You are on page 1of 5

Fruit Quality Management using Image Processing

Shivang Mishra, Saurav Kumar, Sahil Jaglan


E.C.E Department, UIET, Panjab University, Chandigarh

ABSTRACT
Non-destructive analysis refers to the surface testing of fruits and vegetables without any intrusive
technique affecting the food aspect and quality. The non-destructive assessment methods supply
data on food characteristics such as structure, mechanical, physical, and chemical properties.
Customarily reviewing of natural products is performed fundamentally by visual examination
utilizing size as a specific quality trait. Picture handling offers an answer for mechanized organic
product size evaluating to give precise, solid, reliable and quantitative data aside from taking care of
enormous volumes, which may not be accomplished by utilizing human graders. This paper
presents a natural product size identifying and evaluating framework dependent on picture handling.
The early appraisal of organic product quality requires a new apparatus for shading estimation.
Subsequent to catching the natural product side view picture, a few organic product characters are
removed by utilizing identifying calculations. These characters help in the realization of grading.
The analysis shows this implanted evaluating framework has the benefit of high precision of
reviewing, high speed and minimal effort. It will have a decent prospect of use in organic product
quality identifying and evaluating regions.
Keywords: embedded system, size detecting, fruit grading, image processing

INTRODUCTION
So as to improve natural products' quality and creation productivity, lessens the work power, it is
important to look into non-destructive programmed recognition innovation. Organic product non-
destructive identification is the procedure of identifying natural products' inside and outside quality
with no harm, utilizing some distinguishing innovation to make assessment concurring some
standard principles. These days, the nature of organic product shape, default, shading and size, etc
can't be assessed on line by the conventional strategies. With the advancement of technology
especially in image processing, it turns out to be increasingly appealing to distinguish organic
products' quality by utilizing vision detecting innovation. At present, most existing organic product
quality distinguishing and reviewing framework have the drawback of low proficiency, low speed
of reviewing, significant expense and intricacy. So it is critical to bring fast and low-cost organic
product size distinguishing and reviewing framework in the market.
Here two decisions are accommodated evaluating either by shading and size. In the first case, we
are going to sort round formed organic products concurring shading and evaluating is finished by
size. The proposed mechanized characterization and evaluating framework are intended to join
three procedures, for example, include extraction, arranging as indicated by shading and evaluating
as per size. Programming advancement is exceptionally significant in this color classification
framework and for discovering the size of an organic product. The whole framework is planned
over MATLAB programming to review the shading and size of the natural product. Work in this
paper is carried for apple having extraordinary highlights like apple is red and framework can sort
and grade the natural products as indicated by various qualities, for example, shading and size. It
predominantly contains four sections: the framework outline, natural product size recognizing and
evaluating, trials and results, and conclusion.
RELATED WORK

Hongshe Dang, Jinguo Song, Qin Guo [1] have proposed fruit size detecting and grading system
based on image processing. The system takes ARM9 as main processor and develops the fruits size
detecting program using image processing algorithms on the QT/Embedded platform. Authors in
[2] have proposed system which finds size of different fruits and accordingly different fruits can be
sorted using fuzzy logic, here author proposed MATLAB for the features extraction and for making
GUI. John B. Njoroge. Kazunori Ninomiya. Naoshi Kondo and Hideki Toita [3] have developed an
automated grading system using image processing where the focus is on the fruit’s internal and
external defects. The system consists of six CCD cameras. Two cameras are mounted on the top,
two on the right and another two cameras mounted on the left of the fruit. X-ray imaging is used for
inspecting the biological defects. Image processing is used to analyse the fruit’s features; size,
color, shape and the grade is determined based on the features. The developed system is built from a
combination of advanced designs, expert fabrications and automatic mechanical control. J. V.
Frances, J. Calpe, E. Soria, M. Martinez, A. Rosado, A.J. Serrano, J. Calleja, M. Diaz

PROPOSED WORK

Methodology:

This mechanized framework is intended to defeat the issues of manual procedures. Its hardware
system contains a conveyor belt for transport framework, grading assembly, computerized camera,
IR sensor, Arduino Uno and grading assembly.
The picture could be taken utilizing a normal computerized camera. Here we have utilized for
catching picture the i-ball twist cam which is CMOS based camera. The framework comprises a few
stages like feature extraction, sorting, and grading. The black background helps in distinguishing
the products' features easily so black color is used for the conveyor system for grading using first
choice camera position is adjusted in such a way that for capturing live image of a fruit the camera
is continuously scanning the conveyor belt in video mode, when conveyor stops as fruit is detected
by IR system camera can capture top view image of fruit. The captured image is given as an input to
the MATLAB software which extracts (detects) color and size of a fruit, this data is transferred to
ARM-based system using com port and accordingly control action is taken place, later conveyor
starts and Then the fruit is collected in the main plate of grading assembly.
In the process, fruit color is detected according to RGB values, here fruits are sorted according to
color and size. So for e.g. two fruits are considered to say apple having red color and
guava having a green color, so in this step, the framework is going to find out the color of fruit by
using RGB values of an image taken from the camera, this image can be processed by using
MATLAB software and accordingly color can be detected i.e.red.

Working:

Using MATLAB all of the devices (Servo Motor, DC Motor, IR Sensor, Arduino UNO, Webcam)
are configured. First the DC motor starts rotating and moves the conveyor belt. As the apple
approaches the camera placed above the belt, it is sensed by the IR Sensor placed on the side of the
conveyor belt. This gives the instruction to the Arduino to stop rotating DC motor after a specific
period of time and allows the webcam to take snapshot of the fruit. This snapshot is sent to the two
algorithms to extract the mean red color of apple and to extract the diameter of the fruit(apple). If
these conditions are not met fruit is rejected from the conveyor belt by rotating the arm attached to
the servo motor which pushes the fruit away to discard bin and the good fruit moves on in the
conveyor belt.
Fruit Size Detection Algorithm

In order to calculate this diameter, the fruit’s natural symmetry is considered, so the fruit size
detecting algorithm based on its symmetry mainly contains two parts: finding the center coordinates
of fruit’s shape in image and finding fruit’s axis in image. To find fruits size first the RBG image is
converted to grey image which is converted to binary image. Using the ‘regionprops’ instruction the
binary image of fruit is assumed to be a circle.The ‘EquivDiameter’ of the the circle is calculated
by regionprops and displayed on screen.

Fruit Color Detection Algorithm

In the process of fruit color is detected according to RGB values, here fruits are sorted according to
color and size. In this step work is going to find out color of a fruit by using RGB values of an
image taken from the camera, this image can be processed by using MATLAB software and
accordingly color can be detected.
After extracting the values RGB then it is compared with the threshold and if the required criterion
is met then the fruit is passed on the conveyor belt.

APPLICATIONS

 The inner contents of fruit could also be analysed for exotic class fruits using X-rays.
 Shelf life of various food products can be estimated using moisture sensor algorithm.
 For better accuracy we can use open CV instead of MATLAB.
 3 to 4 cameras can be used to create 3-D image of fruit and then image processing can be done
on the whole fruit and not just on one part
 Microwave Radar technology can be used to detect smallest holes in food product.

ADVANTAGES

 High Efficiency
 Low Cost
 Easy to implement in existing industries
 High Accuracy
 High Speed
 No more reliability of manual labour as system is consistent and reliable and can handle large
volumes
DISADVANTAGES

 Needs Electricity supply as long as it is working which is matter of concern in remote or rural
areas.
 Maintenance can be expensive if a failure occurs.
 Occupies space and is heavy due to conveyor belt.
 Complexity

CONCLUSION

The proposed framework is a prototype, so for a huge scale creation, the number of cameras and length of
the transport framework can be altered. This work displays new incorporated systems for arranging and
reviewing of various organic products. For the most part, image capturing is a major test as there is a chance
of high vulnerability because of the outer lighting conditions, so we are exploiting the grey scale image
which is less affected to the outside condition changes and is just as helpful for discovering the size of an
organic product. In the same way, the fruit collecting system over the conveyor belt is somewhat unstable, so
further the system can be modified to overcome this problem. The speed and efficiency of the framework can
be further improved by utilizing an ARM9 or ARM11 processor for a similar cause.

REFERENCES
Hongshe Dang, Jinguo Song, Qin Guo, “A Fruit Size Detecting and Grading System Based on
Image Processing,” 2010 Second International Conference on Intelligent Human-Machine Systems
and Cybernetics,pp83-86.
[2] Harshavardhan G. Naganur, Sanjeev S. Sannakki, Vijay SRajpurohit, Arunkumar R, “Fruits
Sorting and Grading usingFuzzy Logic,” International Journal of Advanced Research inComputer
Engineering & Technology (IJARCET) Volume 1,Issue 6, August 2012,pp 117-122.
[3] John B. Njoroge. Kazunori Ninomiya. Naoshi Kondo andHideki Toita, “Automated Fruit
Grading System using ImageProcessing,” The Society of Instrument and Control
Engineers(SICE2002), Osaka, Japan, August 2002, pp 1346-1351.
[4] J. V. Frances, J. Calpe, E. Soria, M. Martinez, A. Rosado, A.J.Serrano, J. Calleja, M. Diaz,
“Application of ARMA modeling to the improvement of weight estimations in fruit sorting and
grading machinery,” IEEE 2000, pp 3666-3669
[5] Wong Bing Yit, Nur Badariah Ahmad Mustafa, ZaipatimahAli, Syed Khaleel Ahmed, Zainul
Abidin Md Sharrif, “Design and Development of a Fully Automated Consumer-based Wireless
Communication System For Fruit Grading”, ISCIT 2009 , pp 364- 369.
[6] Naoshi Kondo, “Fruit Grading Robot”, Proceedings of the2003 IEEE/ASME International
Conference on AdvancedIntelligent Mechatronics (AIM 2003), pp 1366-1371. [7] R. C. Gonzalez,
R. E. Woods," Digital Image Processing", Pearson Education.IIEd.,2002

You might also like