I. INTRODUCTION:

The quality of the fruits is important for the consumers and become there requirement from the suppliers to provide fruits with high standards quality. So, in the past few years, fruit grading systems have established to fulfill the needs of the fruit processing industry inspection .Besides that, the process of fruits involves several steps that can generally be classified into grading, sorting, packaging, transporting and storage. The grading is considered as the most important steps towards the high standard of quality .Fruits are almost graded manually which is an expensive and time consuming process and labors shortage will affect to the operation during peak seasons. It has become increasingly difficult to hire or train the person who

are willing to handle the monotonous ask of inspection. In the mean while, a cost effective and accurate grading can be performed with automated grading system. Generally, the fruits quality depends on outer parameters like size, color intensity, shape, surface appearances and inner parameters (sugar contents, acid contents) but color and size is the most important factor for grading and sorting of fruits. The fruit grading system is accomplished based on weight, color and size which are accessible in all fruit processing industries. The fruit grading system techniques using computer machine vision and image processing play the important role of quality control in fruit processing industries. From the past few years, different techniques have been enhanced to grade and evaluate the quality of fruits. These methods can help to detect different physical properties of fruits and with certain quality factors. For example, the vision-based systems include CCD or CMOS sensors that are used to estimate the size and shape of fruits. It helps to predict the size of the fruits from its RGB image frame with the help of CCD camera. Software plays an important role in this color classification system. Fruit’s quality is being improved and production efficiency, reduce labor intensity, it is necessary to research non-destructive automatic detection technology. The quality of fruit shape, color and size cannot be evaluated only by the traditional methods. Image processing technology and computer software and hardware; it becomes more attractive to detect fruit's quality by using vision detecting technology. At present, most existing fruit quality detecting and grading system have the disadvantage of low efficiency, low speed of grading, high cost and complexity. High speed and low cost fruit size detecting and grading system is developed. Choices provided for grading are either by color and size. Circular shaped fruits color and grading is done in first case, according to size. Processes such as feature extraction, sorting according to color and grading according to size are classified under grading.

II. LITERATURE SURVEY:

Hongshe Dang, Jinguo Song, Qin Guo[1] have proposed fruit size detecting and classification system based on image processing. The system takes ARM 9 as main processes and develops the fruits size detecting the QT/Embedded platform. Authors in[2] have proposed system in which 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 Kondoand, Hideki Toita[3] have developed an automated focus on the fruit’s internal and external defects. The system consists of six CCD cameras. On the right two cameras are mounted and another two cameras mounted on the left of the fruit. X-ray imaging is used for inspecting the biological defects. Digital Image processing is employed to research the fruit’s features; size, color, shape and the grade is determined based on the features. The developed system is constructed from a mixture of advanced designs, professional fabrications and automatic mechanical management. J.V. Frances, J. Calpe, E. Sofia, M. Martinez, A. Rosado, A.J .Serrano, J. Calera, M. Diaz[4] presented a procedure to improve the performance, whether increasing speed or accuracy, of the load-cell-based weighting sub-system in a fruit sorting and grading machine to achieve an accuracy of kilogram. Wong Bing Yit, Nur Badariah Ahmad Mustafa, Zaipatimah Ali, Syed Khaleel Ahmed, Zainul Abidin Md Sharrif[5] proposed new MMS-based system design and developed with signal processing for fruit grading for consumers. The prototype network architecture, integration of wireless messaging system with signal processing between mobile consumers for development functions was studied, planned and designed.

III. SYSTEM MODEL AND METHODOLOGY:

It's somewhat troublesome to find fruit form, size and color due to poor existing methods. Currently method of exploitation vision technology is used. At present, most existing fruit quality and grading system have the disadvantage of low potency, low speed of grading, high value and complexness. Thus it's important to develop high speed and low value fruit and grading system. Fruits recognition techniques which combine four options analysis techniques that are shape, size, color and texture based morphological techniques, to extend accuracy of recognition. The main connotation of this paper is to introduce a replacement epitome for the detection of fruit grading in assortment of fruits. The image processing is a form of signal processing where the input is an image, like a photograph or video frame, the output of an image processing may be either an image or a video frame or a set of characteristics or parameters related to the image. The acquisition of digital image usually suffers from undesirable camera shakes and due to unstable random camera motions. Hence image enhancement algorithms are required to remove these unwanted camera shakes. This image processing concepts are implemented in Raspberry pi in the application of MAV. Raspberry Pi is an embedded system containing a credit card-sized single board computers developed in the UK by the Raspberry Pi Foundation. The Raspberry Pi is based on the Broadcom BCM2835 system on a chip (SOC) which includes an ARM1176JZF-S Core (ARM V6K)700 MHz CPU processor, Broadcom Video Core IV GPU having 17 pins, 3.5W of power, and 512 MB of RAM memory. The Raspberry Pi system has Secure SD card reader (models A and B) or Micro SD card reader (models A+ and B+) sockets for boot media and persistent storage. The system provides Debian Linux operating system Raspbian image for download. Python is used as main programming language for raspberry pi. MAV is a remote-controlled, Unmanned Aircraft Vehicle (UAV) significantly smaller than typical UAVs that have a size restriction. UAV is an aircraft without a human pilot. It is controlled autonomously on board computers or by the remote control of a pilot on the ground or in another vehicle. Raspberry Pi cam module present on a MAV the efficiency of this air vehicle increases and new fields of applications are available. It is needed in military Operations, in which targets have to be identified. Classification is often done by a human on ground, to minimise the probability of mistakes. But a Raspberry Pi camera module is also helpful if a MAV shall autonomously fly through an arch. The main signal processing chip unit used in Raspberry Pi system is a Broadcom 2835 700MHz Chip in which CPU core is a 32 bit ARM1176JZF-S RISC processor designed by Advanced RISC Machines. This main processing chip joints a camera and display. The Raspberry Pi design does not include a built in hard disk or solid state drive, instead used an SD card for booting and long term storage. This board is intended to run Linux Debian based operating systems. This Raspberry Pi module has a Samsung class 4 micro SD card preloaded with the Raspberry Pi NOOBS (New Out of Box Software) package, and a printed Micro SD card adaptor.

Fig 1: Work flow of fruit grading system

IV. HARDWARE APPROACH:

The sequence of elements and ideas needed during this planned system measure are Raspberry Pi 3, Web Camera, IR Sensor, Motor and Relay.

Fig 2: Hardware Board Devices

a). Raspberry Pi 3:

A budget desktop, however expecting the Pi 3 to match a typical PC, it will lag loading heavier websites and, when browsing demanding sites, having more than a handful of windows open at a time runs the risk of overloading the Pi's memory--causing a lengthy freeze. Everything from loading web pages is to alt-tabbing between applications taking slightly longer than used to. Also, while running every application, one need to rely on web apps. Using more specialized local applications could find they're not supported on the Pi's predominantly Linux-based OS. However, the Pi works well as a thin-client, as when tested it is running as a thin client for Windows 10. There are various options for using the Pi 3 as a media center but the most popular choices are the Kodi-based OSes OSMC or LibreElec. It has the advantage of a faster graphics processor, which the Raspberry Pi Foundation has said is able to play local H.264-encoded video recorded at 1920x1080 resolution and 60 frames per second. Another advantage is built-in support for Wi-Fi makes it easier to stream content to the Pi, while native Bluetooth simplifies the hooking up peripherals. The Raspberry Pi 3's specification are: Chipset-Broadcom BCM2837,CPU-1.2GHz quad-core 64-bit ARM cortex A53, Ethernet-10/100 (Max throughput 100Mbps), USB-Four USB 2.0 with 480Mbps data transfer, Storage- Micro SD card or via USB-attached storage,Wireless-802.11n Wireless Local Access Network (Peak transmit/receive throughput of 150Mbps), Bluetooth 4.1, Graphics- 400MHz Video Core IV multimedia, Memory-1GB LPDDR2-900 SDRAM, Expandability-40 general purpose input-output pins, Video-Full HDMI port, Audio- 3.5mm audio out jack and composite video, Camera interface (CSI), Display interface (DSI).According to tests, the peak power consumption of the Pi 3 when under heavy load is about twice that of the Pi 2 (750mA vs 360mA), though for less-demanding workloads it should be broadly similar to earlier boards. Writing simple programs will allow you to send signals via the pins to control the attached hardware--for example making an LED flash-- or to read a signal sent from the attached hardware via the pins--for example to take a measurement from a sensor.

Fig 3: Raspberry Pi 3

b). Web Camera:

The camera plugs to the CSI connector on the Raspberry Pi. It captures clear 5MP resolution image, or 1080p HD video recording at 30fps. The camera module attaches to Raspberry Pi by a 15 pin Ribbon Cable, to the dedicated 15 pin MIPI Camera Serial Interface (CSI), which was designed especially for interfacing to cameras. The CSI bus is capable of high data rates, and it carries pixel data to the BCM2835 processor.

The method uses the raspberry pi board is the main controller. The latest version of raspbian wheezy is used on to the board. After installation of the OS to the board connect all the necessary hardware components and switch on the power supply. It starts booting up the Board and login the raspberry pi by username and password. It operates on the Linux Debian arch operating system. It mainly works on the python software and checks the network settings to update the python software by commands in the terminal window. Following packages are to be installed for implementing the proposed model.

Enabling the camera settings on the board to capture the image and save it on the folder. The python code is being run to check the enhancement algorithms and remove the noise present in an image.

Fig 4: Web Cam

c). IR Sensor:

There is a semiconductor/chip within the sensor; it has power specification of 3 - 5V to function. Distinct to photocells and FSRs, wherever they act like resistors and therefore are often merely tested with a multimeter. Pin 1 : The output therefore we tend to wire this to a visible Light-emitting diode and

Fig 4: IR Sensor

Resistor. Pin 2: Ground. Pin 3 : VCC, connect to 3-5V.The detector views IR signal, it will pull the output low, turning on the LED - since the LED is red it’s much easier for us to see than IR! We will use 4xAA 1.3V batteries (I use NiMH) so that the voltage powering the sensor is about 4V. 2 batteries (3V) may be too little. Ground goes to the middle pin. The positive (long) head of the Red LED connects to the +6V pin and the negative (short lead) connects through a 200 to 1000 ohm resistor to the first pin on the IR sensor. Grabbing any remote control like for a TV, DVD, computer, etc. terminating it at the detector while pressing some buttons, the LED blink a couple of times whenever the remote is pressed

d). Motor: Motor a class of rotary electrical machines that converts direct current electrical energy into mechanical energy. The most common types rely on the forces produced by magnetic fields. Nearly all types of DC motors have some internal mechanism, either electromechanical or electronic; to periodically change the direction of current flow in part of the motor.

Fig 6 : Motor

e). Relay:

Relay an electromagnetic device is used to isolate two circuits electrically and connect them magnetically. It allows one circuit to switch with another one while they are completely separate. It is often used to interface an electronic circuit (operating at a low voltage) to an electrical circuit which works at very high voltage. Relay switch is split into 2 parts; input and output. Input section contains a coil which generates magnetic field when a small voltage from an electronic circuit is applied to it. This voltage is called the operating voltage. Preferably used relays are available in different configuration of operating voltages like 6V, 9V, 12V, 24V etc. The output consists of contactors which connect or disconnect mechanically. In a relay there are three contactors: normally open (NO), normally closed(N and common (COM).

Fig 7: Relay

V. PERFORMANCE:

A VGA monitor is connected to raspberry Pi with the help of 15 pin VGA cable. The VGA cable is connected to the HDMI to VGA converter which is further connected to Raspberry Pi 3. A HDMI to VGA converter supports upto 1080 (60Hz) signal and needs no external power. First connecting the VGA cable to the desktops 15 pin 3 row connector, which is at backside of the desktop? The other end of the VGA cable is connected to the HDMI to VGA converter. The other end of the HDMI is connected to the HDMI connector of Raspberry Pi 3. The desktop serves as display to the Raspberry Pi 3. Through the Raspberry USB port the camera, mouse and the keyboard is connected. The relay circuit and the IR sensor is connected through GPIO pins of the Raspberry Pi 3. The motor is connected to the relay circuit. After all the devices are connected safely the power supply is given to the Raspberry Pi 3. Then the desktop is turned on. The desktop is now a display for the Raspberry Pi 3 processor. Now Python code is required for functioning of Fruit Grading process. The command prompt is opened where a python code sudo ./project.sh shown in Fig 8 is typed as python code, which is a pre-requisite stored program saved in memory chip. Which functions the windows for resulting output? A real time apple is kept before the camera which scans the fruit and shows the output in Fig 9. Figure 9 shows the output whether the fruit is diseased or good. The IR sensor counts the number of fruit according the diseased or specifications come accordingly in count. The motor turns clockwise when the fruit is good and it turns anticlockwise when it is diseased. Thus the fruit is seperated. It is an Industrial approach for grading of fruit with big Servo motors for detecting fruit.

Fig 8: Python Pseudo Code

Fig 9: Fruit quality detection

VI. CONCLUSION:

Several measures are equipped for determining good quality and sorting of fruits. It has four level of grading (Red, Orange, Green and Turning to Green) with some pictures of tomato. In this paper, the comparison between different method, by different authors and various algorithms in different ways for finding a good quality of fruits have been done. The identification of fine and defected fruits based on quality in image processing using raspberry pi 3 is successfully done with low computational time 0.52 and high accuracy 95%. The use of image processing for identifying the quality can be applied not only to fruits but also to other fruits such as oranges, apples, melons etc. and also vegetables with more accuracy. This work presents new integrated techniques grading of different fruits. So as to boost the practicality and suppleness of the system hardness, softness features can be combined together with shape, size, color and texture feature.

VII. REFERENCES:

1. Hongshe Dang, Jinguo Song, Qin Guo, “A Fruit Size Detection and Grading System Based on Image Process,” 2010 Second International Conference on Intelligent Human-Machine Systems and information processing,pp83-86.

2. Harshavardhan G. Naganur, Sanjeev S. Sannakki, Vijay S. Rajpurohit, Arun kumar R, “Fruits Sorting and Grading using Fuzzy Logic Systems,” International Journal of Advanced Research in Computer Engineering & Technology (IJARCET) Volume 1, Issue 6, August 2012, pp 117-122.

3. John B. Njoroge. Kazunori Ninomiya. Naoshi Kondo and Hideki Toita, “Automated Fruit Grading System using Image Processing,” 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, Zaipatimah Ali, 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. Wang S, Siskins J (2003). Image Segmentation with ratio Cut. IEEE Trans. Patt. Ana. Mach. Intell., 25(6): 675-690.

7. López C, Gómez P (2004). Comparison of color indexes for tomato ripening. Hortic Bras., 22(3): 1-4.

8. Sanz P, Marin R, Sanchez J (2005). Pattern Recognition for Autonomous Manipulation in Robotic Systems. IEEE Trans. Syst. Man. Cybern. C Appl. Rev., pp. 1-35.

9. Tao W, Jin H, Zhang Y (2007). Color Image Segmentation Based on Mean Shift and Normalized Cuts. IEEE Trans. Syst. Man Cybern. 37(5): 1382-1389.

10. Muratore G, Rizzo V, Licciardello F, Maccarone E (2008). Partial dehydration of cherry tomato at different temperature, and nutritional quality of the products. Food Chem., 111: 887-891.

11. Lien C, Ay C, Ting C (2009). Non-destructive impact test for assessment of tomato maturity. J. Food Eng., 91(3): 402-407.

12. W. M. Syahrir, A. Suryanti and C. Connsynn, “Color Grading in Tomato Maturity Estimator using Image Processing Technique”, 2nd IEEE International Conference on Computer Science and Information Technology (ICCSIT 2009), 2009, pp. 276-280.

13. Suresha M., Shilpa N., Soumya B “Apples Grading Based On Svm Classifier International Journal of Computer Applications” (0975 – 8878) On National Conference on Advanced Computing and Communications - Ncacc, April 2012

14. Chen, Xuming and Yang, Simon X. (2013), "A Practical Solution for Ripe Tomato Recognition and Localisation", Journal of Real-Time Image Processing, Vol. 8, No. I, pp. 35-51, January 2013.

Received on 10.06.2019 Accepted on 28.06.2019

Int. J. Tech. 2018; 9(1):27-32.

DOI: 10.5958/2231-3915.2019.00007.5

Automatic fruit Grading System