains," Biosystems Engineering, vol. 95, p.
61–67, 2006.
[43] P. Zapotoczny, M. Zielinska and M. Nitab,
"Application of image analysis for the varietal
classification of barley: Morphological features.,"
Journal of Cereal Science., vol. 48, pp. 104-110,
2008.
[44] A. Masoumiasl, R. Amiri-Fahliani and A.
Khoshroo, "Some local and commercial rice
(Oryza sativa L.) varieties comparison for aroma
and other qualitative properties," International
Journal of Agriculture and Crop Sciences., vol. 5,
pp. 2184-2189, 2013.
[45] H. Utku, " Application of the feature selection
method to discriminate digitized wheat varieties.,"
Journal of Food Engineering., vol. 46, pp. 211-
216, 2000.
[46] S. Majumdar and D. S. Jayas, "Classification of
cereal grains using machine vision: I. Morphology
models.," Transaction of American Society of
Agricultural Engineering., vol. 43, pp. 1669-1675,
2000a.
cereal grains using machine vision: II. Colour
2000b.
cereal grains using machine vision: III. Texture
2000c.
[49] S. a. D. J. Majumdar, "Classification of cereal
grains using machine vision: IV. Combined
morphology, colour, and texture models.,"
Transaction of American Society of Agricultural
Engineering., vol. 43, pp. 1689-1694, 2000d.
[50] A. Arefi, A. Motlagh and A. Khoshroo,
"Recognition of weed seed species by image
processing," Journal of Food, Agriculture and
Environment, vol. 9, pp. 379-383, 2011.
[51] R. Choudhary, J. Paliwl and D. Jayas,
"Classification of cereal grains using wavelet,
morphological, colour, and textural features of
nontouching kernel images," Biosystem
Engineering, vol. 99, pp. 330-337, 2008.
[52] A. Khoshroo, A. Keyhani, S. Rafiee, R. Zoroofi
and Z. Zamani, "Pomegranate quality evaluation
using machine vision.," Proceedings of the First
International Symposium on Pomegranate and
Minor Mediterranean Fruits, pp. 347- 352, 2009.
[53] N. Visen, D. Jayas, J. Paliwal and N. White,
"Comparison of two neural network architectures
for classification of singulated cereal grains.,"
Canadian Biosystems Engineering, vol. 46, pp.
3.7-3.14, 2004.
[54] N. Visen, J. Paliwal, D. Jayas and N. White,
"Image analysis of bulk grain samples using neural
networks," Canadian Biosystems Engineering, vol.
46, no. 7, pp. 11-15, 2004.
[55] R. D. Tillett, "Image analysis for agriculture
processes: a review of potential opportunities,"
Jornal of Agricultural Engineering Research, vol.
50, pp. 247-258, September-December 1991.

Applications of Pattern Recognition Algorithms in Agriculture: A Review

More Related Content

What's hot (20)

Similar to Applications of Pattern Recognition Algorithms in Agriculture: A Review (20)

More from Eswar Publications (20)

Recently uploaded (20)

Applications of Pattern Recognition Algorithms in Agriculture: A Review