International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056

Volume: 06 Issue: 04 | Apr 2019 www.irjet.net p-ISSN: 2395-0072

Automatic Object Sorting Machine

Prof. S. K. Latad1, Pallavi Dhole2, Rupali Sonare3, Akshay Kukadkar4, Pawan Mankar5
1,2,3,4,5Prof.Ram Meghe Institute of Technology and Research Badnera. Amaravati
---------------------------------------------------------------------------***-------------------------------------------------------------------------
Abstract - Sorting of products is a very difficult industrial process. Continuous manual sorting creates consistency issues. This
paper describes a working prototype designed for automatic sorting of objects based on the color. TCS230 sensor was used to
detect the color of the product and the PIC16F628A microcontroller was used to control the overall process. The identification of
the color is based on the frequency analysis of the output of TCS230 sensor. One conveyor belts were used, it controlled by
separate DC motors. The belt is for placing the product to be analyzed by the color sensor, having separated compartments, in
order to separate the products. The experimental results promise that the prototype will fulfill the needs for higher production
and precise quality in the field of automation.

Keywords: Color sorting, Conveyor belt, DC motor, PIC 16F628A, TCS230 color sensor.

1. Introduction

Machines can perform highly repetitive tasks better than humans. Worker fatigue on assembly lines can result in reduced
performance, and cause challenges in maintaining product quality. An employee who has been performing an inspection task
over and over again may eventually fail to recognize the color of product. Automating many of the tasks in the industries may
help to improve the efficiency of manufacturing system. The purpose of this model is to design and implement a system
which automatically separates products based on their color. This machine consists of three parts: conveyor belt, color
sensor, and dc motor. The output and input of these parts was interfaced using PIC microcontroller. To reduce human efforts
on mechanical maneuvering different types of sorting machines are being developed. These machines are too costly due to
the complexity in the fabrication process. A common requirement in the field of color sorting is that of color sensing and
identification.

2. Literature Survey

Color sensor systems are increasingly being used in automated applications to detect automation errors and monitor quality
at the speed of production line. They are used in assembly lines to identify and classify products by color. The objectives of
their usage include to check the quality of products [1-3], to facilitate sorting and packaging [4-6], to assess the equality of
products in storage [7,8], and to monitor waste products [9]. Consequently, there is an abundance of color sensors and the
choice is often application-driven [10, 11]. Low cost and simple color sensors are preferred over sophisticated solutions for
less demanding applications where the top priority is cost and power consumption. Color names can be used and conjure
reasonably consistent perceptions. There have eleven basic color names that have been identified such as white, gray, black,
red, yellow, green, blue, orange, purple, pink, and brown. Most or all colors can be described in terms of variations and
combinations of these colors [12].

Due to the fact that human color vision is accomplished in part by three different types of cone cells in the retina, it follows
that three values are necessary and sufficient to define any color. Color theory describes that there are three values that can
be thought of as coordinates of a point in three-dimensional space, giving rise to the concept of color space. Hue, saturation,
luminance is one such color co-ordinate system, or color space.

Beckman, et al [13] described in the article about automated color-sorting using optical technology that has evolved from
early designs intended to remove ceramic contaminants. The system configuration is similar to automated ceramic removal
equipment, but color-sorting equipment used a different light source. Automated systems can generally be instructed to
remove any one or a combination of the three glass colors.

Industrial applications require some sort of automated visual processing and classification of items placed on a moving
conveyor. Bozeman and Yal-cin [14] state that items may be randomly positioned and oriented while moving on a conveyor.

© 2019, IRJET | Impact Factor value: 7.211 | ISO 9001:2008 Certified Journal | Page 3259
 International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 06 Issue: 04 | Apr 2019 www.irjet.net p-ISSN: 2395-0072

A camera located above the conveyor views the items orthographically. Boukouvalas et al [15] describes an integrated system
developed for the detection of defects on color ceramic tiles and for the color grading of defect-free tiles.

The integrated system developed under the ASSIST project (automatic system for surface inspection and sorting of tiles) is
used for the detection of defects on color tiles and for the color grading of defect-free tiles. Many have proposed advanced
solutions for the sorting of recyclable packaging towards process automation. Matt one et al had explained about a technique
for detecting and classifying objects. Most of the authors prefer to use 2D Vision techniques to separate the objects from the
known belt background and to get some of their geometrical parameters.

3. Proposed System

The proposed system is designed for automatic sorting of Red or Green or Black colored products. The prototype consists of
two DC motors, two conveyor belt, a PIC and a color sensing circuit using TCS230 (Fig. 3). DC motors are used to control the
conveyor belts. After integrating the programmed PIC and the TCS230 circuitry with the structure of the model, we measure
the frequency of signals corresponding to each color by observing them on a CRO. Based on this study the timer delay value is
adjusted by reprogramming the PIC. The time required for the product to reach the corresponding container in the separator
placed on second conveyor belt is also considered. L293D Hybrid IC is used to drive the second motor both in clock wise and
anti-clock wise direction, which provides the to and fro movement of the container of dimensions 9cm x 30cm x 7cm (Width x
Length x Height). Separators were used to create compartments of equal sizes meant for collecting objects of same color. The
end section consist of a DC motor (12V, 30rpm), which is used to control the movement of the second conveyor belt in order
to position the separator according to the sensor output. The whole framework, excluding the DC power supply, weighed
1.5kg approximately and was 27cm tall, 35cm wide and 77cm long.

4. Specifications

4.1. TCS230 Color Sensor

The TCS230 is a programmable color sensing module equipped with GY-31 light-to frequency converter that combines
configurable 8x8 silicon photodiode array as single monolithic CMOS integrated circuit. The output is a square wave (50
percentage duty cycles) with frequency directly proportional to light intensity (irradiance). The full scale output frequency
can be scaled by one of three preset values via two control input pins. Digital inputs and digital output allow direct interface
to a microcontroller or other logic circuitry. Output enable (OE) places the output in the high impedance state for multiple
units sharing of a microcontroller input line. The light-to-frequency converter reads an 8 x 8 array of photodiodes. Sixteen
photodiodes have blue filters, 16 photodiodes have green filters, 16 photodiodes have red filters, and 16 photodiodes are
clear with no filters. The four types (colors) of photodiodes are inter-digitized to minimize the effect of non-uniformity of
incident irradiance. All 16 photodiodes of the same color are connected in parallel and which type of photodiode the device
uses during operation is pin-selectable. Photodiodes are 120 mm x 120 mm in size and are on 144-mm center.

Fig. Color sensor TCS230

© 2019, IRJET | Impact Factor value: 7.211 | ISO 9001:2008 Certified Journal | Page 3260
 International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 06 Issue: 04 | Apr 2019 www.irjet.net p-ISSN: 2395-0072

4.2 Dc Motor

DC motors are electric motors that are powered by direct current (DC), such as from a battery or DC power supply. Their
commutation can be brushed or brushless. The speed of a brushed DC motor can be controlled by changing the voltage alone.
By contrast, an AC motor is powered by alternating current (AC) which is defined by both a voltage and a frequency.
Consequently, motors that are powered by AC require a change in frequency to change speed, involving more complex and
costly speed control. This makes DC motors better suited for equipment ranging from 12VDC systems in automobiles to
conveyor motors, both which require fine speed control for a range of speeds above and below the rated speeds.

When selecting DC motors, industrial buyers need to identify the key performance specifications, determine design and size
requirements, and consider the environmental requirements of their application. This selection guide is designed to help with
this process.

Fig. Dc Motor

5. Methodology

When a supply of 3.4V is given to the DC motor (12V, 3.5rpm) it starts to rotate. It will control the movement of the conveyor
belt on which the product is placed. When the light falls on the product it is reflected back to the color sensor. As mentioned
before, color sensor TCS230 has 4 color filters for green, red, blue and black (no color), which is opted by its select pins.
Filters are selected by the program saved in the microcontroller. Frequency output from color sensor depends on the color of
the object as well as the select pin configuration input from microcontroller.

Select pin can select one of the four photo diode filters which can give output according to the color of the object. When there
is no object in front of sensor it produces an output of 330Hz range frequency and when there is an object it produces an
output frequency of 7-14 KHz. If there is an object the sensor produces an output frequency which is proportional to the color
of the object and the selected photo diode configuration in such a way that it provides maximum frequency for the respective
color to the respective photo diode. Hence sensor gives maximum frequency for red colored object when red filter is selected,
and in the same way other colored object are also sensed by corresponding filters.

Frequency received during each filter selection is counted and saved to separate registers and these values are examined for
taking the greater one, in order to identify the color of the object. The container has three sections; first section for Green,
middle for Black, and third for Red. According to the color, the container will be moved in forward or backward direction by
the conveyor belt, which is made possible by connecting the DC motor to L293D hybrid IC. The products will finally fall to the
corresponding sections in the container. The PIC has 18 pins, out of which five pins are connected to TCS230 color sensor,
and two pins are connected to L293D IC.

© 2019, IRJET | Impact Factor value: 7.211 | ISO 9001:2008 Certified Journal | Page 3261
 International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 06 Issue: 04 | Apr 2019 www.irjet.net p-ISSN: 2395-0072

Fig. Block Diagram

6. Application

1)Color Detection & Sorting operations like brick sorting, tablets separation, etc.

2)Process control to printed materials.

3)Ambience light detection.

4)Robotics color detection.

5)Packet sorting

6)Toys sorting

7)Bottle sorting

© 2019, IRJET | Impact Factor value: 7.211 | ISO 9001:2008 Certified Journal | Page 3262
 International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 06 Issue: 04 | Apr 2019 www.irjet.net p-ISSN: 2395-0072

7. Result

8. Conclusion

As mentioned before we have used objects of standard size and having any of the three colors for testing our
prototype. We conducted a continuous trail with 100 objects and we got 100% correct detection. As long as the colors
of the objects do not deviate from the preset values and as long as the placement on the belt is perfect, the detection
process seldom fails.

9. Future Scope

It is very useful in wide varieties of industries along with the help of PLC and SCADA, especially in the packaging
section. Automatic sorting machine enhances efficiency, practicality, and safety of operators. It ensures remarkable
processing capacity as well as peerless performance including color detection. Of course we need to add high speed
DC motors and sensors with appreciable response to speed up the system for industrial application. The model can be
improved by making some changes in the program and components. Some suggestions are given below.

•We can also add a counter for counting the number of products

•We can add a load cell for measurement and control of weight of the product

•The sensor can be changed according to the type of product

•The PIC can be replaced with PLC

•Speed of the system can be increased accounting to the speed of production

•The system can be used as a quality controller by adding more sensors

•The DC motor can be replaced with stepper motor

© 2019, IRJET | Impact Factor value: 7.211 | ISO 9001:2008 Certified Journal | Page 3263
 International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 06 Issue: 04 | Apr 2019 www.irjet.net p-ISSN: 2395-0072

9. References

[1]C. Zhang and K. S. Seasick, “Colorimetric sensor array for soft drink analysis”, J. Agric Food Chem., vole 55, pp 237-
242, 2007.

[2]P. D. Oram and J. Strine, “Color measurement of a solid active pharmaceutical ingredient as an aid to identifying key
process parameters”, J. Pharm Biomed Anal, vol 40, pp 1021-1024, 2006.Author, Title of the Book, Publisher, (200X),
pp:XXX-YYY.

[3]M. Frank, Nobert Kaiser, Wolfgang Buss, Ramona Eberhardt, “High-speed industrial color and position sensors”,
Electronic Imaging’99, pp 50-57, 1999.

[4]A. Acquit, J. Frisbee, Danny Diamond, King Tong Lau, Alan Farrell, Bird Quilt, Dermot Diamond,

“Development of a smart packaging for the monitoring of fish spoilage”, Food Chem., vole 102, pp 466-470, 2007.

[5]E. J. Cadieux Jr, “System and method for visually inspecting a cigarette packaging process”, ed:

Google Patents, 2002.

[6]D.J. Lee and R. S. Anbalagan, “High-speed automated colorsorting vision system”, in Optical Engineering
Midwest’95, pp 573-579, 1995.

[7]J. V. Popov-Ralji, et al, “Investigations of bread production with postponed staling applying instrumental
measurements of bread crumb color”, Sensors, vol 9, pp 8613- 8623, 2009.

[8]J. V. Popov-Ralji and J. G. Lalii-Petronijevi, “Sensory properties and color measurements of dietary chocolates with
different compositions during storage for up to 360 days”, Sensors, vol 9, pp 1996-2016, 2009.

[9]R. Baribeau, “Color reflectance modeling using a polychromatic laser range sensor”, IEEE T Pattern.

Anal., vol. 14, pp. 263- 269, 1992.

[10]H. Escid, et al., “0.35 mm CMOS optical sensor for an integrated transimpedance circuit”, the International Journal
on Smart Sensing and Intelligent Systems, vol. 4, no. 3, pp. 467481, September 2011.

[11]Norfazlinda Binti Daud, ”Application of colors sensor in an automated system”, Technical University Malaysia,
May 2007.

[12]Bickman, Josh, “Automated Color-Sorting uses optical technology”, vol. 13, 1996.

© 2019, IRJET | Impact Factor value: 7.211 | ISO 9001:2008 Certified Journal | Page 3264