In our daily life, our vision and actions are influenced by an abundance of geometry and color information to identify a technical apparatus for identification. Only by analyzing color information do we subsequently decide whether we are to continue. In the industry, color processing is gaining a greater importance in digital signal processing. With technological advancement optical sensors for reading operations in the visible spectrum with integrated electronic readout circuits are getting affordable, that’s why we chose to use a low cost color light to frequency converter to build a color sorting machine. Our project was made in for pedagogical use but a further development using robust materials and components could lead to an industrial color sorting machine for a use in the color sorting industry of recyclable materials.

1. Color sorting machine using color light to frequency converter
Assia HAMZA
assia.hamza@etu.usthb.dz
Merwan KHENAK
merwan.khenak@etu.usthb.dz
University of science and technology Houari Boumediene
BP 32, El Alia, Bab Ezzouar, 16111 Algiers, Algeria
Department of instrumentation and control
December 20, 2018
Abstract
In our daily life, our vision and actions are influenced by an abundance of geometry and color infor-
mation to identify a technical apparatus for identification. Only by analyzing color information do we
subsequently decide whether we are to continue.
In the industry, color processing is gaining a greater importance in digital signal processing.
With technological advancement optical sensors for reading operations in the visible spectrum with inte-
grated electronic readout circuits are getting affordable, that’s why we chose to use a low cost color light to
frequency converter to build a color sorting machine.
Our project was made in for pedagogical use but a further development using robust materials and compo-
nents could lead to an industrial color sorting machine for a use in the color sorting industry of recyclable
materials.
Keywords : Color sorting machine - Programmable color light to frequency converter - TCS230 - Op-
tical detector - Smart sensor - Arduino Robot - Colorimetry
1 Introduction
Humans have been quite interested in color for many
centuries. However, the scientific beginning of color
studies goes back only to Newton when he performed
his classic experiment with a prism.
Light is a narrow range of electromagnetic waves that
the eye can detect. Electromagnetic waves can have
many different wavelengths and frequencies in a range
known as the electromagnetic spectrum. Different
wavelengths of light produce different perceptions of
color. The longest wavelengths produce the percep-
tion of red, while the shortest ones produce the percep-
tion of violet. The spectrum in the visible, ultraviolet
(UV), and infrared (IR) regions is classified in a spec-
ified way.
The measurement of color is a very important issue
with applications in many types of industrial activi-
ties.
The light to frequency converter TCS230 is a small,
highly integrated color sensing device packaged in a
clear plastic 8-pin. It reports, as analog frequency, the
amount of short-wave (blue), medium-wave (green),
long-wave (red), and wideband (white) optical power
incident onto the device. It can be used in a variety of
color sensing applications.
In our case we use it to command a smart color sort-
ing machine with Arduino Nano controller. We tried
to make the most compact prototype and give it a prac-
tical use. We are going to introduce you how we made
a color sorting machine that which sorts bottles caps
in different boxes.
2 Materials and methods
In our project we only use Arduino Software IDE, but
a multitude of hardware components such as :
Light to frequency converter TCS230 :
The TCS3200 programmable color light-to-frequency
converter combines configurable silicon photodiodes
and a current-to-frequency converter on a single
monolithic CMOS integrated circuit. The output is
a square wave with frequency directly proportional to
light intensity (irradiance). The full-scale output fre-
quency can be scaled by one of three preset values via
two control input pins. Digital inputs and digital out-
put allow direct interface to a microcontroller or other
logic circuitry. Output enable (OE) places the output
1

2. in the high-impedance state for multiple-unit sharing
of a microcontroller input line. In the TCS3200, the
light-to-frequency converter reads an 8*8 array of pho-
todiodes. Sixteen photodiodes have blue filters, 16
photodiodes have green filters, 16 photodiodes have
red filters, and 16 photodiodes are clear with no fil-
ters. The four types (colors) of photodiodes are inter-
digitated to minimize the effect of non-uniformity of
incident irradiance. All photodiodes of the same color
are connected in parallel. Pins S2 and S3 are used to
select which group of photodiodes (red, green, blue,
clear) are active.
The benefits and features of TCS3200 Programmable
Color Light-to-Frequency Converter are that it :
- Enables High-Resolution Conversion of Light Inten-
sity to Frequency.
- Disables the Output into a Hi-Impedance State when
OE Input Pin is Low.
- Enables Output Range to be Optimized for a Variety
of Low-Cost Measurement Techniques.
- Reduces Board Space Requirements while Simplify-
ing Designs.
- Nonlinearity Error Typically 0.2 at 50kHz.
- Communicates Directly with a Microcontroller.
The functional blocks of this device are shown below
in figure 1:
Figure 1: Functional blocks of the TCS3200
And the TCS3200 pin assignments are described
below in figure 2: figure 1:
Figure 2: TCS3200 pin assignments
Arduino Nano
Arduino Nano is a small microcontroller good for
compact prototypes, it comes with a crystal oscillator
of frequency 16 MHz. It is used to produce a clock of
precise frequency using constant voltage.
Two Sg90 servo motor
We chose this specified servo motor because it is the
lightest on sale. It weights 90g and its engine couple
is up to 1,6 kg.cm on 4,8 Vcc ideal for our compact
prototype.
we use two servo motors one on the top and the 2nd at
the bottom. The top servo is used to bring the bottle
caps under our sensor, and the second servo is used to
deliver the bottle cap to the exact box of it’s color.
5V AC power battery
We chose to make an on-board power supply system
to power our electronics easily anywhere we want to
expose our prototype.
Switch
To disconnect or connect the conducting path in an
electrical circuit, interrupting the electric current or di-
verting it from one conductor to another and make the
machine ON or OFF.
Breadboard
The breadboard lets you easily insert electronic com-
ponents to to build and test an early version of the elec-
tronic circuit of the prototype.
Wires
It is necessary to connect our components in order to
make communication.
We connect the components together according to the
following circuit schematics in figure 3:
Figure 3: Schematics of the color sorting machine
The most important part beside the schematics is
how we calibrate the program to order the servo motor
to move for each color a specified direction.
So we chose 3 different colors bottle caps, red, white
and blue.
2

3. We pass each color under the sensor different times
and each time in the arduino real time monitor we reg-
ister the RGB colors variations for each bottle cup. By
recovering these datas we can put in our program spec-
ified RGB variation for each bottle cap color so the
servo motor moves to the correct wanted position each
time the sensor detects a bottle cap.
3 Results
The final machine prototype made is illustred in fig-
ure 4: The manner how it mechanically works is that
nitially, the colored bottle caps which are held in the
charger drop into the platform attached on the top
servo motor.
Then the servo motor rotates and brings the colored
bottle caps to the color sensor which detects its color.
After that the bottom servo motor rotates to the partic-
ular position and then the top servo motor rotates again
till the skittle drop into the guide rail.
Figure 4: Our final color sorting machine prototype
4 Conclusion
An automated color sorting machine was made at the
end of this project. Using color to frequency converter
sensor technology we learned how to make a low cost
equipment.
It’s important to admire how much sensor supplements
technology has developed in the last decade, and it’s
all fun to learn how to deal with it.
Color perception is so important for human interac-
tions, so as the industry. So making intelligent auto-
mated machine that can do tasks alone is a big invest-
ment of time for industries.
As many applications in the future require sensor
systems we want to participate in the creation of those
systems, and we won’t be able to achieve our goals
unless we master a variety of sensors.
Acknowledgement
The authors would like to express their special thanks
of gratitude to mr Dejan Nedelkovski founder of ’How
to mecatronics’ who made this work open source to
everyone wanting to learn about it.
3

4. References
Gaurav Sharma 2003, Electrical engineering and applied signal processing series, Digital Color Imaging Hand-
book
Andreas Koschan, Mongi Abidi 2008, Digital color image processing
Christine Fernandez - Maloigne 2013, Advanced Color Image Processing and Analysis
Asch, Georges 2015, Acquisition de donnees : Du capteur aa l’ordinateur
David C. Swanson 2000, Signal Processing for Intelligent Sensor Systems
Forrest M. Mims III 1996, Engineer’s Mini-Notebook: Sensor Projects
Daniel Malacara 2011, Color Vision and Colorimetry THEORY and APPLICATIONS
4