Automated Rubber Seed Clones Identification Using Reflectance Sensors
And PIC
Mohd zafran abdul aziz,Fairul nazmie osman,Hadzli hashim
Faculty of Electrical Engineering
Universiti Teknologi MARA
40450 Shah Alam
MALAYSIA
[email protected],[email protected], [email protected]
Abstract: - This paper studies automated rubber seed clones identification using Reflectance sensors by create an
intelligent and simple technology using application of Microcontroller PIC 16F877A. In order to recognize
selected best rubber tree series clones based on reflectance at their surface. There are five types of clones from the
same species of rubber seed have been used as samples in this paper which are PB360, RRIM2009, RRIM2011,
RRIM2016 and RRIM2025. There three sensor reflectance (QTR-1A) was used to ensure that all surface samples
taken the reading. Every surface reflects differently of light. The device measure the percentages of reflectance
based on the intensity of light reflected from the seed surface. The Microsoft Office Excel was used to analysis the
average voltage that converts from light reflectance of wavelength by takes 30 samples readings from 5 difference
clones. Analysis results showed the average voltage derived from the data and graph, there are small differences in
voltage for each type of clone. It was found that there are the brightness of brown color at front surface can be used in
order to classify the types of rubber tree clones compare with back surface.
Key-Words: - infrared(IR),
1 Introduction Rubber is an important economic resource for the
world,especially in Malaysia, lots of studies have been
done to improve the quality of latex production in
rubber tree, In Malaysia there are more than 30 clones
of rubber tree, producing clone of so many causes
farmers and agriculture officers difficult to identify
rubber seeds clone. Rubber seeds have different
patterns on the surface, for experienced it is easy for
them to identify to see the difference in the pattern
of clones on the surface of seeds. However, these
methods will consume time, percentage accuracy and
as well as cost in order to trained new worker or farmer
with regards to the identification of rubber seed clones.
Before this the best rubber series clone was recognized
based on range of reflectance of wavelength, shape and
texture of the rubber series seeds through the image
p r oc e s s i n g a nd wave l e t t e c hn i q u e [ 1 -4 ] .
This paper present to classify rubber seed clones by
identify the diffrence color reflected from color and
pattern by analyze the reflectance of wavelength for 5
rubber seed clones types. The 5 clone were PB360 ,
RRIM2009, RRIM2011, RRIM2016 and
RRIM2025.The main objective of this paper is to
determine difference output voltage for each clone
based on light reflectance.Distinguish each of the
rubber seed were analyzed by determining the average
voltage is taken from reflectance sensor.The outcomes
data will be statically analyzed.
1.1 Clone characteristic
(a) RRIM2009 (b)RRIM2011 (c) RRIM2016
(d) RRIM 2025 (e) PB360
Figure 3 Types of rubber seed clones
Recent Researches in Circuits, Systems, Electronics, Control & Signal Processing
ISBN: 978-960-474-262-2 55
A) RRIM2009 (recommended): The seed is small in
size,smooth and shining with light brownish seed
coat. It has square to slightly rounded shape.
Overall growth and seed production of this clone is
good. This is the best type of clone and was
recommended for both latex and timber.
B) RRIM2011: The seed feature and size of this
clone mostly like same with RRIM2009 but the
RRIM2009 more bright compare this clone. Overall
growth and seed production of this clone is good.
This clone also recommended for both latex and
timber production [1].
C) RRIM2016: The seed features of this clone are
most likely same with RRIM2009 but it has square
shaped of seed. Overall growth and seed production
of the clone isvery good. This clone was also
recommended for latex andtimber production [1, 2].
D) RRIM2025: The seed is small in size, smooth and
shining with dark brownish seed coat. It has square to
slightly rounded shape. Overall growth and seed
production of this clone is good. This is the also best
type of clone and was recommended for both latex
and timber production [1].
E) PB360: The seed size of this clone is medium. It
was smooth and brownish seed coat. The shape is
square to slightly rectangular. The color is more
bright compare with other clones. This clone not
commercial the production but used as sample in
research of rubber seed clones.
2 Hardware Design There are 4 important components in completing this
paper, consisting of analog sensors, the internal ADC,
the Microcontroller and LCD as a display.
Figure 1 controller and display
Figure 1 shows the hardware design. PIC 16F877A is
a microcontroller operated based on program created
by using MPlab software and it interfaces with three
reflectance sensor QTR-1A, The sensors are used to
measure total of light reflected from the rubber seeds
surface using analog-to-digital (ADC) .These values
are subsequent by converted into voltage value by
using equation 1. PIC 16F877A has five different
port, A, B, C, D and E. To read analog input QTR-1A
must interface with port A.The QTR-1A reflectance
sensor carries a single infrared LED and
phototransistor pair. The phototransistor is connected
to a pull-up resistor to form a voltage divider that
produces an analog voltage output between and
(which is typically ) as a function of the
reflected IR.Lower output voltage is an indication of
greater reflection. Figure 5 shows QTR-1A
sensor.LCD Display 16x2 is used as indicators to
Display the type rubber seeds clone.
(1)
Figure 5 QTR-1A sensor
3 Experiment Figure 2 and 3 shows the overview of the rubber seed
clones detector.
Figure 2. Rubber seed clones detector
Recent Researches in Circuits, Systems, Electronics, Control & Signal Processing
ISBN: 978-960-474-262-2 56
Figure 3. Rubber seed clones detector
Various experiment has been conducted to realiaze
the proposed technique.The maximum light
reflection is very important when obtaining
readings. The sensors are arranged in parallel, to get
three different readings on the position, the top
(AN2), middle (AN0) and bottom (AN1). The three
sensors connected to port A 0, 1 and 2 because the
paper using the internal ADC is available on port A
for 16F877A.The sensors will scan the sample 255
times to get an average reading of the ADC. The
ADC values have been obtained will be changed to
voltage output by using the equation 1 :
Figure 6 ADC and voltage value
The scanning process was done at the Advanced
Signal Processing (ASP) and robotic Lab Faculty of
Electrical Engineering UiTM Shah Alam. The black
background is choosing to prevent the effect of
reflectance of light from environment surface. The
scanned light reflectance information was directly
extracted using MPlab software.Then the data will
be arranged to Microsoft Office Excel.
The Microsoft office Excel was used as statistical
measurement to represent the average reflectance of
wavelength by using graphical method.
For graphical method, a test was designed to study
different voltage for each clone. The average value of
the voltage difference will be used to create program
Lighting and position of rubber seed clones for each
experiment very important to get accurate readings.
4 Result Graph below shows the results that are taken during
experiment session. The reading involved 30 samples
for each clone which are PB360, RRIM2009 ,
RRIM2011 , RRIM2016 and RRIM2025.
Figure 7, 8, 9, 10 and 11 shows the voltage value
from three sensors, AN1, AN2 & AN3. The voltage
values were obtained from 30 different samples.
From the results, it can be concluded that the voltage
values obtained from AN1 & AN2 do not differ
much, except for AN3, the voltage are much higher.
Figure 7 RRIM2009
Figure 8 RRIM2025
Figure 9 PB360
Recent Researches in Circuits, Systems, Electronics, Control & Signal Processing
ISBN: 978-960-474-262-2 57
Figure 10 RRIM2011
Figure 11 RRIM2016
These experiments are performed to collect
frequencies of wavelength that have been reflected.
In order to record the necessary range of average
voltages, all the possibilities had taken.Table 1
tabulates the average voltage of rubber seeds clone.
Figure 12, 13 and 14 show three different graphs
plotted on the difference of voltage reading for 5
clones taken from three different sensors, namely
AN0 (middle sensor), AN1 (bottom sensor) and AN2
(top sensor).
Table 1
CLONE AVERAGE
AN0 AN1 AN2
RRIM2009 2924.667 3051.667 3638.633
RRIM2025 2974.500 3047.667 3691.833
PB360 1908.167 2238.500 2869.667
RRIM2011 2869.167 2956.000 3589.667
RRIM2016 2625.033 2757.833 3451.167
Figure 12 Middle Sensor (AN0)
Table 2, 3 & 4 tabulate the summarization of graph
plotted in Figure 12, 13 and 14 respectively. It depicted
the voltage range for 5 different clones.
Table 2
NO TYPE RANGE VOLTAGE (mV)
1 RRIM2009 2700-3300
2 RRIM2025 2900-3100
3 PB360 1500-2500
4 RRIM2011 2000-3300
5 RRIM2016 2100-2700
Figure 13 Bottom Sensor (AN1)
Table 3
NO TYPE RANGE VOLTAGE (mV)
1 RRIM2009 2700-3200
2 RRIM2025 2900-3300
3 PB360 2000-2500
4 RRIM2011 2600-3300
5 RRIM2016 2500-3000
Recent Researches in Circuits, Systems, Electronics, Control & Signal Processing
ISBN: 978-960-474-262-2 58
Figure 14 Top Sensor (AN2)
Table 4
NO TYPE RANGE VOLTAGE (mV)
1 RRIM2009 3500-3800
2 RRIM2025 3600-3700
3 PB360 2800-3300
4 RRIM2011 3100-3900
5 RRIM2016 3400-3600
TABLE 5 tabulates the average voltage range
suitable for 5 types rubber seed clones. Types PB360,
RRIM 2009 and RRIM16 have different voltage
range but for RRIM 2025 and RRIM 2011 but for a
and b of the voltage range they are very close to each
other. Figure 15 shows example for 2 types of clones
PB360 and RRIM 2016.
Table 5
NO RANGE VOLTAGE (mV) TYPE
1 1500-2500 PB360
2 2800-3400 RRIM2009
3 3600-4000 RRIM2025
4 2500-2700 RRIM2016
5 3500-3700 RRIM2011
(a) Result for PB360
(b) Result for RRIM 2016
Figure 15 Example result display
5 Conclusion The five types of rubber seed clones
PB360.RRIM2009, RRIM2011, RRIM2016 and
RRIM 2025 are being tested in order to observe the
differences among the clones in terms of maximum
percentage reflectance of wavelength. Data extracted
from hardware design were analyzed with statistical
method. From the observation, the surface of rubber
seeds has different reflection percentage. By taking a
significant difference on the 3 surface, the process of
differentiation will be easier. The maximum
percentage reflectances of wavelength are
distinguishing each type of clones series can be done
by for front surface. The voltage range of PB360,
RRIM2016 and RRIM2009 clones series distinguish
each other but RRIM2011 and RRIM2025 clones
series have same range of voltage. Therefore, it can
conclude that analyzing more surfaces of rubber
seeds can be used in order to classify the types of
rubber seed clones.
References:
[1] M. Z. A. Aziz, D. R. Othman, D. M. Benong, and
D.O. S. Huat, RRIM2000 SERIES CLONES :
Characteristic and Description: RUBBER
RESEARCH INSTITUTE OF MALAYSIA,
1997.
[2] H. M. Salleh, "An Automated Shape Recognition
for Rubber Seed Cloth rough Imaging
Techniques," in Faculty of Electrical Engineering,
vol. Bsc.Eng(Hons.) Electrical. Shah Alam:
Universiti Teknologi MARA, 2007, pp. 9.
[3] A. E. Darani, "A Study on LED Reflectance for
Various Rubber Seeds Clone," in Faculty of
Electrical Engineering, vol. Bsc.Eng (Hons.)
Electrical. Shah Alam: Universiti Teknologi
MARA,pp.10.
[4] E. A. Anua, "An Automated Texture Recognition
for Rubber Seed Clones through Wavelet
Techniques, " in Facultyof Electrical Engineering,
vol. Bsc.Eng (Hons.) Electrical. Shah Alam:
Universiti Teknologi MARA,pp10.
[5] Li Jinwei, Liao Guiping, Xiao Fen, "Rapeseed
Seeds Colour Recognition by Machine Vision," in
Institute of Agricultural Information, Human
Agricultural University, Changsa 410128, P. R.
China.
Recent Researches in Circuits, Systems, Electronics, Control & Signal Processing
ISBN: 978-960-474-262-2 59