Scalable Computing: Practice and Experience, ISSN 1895-1767, http://www.scpe.

org
© 2020 SCPE. Volume 21, Issues 4, pp. 583–589, DOI 10.12694:/scpe.v21i4.1699

A REVIEW ON THE ROLE OF MACHINE LEARNING IN AGRICULTURE

SYAMASUDHA VEERAGANDHAM∗ AND H SANTHI†

Abstract. Machine learning is a promising domain which is widely used now a days in the field of agriculture. The availability
of manpower for agriculture is not enough and skill full farmers are less. Understanding the situation of the crop is not that
much easy to detect and prevent the diseases in the crop. It is also widely employed in various agricultural fields such as topsoil
management, yield management, water management, disease management and climate conditions. The machine learning models
facilitate very fast and optimal decisions. The model of machine learning involves with training and testing to predict the accuracy
of the result. The use of machine learning in agriculture helps to increase the productivity and better management on soil
classification, disease detection, species management, water management, yield prediction, crop quality and weed detection. This
article aims at providing detailed information on various machine learning approaches proposed in the past five years by emphasizing
the advantage and disadvantages. It also compares different machine learning algorithms used in the modern agricultural field.

Key words: Machine Learning, Agriculture, Data Analysis, Training Methods and Sensors.

AMS subject classifications. 68T05

1. Introduction. Now a days, agriculture is the major source in all over the world, it plays a vital role in
the global economy. Many of the research projects and funding projects are continuously implement with the
latest technologies. Increasing the yield of the crop, automating the work done for the crop, minimizing the
manpower, reducing the disease ratio by detecting on initial stages with existing matched patterns and finally
harvesting with machinery in minimum time [1]. Machine learning in agriculture implemented with different
crops from sowing to harvesting using many techniques from different technologies like big data, artificial
intelligence, drones and data mining. Using these technologies, for mapping with existing data to identify and
fix the solution of frequent problems. And defining the machine with related scenario based on the weather
conditions and the moisture position using sensors detectors.
Agriculture in normal processing with manpower may have many problems with less expertise for knowing
the crop complete processing, disease identification, appropriate pesticides usage on initial stages instead of
spreading on whole crop, huge manpower utilization from sowing to harvesting of the crop. There may be many
consequences to know everything about the crop and all types of crop manually, without having any experience.
So, the new people working on the crop have difficulties in knowing everything. Mainly the cost of the crop
is very high because of utilizing everything on manpower. Even now a days many machineries are using in
cultivating crops in agriculture and also manpower is needed to form sowing to harvesting, but many crop
automation is not possible because of the crops depends on environment conditions and soil management [1].
The Machine learning algorithms are processed on deep learning and artificial intelligence algorithms based
implementations. Train the system with deep learning and automated with artificial intelligence combination
mechanism. Using machine learning in agriculture trained the system with deep learning mechanism inbuilt with
sensors intermediate for collecting and processing the data. Once the current status data is processed by the
sensor then the machine understands the process of deep learning and also mapping with artificial intelligence.
Machine learning in agriculture mainly depends on the existing dataset mapping with new data set processed
by the sensor. Deep learning in agriculture also is a trend nowadays. Using the latest technologies on image
processing with data analysis of existing models which matches with the results and its excellent outcome.
∗ Research scholar, School of Computer Science and Engineering, Vellore Institute of Technology, Vellore, Tamilnadu, India

([email protected]).
† Correspondence author: Associate Professor, School of Computer Science and Engineering, Vellore Institute of Technology,

Vellore, Tamilnadu, India ([email protected]).

583
584 Syamasudha Veeragandham and H. Santhi

These days we can overcome many problems and challenges on different agriculture fields and food production
using deep learning mechanism. Deep learning used on different crop fields with natural language processing,
speech recognition with learning, sensors to detect temperature on soil based on the crop and weather condition
from sowing to harvesting of the crop [2].
The usage of machine learning helps farmers to collect the information and data by using it in agriculture
with the aid of information technology to make the best decisions on high output from the farms [3]. Machine
learning algorithms can be used for many applications in agriculture; like crop suggestions based on the pest
detection in plants, soil fertility, weed detection, yield cultivation of crops and plant disease detection based
on the disease identification on early stages will recovery the plant so that automatically increase the crop. It
is very important to minimize the utilization of pesticides as considering food quality and health measures of
the people. Apart from monitoring environmental conditions on a farm, intelligent agriculture must analyse
understand how weather circumstances cause environmental changes at the farm and how long-term crop
cultivation brings about soil erosion or changes in the soil structure. Through minimizing water management
we can save the water instead of wastage and use the same water for another crop. Monitoring a farm using
machine learning can prevent low productivity of crops [4]. Using crop management improves the yield of the
crop can take care from sowing to harvesting.
2. Literature survey. In agriculture, using machine learning indicates using many crops. A previous
study [5] explains the harvesting in date fruit orchard using robotics and Deep Learning mechanism. There are
two pre-learning CNN mechanisms; namely, AlexNet and VGG-16. To construct a study machine imaginative
and prescient system, also based on the high-quality image dataset of 5 data types for all maturity stages. The
suggested method accomplish extremely good classification based on the difficult dataset with matching ration.
The high-quality pixel images data sets are used in future to improve mapping of different date fruit orchards.
In another study [6] conducted on the graphical representation of modelling, primarily with the references of
sparse linear additive and proposed processes to discover a sparse in part linear additive shape on development
of directed open-chain graphical representation. The study updates the outcomes as a case study the use of
variable dataset accumulated on the runtime of the plant manufacturing and also proposed view is tremendous
for discovering strengthened variable or fixed based on temporary with proper output graph. They proposed
a method regarding the energy-efficient improvement management. Anyhow, the new approach that the team
invented predicted to be applicable for various surveys which are primarily based on the statistical methods.
Here, they mainly focused on the classification of normal additive samples, but their scheme can be without
difficulty, moved further with elaboration and naturally through involving arbitrary link models. Here they
assumed the hill mountaineering approach for getting associated best results inside the fixed period.
Machine learning is famous with its ability to achieve maximum on many domain-based technologies.
Machine learning can be regarded the top-rated analytic tool for fog computing applications. Instead of modern
day’s achievements, machine learning purposes and literature also plays a major role. The latest research
gap defines fog computing. The research achievement of machine learning in three elements are resource
management, accuracy and security. Machine learning is implementing mainly on resource management instead
of accuracy and security in fog computing. Machine learning include cloud computing in one of the layers. Even
many problems and challenges have been open-ended with these combinations. Even most of the challenging
problems regarding safety measures used in cloud computing. Supervised learning in fog computing is one
of the famous machine learning assignment. In the healthcare domain most of the applications have utilized
machine learning and also many open challenges and issues are there in fog computing in machine learning [7].
A prior study [8] has planned a correct and strong algorithm for a new mechanism to critically find the
growth of cucumber using robotic harvesting automated process in agriculture. This algorithm is a different
sort of implementations and mining methodologies of existing data to gain of cucumber field with extraordinary
components. This mechanism combines vector elements with image pixels match to get the starting stage itself
on next level onwards. Many outcomes of the yield were taken as the feedback as input for modelling and testing
the final algorithm. This algorithm outcome of the application is more efficient for harvesting the cucumber.
Deep Learning resolves many diseases from detecting with proper data set and using pesticides with minimal
quantity for curing the plants of the crop [9]. Another study [10] explains crop prediction, implementing
many techniques including artificial neural networks. Based on the artificial intelligence finding the soil state
 A Review on the Role of Machine Learning in Agriculture 585

depending on the weather condition with maximizing the yield of the crop. Later, a study [11] referred around
40 research papers defined as a survey on all the aspects of machine learning in agriculture. The reference papers
appeared in standard journals with high cited papers and many are the reputed papers with implemented on
all over the world on machine learning in agriculture. Machine learning in agriculture using sensor data on the
agricultural field using artificial intelligence with high suggestions from sowing to harvesting.
3. Machine Learning In Agriculture. The latest trend in agriculture using machine learning are im-
plementing smart farming with latest techniques. Many recent mechanisms are using for farming to find soil
moisture based on the type of crop with water management, disease detection and selection of pesticide with
existing patterns, crop management. Machine learning consists of five major components as shown in Figure
3.1. These five major components are 1). Collecting data from the farm 2). Stored data 3). Data pre-processing
4). Train the model and 5). Performance metrics.

Fig. 3.1. General process of machine learning

3.1. Collecting data from the farm. Most of the researchers used different sensors such as temperature
sensor, passive Infrared sensor, ultraviolet sensor, pH sensor, soil moisture sensor, humidity sensor, gas sensor,
barometric Pressure, hyperspectral camera, multispectral camera, DSLR camera, NPK sensor, obstacle sensor,
acoustic sensor, water level sensor, water quality Sensor, GPS sensor and MooMonitor sensor to collect the data
about the soil fertility, yield of crops, climate condition, pest detection in plants, weed detection, identifying
diseases, etc. This data gives more accurate results, but it is cost-oriented, time-consuming and also more
difficult to collect. A study [12] developed a robot with different sensors and it is collected various environmental
factors that effects on soil such as soil temperature, pH value in soil, the intensity of sunlight, soil moisture,
humidity etc. to grow a plant. Later, another study [13] proposed and designed an optical transducer to measure
soil nutrients such as Nitrogen, Phosphorus and Potassium, and these nutrients are called primary nutrients.
Soil nutrient not only depends on primary nutrients and also having macronutrients and micronutrients [14].
Generally, the soil fertility and environmental factors both are interrelated.
Another study [15] developed a smart irrigation system for agriculture by using temperature, humidity
and UV sensors. Soil fertility, types of crop and climate condition are the essential parameters for designing a
smart irrigation system. So, while collecting soil nutrients, also collect environmental factors and type of crop,
it will generate more accurate results for decisions from the data analysis. Hyperspectral Image of different
crops by using hyperspectral cameras for crop classification have used in another study [16]. Later a study
[17] has used the acoustic sensor for detecting insect’s pests in the underground. This sensor is mostly useful
for detecting insects in underground crops like carrot, potato, groundnuts, onion, taro, turmeric, garlic, etc.
A research collected [18] 83,260 and 16,652 colour images format as JPEG from wheat planting of Shandong
586 Syamasudha Veeragandham and H. Santhi

Province, China, to train and test a model respectively for winter wheat leaf diseases by using Canon EOS 80D
camera.
3.2. Standard data repositories. The main benefit of the existing data is the researcher doesn’t want to
spend money on collecting data. This data takes less time and also easy to collect, but it gives the less accurate
results [19]. This data is especially available in various websites. The researchers can download from the
websites and performs the data analysis. The forms of data include images, tables, text, graphics, audios and
videos. The following industries are mainly providing datasets for analysing the data by using machine learning
algorithms in agriculture. UCI, MIT, Kaggle, University of Arkansas, Live tree, China Agro. and Econ. Data,
Open Government Data, OpenAg, GitHub, Data. world, Knoema, USDA-ARS, V2 Plant Seedlings Dataset,
Food and agriculture data, Pesticide Use in Agriculture and Soil Resources Development Institute are providing
data to the data analysis.
A study [20] used soil related data of upazila of Khulna district in Bangladesh for soil classification and
crop suggestion, collected data from soil resources development institute, bangladesh. Another study [21] are
collected from 3010 images of rice plants with diseases from the high-standard rice experimental field of the
hunan rice research institute in China to detect the rice plant diseases. Another study [18] collected eight
categories of 16652 images from Shandong Province, China to identify 8 different diseases from the wheat
winter crops. And another study [22] collected 1070 real-time mango tree leaves images from shri mata vaishno
devi university in the district of Katra located in Jammu and kashmir, India to identify the fungal disease
named as anthracnose. This method is less expensive, easier to collect and cost-effective but the error rate is
high and not suitable for all areas.
3.3. Data pre-processing. The major issues in real-time data are inconsistent, duplication, noise and
missing values. This type of dataset is very critical for analysing the process and increases the error rate.
Data pre-processing performs on the raw data for further processing to enhance the quality of the data. Data
pre-processing is major crucial step in machine learning to improve efficiency while data processing. Pre-
processing can removing the noise, inserts the missing values, the appropriate data range, and extracting the
functionality etc. A study [21] used Two-Dimensional Filter Mask Combined with Weighted Multistage Median
Filter (2DFM-AMMF) to remove the noise from the rice plant images. Majority of image related works are
used segmentation and feature enhancement to improve the quality in data. Another study [23] removed salt
and pepper noise by using Gaussian Median and Gaussian filter respectively to enhance the image quality to
the 4-different crops and 2-weeds namely Paragrass and Nutsedge are chosen for classification. And another
study [24] used Principle Component Analysis to remove the dimensionality and multicollinearity problems for
water supply forecasting in the US West.
3.4. Train the model by using machine learning algorithms. Machine Learning algorithms are
classified into two types; Supervised Learning and Unsupervised learning. The use of machine learning in
agriculture helps to a). Soil classification b). Disease Detection c). Species management d). Water Management
e). Yield Prediction f). Crop Quality g). Weed Detection.
3.4.1. Soil classification. Soil is classified based on its strength and property, it can be helps to grow
the crop. Former uses the soil classification system for predicting the soil behaviour. Based on the chemical
and physical properties of the layers of soil, soil can be classified and named. Soil classification can be used
to identify the best crops and type of fertilizer based on the type of soil. Climate changes also plays a major
role in soil management with water management. Using machine learning techniques, suggests associated
procedures, moisture techniques concerning the temperature. A study[20] used Machine learning algorithms
such as Gaussian kernel-based Support Vector Machines (SVM), k-Nearest Neighbour (k-NN), and Bagged
Trees are used for soil classification, but proposed Gaussian kernel-based Support Vector Machines (SVM)
based method performs better than the k-Nearest Neighbour (k-NN), and Bagged Trees.
3.4.2. Disease Detection. Disease mainly depends on the weather and climatic conditions, soil charac-
teristics and plant strength. Detection of disease based on climatic conditions and viral diseases are broadly
pest and managing of disease utilization simultaneously. Disease detection also compares with a dataset as
per the weather conditions and age of the plant. A study [21] compared two machine learning algorithms
 A Review on the Role of Machine Learning in Agriculture 587

such as combination of Fuzzy C-means and K-mean clustering(FCM-KM) + faster Region Convolutional Neu-
ral Network(R-CNN) and faster Region Convolutional Neural Network(R-CNN) for detecting the rice plant
disease such as rice blast, bacterial blight and sheath blight, but proposed combined of Fuzzy C-means and
K-mean clustering(FCM-KM) + faster Region Convolutional Neural Network(R-CNN) performs better than
the faster Region Convolutional Neural Network(R-CNN). This method is not suitable for large scale rice plant
disease detection. A another study [3] compared 5-different machine learning algorithms such as Support Vector
Machines (SVM), Artificial Neural Networks (ANN), K-Nearest Neighbour (KNN), Fuzzy and Conventional
Neural Networks (CNN) for identification of different diseases in different crops. Among these 5-machine learn-
ing algorithms, Conventional Neural Network classification is given more accurate results and also suitable for
more crops. And another study [18] used Machine Learning algorithm such as Matrix-based Convolutional
Neural Network (M-bCNN) for winter wheat leaf diseases by classifying 8 different leaf diseases such as normal
leaf, mechanical damage leaf, powdery mildew, cochliobolus heterostrophus, bacterial leaf streak, bacterial leaf
blight, leaf rust and stripe rust.
3.4.3. Species management. Species Breeding: The selection of species is an important mechanism,
based on the soil along with region weather conditions and water associated vitamins with good taste. Using
deep learning procedures existing patterns, data sets are mapped for solving the many challenges instead of
assumptions. A study [25] used Cascaded Support Vector Machine algorithm for classifying 9-different sunflower
seeds.
Species Recognition: Manual selection of plants can be based on the leaves colour, age and shape of the
plant. The selection of plant will be the first step of the crop and the roots of the plant concerning the age and
veins colour on the leaves. A study [26] used 3- different classifiers like Color-Shape-Texture, Pixel and SIFT
based to classify 5 categories of species like Flowers, Fruit, Leaf, LeafScan and Stem. A another study [16] used
combination of Feature Band Set (FBS) and Object Oriented Classification (OOC) to classify different crops
using Hyperspectral Images.
3.4.4. Water Management. Water management plays a major role in every crop. Using a machine
learning mechanism we can efficiently use the water so that excess water will be used for another crop. Based
on the crop and soil type we can provide the water daily, weekly and monthly. A study [27] used soil temperature,
moisture and pH sensors to find the soil water content level in automatic water dripping system for agriculture.
The main benefit of an automatic irrigation system is to correct water usage, power and time saving but it
consumes more money from farmers.
3.4.5. Yield Prediction. Every crop we suppose to concentrate as per yield prediction. Yield prediction
defines mainly mapping of yield, demand based on the crop outcome and evaluation. The yield prediction can
be defined from the earlier dataset and what type of latest technologies available and applicable on all the ways
based on the current crop, climatic and financial situations for improving the yield. A study [28] used Tensor
Flow with Convolutional Neural Networks and Linear Regression for estimating the yield from Sorghum field.
3.4.6. Crop Quality. Crop quality is the way to finalize the crop outcome in the form of financial. Based
on the final quantity of yield, minimal wastage and the quality of the crop after harvesting can be detected. As
per these parameters, we can define the crop quality and also compare with the dataset. A study [29] connected
different sensors to the drone to monitor the crop quality. This drone can monitor the crop, gives alert to the
farmer when any issues identify on the crop.
3.4.7. Weed Detection. Weed detection is the main problem on every crop, based on this the final yield
defines. It is a very important threat on the crop that effects on the yield. Concerning the age of the plant and
the weed, the stage needs to detect. Once we minimize the weed then only the yield will be good otherwise the
fertilizers and the pesticides also not working on the crop. The work of the weed is to eat the whole energy of
the soil. So the crop quantity will be by default minimum. Machine learning has many mechanisms to detect
the type of weed on every crop and intimate. A study [30] used to Support Vector Machine and Conventional
Neural Network for detecting Broad-leaf weed detection in the pasture from the images. Another study [23]
used SVM, ANN and CNN for classifying the 4- different crops and 2- different weeds, but CNN gives better
results compared with the remaining methods.
588 Syamasudha Veeragandham and H. Santhi

3.5. Performance metrics. The best method is decided by using the accuracy. The method which has
the highest accuracy is the best. The more accurate models can give the better decisions as an outcome. The
following Table 3.1 shows different machine learning algorithms used in agriculture and with its accuracy.
Table 3.1
Comparison of different machine learning algorithms are used in agriculture.

S.No Subdomain Ref.No Crop Algorithm Accuracy

1. Soil Classification [20] —– J48 92.30
Support Vector Machines 94.95
2. Disease Detection [21] Paddy Diseases Faster R-CNN 91.28
FCM-KM+ Faster R-CNN 97.50
3. Disease Detection [18] Wheat Diseases Matrix-based CNN 96.50
4. Species Breeding [25] Sunflower Seeds Support Vector Machines 98.82
5. Yield Prediction [28] Sorghum field CNN and Linear Regression 74.50
6. Weed Detection [30] Pasture Support Vector Machine 89.40
CNN 96.88
7. Species Recognition [26] 5-Species SIFT based 98.00
8. Weed Detection [23] 4-crops,2-weeds SVM, ANN and CNN CNN best
9. Disease Detection [3] 6-crops SVM 92.31
2-crops ANN 93.70
2-crops KNN 88.75
1-crop FNN 88.00
4-crops CNN 98.62

4. Conclusion. Machine learning is widely used in modern agriculture. In addition to the Machine
learning, deep learning, artificial intelligence and robotics using much more for minimizing the manpower and
manual mistakes and almost everything needs to do automation form sowing to harvesting the crops. Using
the latest technologies and mechanisms for minimizing the manual mistakes for detecting the type of crop
to pesticides selection concerning the dataset mapped. The use of machine learning in agriculture helps in
the different sub areas like soil classification, disease detection, species management, water management, yield
prediction, crop quality and weed detection processing also implanting with machine learning. In this paper our
focus is to provide detailed survey about how various machine learning algorithms were used in different fields
of modern agriculture. This paper provides a detailed comprehensive comparative analysis of various machine
learning algorithms.

REFERENCES

[1] Seven Reasons Why Machine Learning Is a Game Changer for Agriculture, Available in:
https://towardsdatascience.com/7-reasons-why-machine-learning-is-a-game-changer-for-agriculture-1753dc56e310.
[2] What is ML and why do farming entrepreneurs care, Available in: https://medium.com/sciforce/machine-learning-
in-agriculture-applications-and-techniques-6ab501f4d1b5.
[3] Shruthi, U., Nagaveni, V., & Raghavendra, B. K. (2019, March). A review on machine learning classification techniques
for plant disease detection. In 2019 5th International Conference on Advanced Computing and Communication Systems
(ICACCS) (pp. 281-284). IEEE.
[4] Tseng, F. H., Cho, H. H., & Wu, H. T. (2019). Applying big data for intelligent agriculture-based crop selection analysis.
IEEE Access, 7, 116965-116974.
[5] Altaheri, H., Alsulaiman, M., & Muhammad, G. (2019). Date fruit classification for robotic harvesting in a natural
environment using deep learning. IEEE Access, 7, 117115-117133.
 A Review on the Role of Machine Learning in Agriculture 589

[6] Fujimoto, Y., Murakami, S., Kaneko, N., Fuchikami, H., Hattori, T., & Hayashi, Y. (2019). Machine Learning
Approach for Graphical Model-Based Analysis of Energy-Aware Growth Control in Plant Factories. IEEE Access, 7,
32183-32196.
[7] Abdulkareem, K. H., Mohammed, M. A., Gunasekaran, S. S., Al-Mhiqani, M. N., Mutlag, A. A., Mostafa, S. A., et
al. (2019). A Review of Fog Computing and Machine Learning: Concepts, Applications, Challenges, and Open Issues.
IEEE Access, 7, 153123-153140.
[8] Fernandez, R., Montes, H., Surdilovic, J., Surdilovic, D., Gonzalez-De-Santos, P., & Armada, M. (2018). Auto-
matic detection of field-grown cucumbers for robotic harvesting. IEEE Access, 6, 35512-35527.
[9] Yang, X., & Sun, M. (2019, April). A Survey on Deep Learning in Crop Planting. In IOP Conference Series: Materials
Science and Engineering (Vol. 490, No. 6, p. 062053). IOP Publishing.
[10] K, Sriram. (2019). , A Survey on Crop Prediction using Machine Learning Approach. International Journal for Research
in Applied Science and Engineering Technology. 7. 3231-3234. 10.22214/ijraset.2019.4542.
[11] Liakos, K. G., Busato, P., Moshou, D., Pearson, S., & Bochtis, D. (2018). Machine learning in agriculture: A review.
Sensors, 18(8), 2674.
[12] Krishna, K. L., Silver, O., Malende, W. F., & Anuradha, K. (2017, February). Internet of Things application for
implementation of smart agriculture system. In 2017 International Conference on I-SMAC (IoT in Social, Mobile,
Analytics and Cloud)(I-SMAC) (pp. 54-59). IEEE.
[13] Masrie, M., Rosman, M. S. A., Sam, R., & Janin, Z. (2017, November). Detection of nitrogen, phosphorus, and potassium
(NPK) nutrients of soil using optical transducer. In 2017 IEEE 4th International Conference on Smart Instrumentation,
Measurement and Application (ICSIMA) (pp. 1-4). IEEE.
[14] About the soil nutrients, available in: https://emeraldlawnsaustin.com/macronutrients-micronutrients-soil/.
[15] Goap, A., Sharma, D., Shukla, A. K., & Krishna, C. R. (2018). An IoT based smart irrigation management system using
Machine learning and open source technologies. Computers and electronics in agriculture, 155, 41-49.
[16] Zhang, X., Sun, Y., Shang, K., Zhang, L., & Wang, S. (2016). Crop classification based on feature band set construction
and object-oriented approach using hyperspectral images. IEEE Journal of Selected Topics in Applied Earth Observations
and Remote Sensing, 9(9), 4117-4128.
[17] Bayrakdar, M. E. (2019). A Smart Insect Pest Detection Technique With Qualified Underground Wireless Sensor Nodes
for Precision Agriculture. IEEE Sensors Journal, 19(22), 10892-10897.
[18] Lin, Z., Mu, S., Huang, F., Mateen, K. A., Wang, M., Gao, W., & Jia, J. (2019). A unified matrix-based convolutional
neural network for fine-grained image classification of wheat leaf diseases. IEEE Access, 7, 11570-11590.
[19] Farooq, M. S., Riaz, S., Abid, A., Abid, K., & Naeem, M. A. (2019). A Survey on the Role of IoT in Agriculture for the
Implementation of Smart Farming. IEEE Access, 7, 156237-156271.
[20] Rahman, S. A. Z., Mitra, K. C., & Islam, S. M. (2018, December). Soil classification using machine learning methods
and crop suggestion based on soil series. In 2018 21st International Conference of Computer and Information Technology
(ICCIT) (pp. 1-4). IEEE.
[21] Zhou, G., Zhang, W., Chen, A., He, M., & Ma, X. (2019). Rapid Detection of Rice Disease Based on FCM-KM and
Faster R-CNN Fusion. IEEE Access, 7, 143190-143206.
[22] Singh, U. P., Chouhan, S. S., Jain, S., & Jain, S. (2019). Multilayer convolution neural network for the classification of
mango leaves infected by anthracnose disease. IEEE Access, 7, 43721-43729.
[23] Sarvini, T., Sneha, T., GS, S. G., Sushmitha, S., & Kumaraswamy, R. (2019, April). Performance Comparison of
Weed Detection Algorithms. In 2019 International Conference on Communication and Signal Processing (ICCSP) (pp.
0843-0847). IEEE.
[24] Fleming, S. W., & Goodbody, A. G. (2019). A Machine Learning Metasystem for Robust Probabilistic Nonlinear
Regression-Based Forecasting of Seasonal Water Availability in the US West. IEEE Access, 7, 119943-119964.
[25] JayaBrindha, G., & Subbu, E. G. (2017). Ant colony technique for optimizing the order of cascaded SVM classifier for
sunflower seed classification. IEEE Transactions on Emerging Topics in Computational Intelligence, 2(1), 78-88.
[26] Purohit, S., Viroja, R., Gandhi, S., & Chaudhary, N. (2015, December). Automatic plant species recognition technique
using machine learning approaches. In 2015 International Conference on Computing and Network Communications
(CoCoNet) (pp. 710-719). IEEE.
[27] Padalalu, P., Mahajan, S., Dabir, K., Mitkar, S., & Javale, D. (2017, April). Smart water dripping system for
agriculture/farming. In 2017 2nd International Conference for Convergence in Technology (I2CT) (pp. 659-662). IEEE.
[28] Zannou, J. G. N., & Houndji, V. R. (2019, April). Sorghum Yield Prediction using Machine Learning. In 2019 3rd
International Conference on Bio-engineering for Smart Technologies (BioSMART) (pp. 1-4). IEEE.
[29] Saha, A. K., Saha, J., Ray, R., Sircar, S., Dutta, S., Chattopadhyay, S. P., & Saha, H. N. (2018, January). IOT-based
drone for improvement of crop quality in agricultural field. In 2018 IEEE 8th Annual Computing and Communication
Workshop and Conference (CCWC) (pp. 612-615). IEEE.
[30] Zhang, W., Hansen, M. F., Volonakis, T. N., Smith, M., Smith, L., Wilson, J., ... & Wright, G. (2018, June).
Broad-leaf weed detection in pasture. In 2018 IEEE 3rd International Conference on Image, Vision and Computing
(ICIVC) (pp. 101-105). IEEE.

Edited by: Rajkumar Rajasekaran

Received: Feb 21, 2020
Accepted: Apr 2, 2020