RPL: Soil Analyzer and Plant Suggestion

Software Development 2
P-29

SUBMITTED TO
Sir. Jeil Francisco

SUBMITTED BY
Jhon Paul A. Gervacio

Date: April 1, 2021

[1] Alasco, The analysis of this study This study was The As the rationale, it uses
Ricardo et was done through the uses aimed to develop technique artificial neural network, it
al., 2018., of digital image processing the SoilMATTic used in the was offered fast and
SoilMATTic: technique to efficiently for faster and study was accurate performance for
Arduino- identify (1) Nitrogen, (2) accurate soil Digital Image the image processing. The
Based Phosphorus, (3) Potassium analysis Processing system data base stored
Automated and (4) pH level of Philippine compared with and Artificial and manages 356
Soil Nutrient farmlands. The system was conventional Neural captured images where
and pH Level composed of five stages method to guide Network with 70% is for training, 15% for
Analyzer namely: automated soil farmers on crop Arduino- testing and 15% for
using Digital testing, image acquisition, suitability and based validation. Results of this
Image image processing, training increase farm prototype. this study showed 96.67
Processing system, and productivity and The Arduino- accuracy in identifying soil
and Artificial recommendation. crop yield. based macronutrient and pH level
Neural Specifically, this prototype, and gives fertilizer
Network study aimed to automated recommendation for
further develop the whole Inbred rice plant, Inbred
the conventional process of corn, Tobacco, sugarcane,
soil testing by macronutrient Pineapple, Mango,
automating the and pH Coconut, Abaca, Coffee,
whole process. analysis banana through a
The chemical of soil from generated report in printed
testing on the soil soil testing form.
is to be done by procedures
an automated up to fertilizer
machine using recommendat
gear motor and ion.
pumps controlled
by the Arduino.

[2] Arago, The analysis of this study This study was The As the rationale, the
Nilo et al., was done through the used aimed to provide technique Artificial neural network
2017.,SoilMA digital image processing more efficient used in the was applied in this study
Te: Soil technique to efficiently analysis of soil study was for its features that make it
macronutrie identify the Macronutrients Nitrogen, Digital Image well suited in offering fast
nts and pH and pH level of Soil in the Phosphorus and Processing and accurate performance
Level farmland of Philippines: (1) Potassium and and Artificial for the image processing.
Assessment Nitrogen,(2) Phosphorus,(3) pH level since the Neural The system will base on
for Rice Potassium and (4) pH. The project will do the Network. 448 captured image data,
 Plant composition of the system is analysis digitally 70% for training, 15% for
through made of four sections using image testing and 15% for
Digital Image namely, image acquisition, processing validation. Based on the
Processing image processing, training supported by result, the program will
using system, and result. different generate a report in
Artificial algorithms printed form. Overall, this
Neural through Artificial study identifies the soil
Network Neural Network macronutrient and pH level
(ANN). of the soil and gives
fertilizer recommendation
for inbred rice plant and
was proven 98.33%
accurate.

[3] Puno, The analysis of this study In general, this The As the rationale, the use of
John Carlo was done through the used study aims to technique Artificial Neural Network
et al., 2017., of image processing and develop a Soil used in the was used to hasten the
Determinatio artificial neural network to Nutrient Analyzer study was performance of image
n of Soil efficiently identify the using Soil Test Kit Digital Image processing in giving
Nutrients nutrients and pH level of soil and Rapid Soil Processing accurate result. The
and pH Level with the use of Soil Test Kit Testing of the and Artificial system will base on
using Image (STK) and Rapid Soil Bureau of Soils Neural captured image data, 70%
Processing Testing (RST) of the Bureau and Water Network with for training, 15% for testing
Using of Soils and Water Management the use of and 15% for validation as
Artificial Management: (1) pH, (2) through color Soil Test Kit default of neural network
Neural Nitrogen, (3) Phosphorus, recognition. (STK) and tool of MATLAB. Based on
Network (4) Potassium, (5) Zinc, (6) Rapid Soil the result, the program will
Calcium, and (7) Specifically, this Testing (RST) show the qualitative level
Magnesium. The study aims to of the Bureau of soil nutrients and pH.
composition of the system is develop a of Soils and Overall, this study
made of five sections program that will Water identifies the soil nutrient
namely soil testing, image give soil pH and Management. and pH level of the soil and
capturing, image six different was proven accurate.
processing, training system nutrient level
for neural network, and namely Nitrogen,
result. Phosphorus,
Potassium, Zinc,
Calcium, and
Magnesium using
image processing
and artificial
neural network
via MATLAB.
 INTERNATIONAL
Title/Author Analysis Objectives of the Technique Rationale
Study Used in the
Study

[4] Abimala, The analysis of the study This study was The As the rationale, HSV and
T.,2020.,Soil was done to support aimed to easily technique GLCM are used to perform
Classificatio agriculture by classifying 7 classify soils in used in the colour based feature
n & Crop different types of soils like different regions study was extraction. Gabor filters
Suggestion Clay, Clayey Peat, Clayey of India namely: Digital Image are used to perform
Based on Sand, Humus Clay, Peat, Clay, Clayey Processing texture based feature
HSV, GLCM, Sandy Clay and Silty Sand, Peat, Clayey and the pre- extraction. The features
Gabor and in suggesting suitable Sand, Humus processing is obtained from the test
Wavelet crops that could be grown in Clay, Peat, Sandy done by image are then compared
Techniques those particular soils using Clay and Silty using Low with the features obtained
and image processing. Pre- Sand with Pass filter. from the images in the
Decision processing is done by using required accuracy HSV, GLCM, dataset. Matching of
Tree Low Pass filter. HSV, and particularly Gabor image features is achieved
Classifier in GLCM, Gabor Wavelet suggesting Wavelet by training the Decision
Image algorithms are used for suitable crops algorithms Tree classifier with
Processing feature extraction. using image are used for statistical measurements
processing. feature like mean, standard
extraction. deviation, skew and
kurtosis. Finally the soil is
predicted with the help of
segmented images that
are given as input for
simulation using Matlab
R2018a and is followed by
crop suggestion.

[5] The analysis of the study This study was The As the rationale, from the
Rajendran, was done through a soil- aimed to develop technique statistical analysis, the
Ganesh.,202 based crop selection and a Soil Test Based used in the data was collected in a
0., Soil Test fertilizer management Smart Agriculture study was various soils and then
Based Smart system has with a Management Digital Image identified which crop was
Agriculture normalized approach to System for crop Processing suitable for cultivation to
Management characterize the identification and and Artificial produce maximum yield
System composition of bare soil, its type in some Neural for a particular field.
metalloids, and mesological agricultural land Network with Various sensors, such as
parameters. in India Fuzzy humidity sensor,
particularly at Modeling and temperature, and camera,
Tanjavur, Decision were used to control and
Tamilnadu. Algorithm. monitor the agriculture
 field. In this study, the area
around Tanjavur,
Tamilnadu has been
considered for the crop
selection of that area.
Irrigation is controlled by
water level sensors, GSM,
and a controller. An APP
has developed to identify a
suitable crop for the
agriculture area, and it can
easily install in the farmer's
mobile phone itself.
Simulation has done using
Matlab for four different
crops. The results indicate
that the proper utilization
of fertilizers protects the
agricultural field and
increases productivity. An
experimental setup is
developed and tested
under different test
conditions.

[6] Raut, The analysis of the study This study was The As Rationale, IoT
Swapnil.,202 was taken through soil tests aimed to develop technique associates the entire world
0., Soil from various yield to check Soil Monitoring used in the with the assistance of
Monitoring and observing of soil. The and Testing Using study was sensors and other installed
and Testing PH sensor is utilized to IoT, what's more, utilizing data gadgets. The diverse soil
Using IoT for figure hydrogen particle- with utilized of mining or tests will be taken from
Fertility based action and to check this examined Matlab various fields and soil
Level and acidic and alkalinity in the information, technique. esteems will be
Crop soil. The PH sensor is regardless of This also determined to utilize a PH
Prediction associated with Arduino whether the done with sensor where
microcontroller, for the harvest will ripe in Arduino supplements worth will be
information will be detected this soil or not all microcontrolle separated from it. At that
and monitored. around r, PH sensor point, live information will
anticipated and Wi-Fi be sent to the cloud
through a modulle. For database where
grouping of soil. this soil information will be put
In light of this examples away. At that point
worth, farmers esteem which dependent on information
can show signs of put away in or data it will be broken
improvement for the database down lastly we show signs
 cultivating or will be of improving crop yield.
gathering of categorized. Here for information broke
harvests in yield. Categorizatio down information mining
Through this n is the method is utilized.
Indian farmer will demonstratio Information mining in
get crop list on n or agriculture assumes an
the versatile procedures of imperative job in yield
application and isolating expectation, soil
furthermore soil things into fruitfulness and plant
esteems and different illnesses and so on.
significantly more. gatherings as Farmers can screen
indicated by information on versatile by
their sorts. a website. What's more,
farmers will likewise get
crop list dependent on that
information which harvests
will be conceivable to yield
in that soil.
Bibliography

[1] R Alasco, R., Domdoma, E., Doria, K., Ricafranca, K., Saito, Y., Arago, N., Reyes, R.,
& Galido, E. (2018, November 29). SoilMATTic: Arduino-Based Automated Soil Nutrient and pH
Level Analyzer using Digital Image Processing and Artificial Neural Network.
https://Ieeexplore.Ieee.Org/Document/8666264. https://ieeexplore.ieee.org/document/8666264
[2] Arago, N., Orillo, J.W., Haban, J.J., Juan, J., Puno, J.C., Quijano, J.F., & Tuazon, G.M.
(2017). SoilMATe: Soil Macronutrients and pH Level Assessment for Rice Plant through Digital Image
Processing Using Artificial Neural Network. Journal of Telecommunication, Electronic and Computer
Engineering, 9, 145-149.
[3] Puno, J.C., Sybingco, E., Dadios, E., Valenzuela, I.C., & Cuello, J. (2017). Determination
of Soil Nutrients and pH level Using Image Processing and Artificial Neural Network. 2017IEEE 9th
International Conference on Humanoid, Nanotechnology, Information Technology, Communication
and Control, Environment and Management (HNICEM), 1-6.

[4] Abimala, T., Sashya, S., & Sripriya, K. (2020, June 3). Soil Classification and Crop
Suggestion using Image Processing. https://easychair.org/publications/preprint/5gZm
[5] Ganesh Babu, R., C., C., G., S., DanielRaj, T., V., K., & MullaM, A. (2020). Soil Test Based
Smart Agriculture Management System. 2020 7th International Conference on Smart Structures and
Systems (ICSSS), 1-6. https://www.semanticscholar.org/paper/Soil-Test-Based-Smart-Agriculture-
Management-System-GaneshBabu-C./55f64e3be6fc07a6a0a7ae3776114949e878605b

[6] Raut, Swapnil and Chitre, Vidya, Soil Monitoring and Testing Using IoT for Fertility
Level and Crop Prediction (April 8, 2020). Proceedings of the 3rd International Conference on
Advances in Science & Technology (ICAST) 2020, Available at
http://dx.doi.org/10.2139/ssrn.3561682