1.1. OVERVIEW OF RICE SORTING MACHINE
Among the cereals, rice is the major food stuff for a large part of the world’s population. Rice needs utmost care during post harvest handling and processing,because in most cases, it is considered as whole kernel. All varieties of rice can be processed as either white or brown rice. Milling of rice post harvest always leads to some grains being broken. Quality factor are related to grain length, stickiness, aroma, texture and flavour. Presently, rice type and quality rapidly assessed by visual inspection. This evaluation is tedious and time consuming, also, the decision taken by human inspectors may be effected by external factors like tiredness, revenge, bias or human psychological limitation this can be overdone by using rice sorting . ice processing begins in a millimg plant, where the harvested grains run through a production line where the rice is husked and shelled.
1.2 WORKING PRINCIPLE OF RICE SORTING MACHINE
Rice processing begins in a milling plant, where the harvested grains run through a production line where the rice is dried, de-stoned, husked and shelled. It is then taken to the sorter machine. At this point, the rice mixture will travel by elevator belt into a hopper on the top of the machine, from which it will flow down chutes in the sorter, streamlining their flow so that they may be scanned by the sensor.
The moment the camera detects any defects, the camera instructs ejectors fitted in the machine to open the nozzle. The nozzle is connected to the valves containing compressed air. This air is then used to shoot out the defected material from the input rice. The types of defects in rice include black tipped and partially black tipped.
1.3 RICE SORTING PROCESS
The below images shows the rice sorting process, first the rice is inserted into the machine it is then flow into the vibrator it fills the rice into the chute. The chutes ranges as 16, 20, 32, 40 and 63 the rice gets detected by the camera and at last it detect whether the given is defective or non defective one.
Fig 1.1 Inserting rice into the machine Fig 1.2 Rice flows into the vibrator
Fig 1.1 and Fig 1.2 shows that the rice is first inserted in to the machine then it may flows in to the vibrator.
Fig 1.3 Chute process till 63
Fig 1.3 shows the chute process, there are totally five types of chute to be present range: 16,20,32,40 and 63 where the rice flows in it.
Fig 1.4 Separating defective and non-defective rice
Fig 1.4 shows the separating process which is used for detecting whether the given rice is defective or non defective.
1.4 CHUTE PROCESS
In a chute rice sorting machine , rice grains drop onto and slide down on anodized aluminium chute . the purpose of the chute is to separate the grain and provide a controlled distribution. At the bottom of the chute the grains are examined optically and contaminants grains are removed from the stream by jets a air .
The machine has the ability to sort low quality rice which contains a large element of contaminants such as husk. The husk is extremely abrasive can lead to as reduction in the life of the chute by wear of the surface.
1.5 TYPES OF RICE
The rice sorting machine categorizes rice as follows
Fully defective ( Black)
Partially defective (Black and White)
Non defective (White)
Three types of rice can be detected by using this rice sorting machine, it checks whether the given rice is fully detective (black), partially defective (black and white) or non defective (white) by the use of the calibration input output, wave generation, clock generation , ADC settings, ADC configuration and ADC process.
1.6 OBJECTIVE OF THE PROJECT
This project aims to design and implement the following modules of the rice sorting machine.
Calibration input output
1.7 ORGANIZATION OF THE THESIS
The organisation of the thesis is as follows
Chapter 1 gives an introduction to the rice sorting machine. Chapter 2 gives a literature survey on various papers which are essential to know the existing rice sorting machine techniques and their significance on sorting machine. Chapter 3 lists the various modules in the rice sorting machine and chapter 4 discusses about the modules implemented in this project. Chapter 5 discusses about the results of modules implemented. Chapter 6 concludes this project work.
A literature survey is done for various papers which are essential to know the previously available techniques and their significance and limitations of sorting machine.
2.2 LITERATURE REVIEW
Md. Hazrat aii, N.Mir-Nasiri (2017), have proposed a Automated Pepper Sorting Machine (APSM) for industrial application. The design consists of several mechanical and electrical components such as motors, a microcontroller, hoper mechanism, suction tube, belt and pulley. In addition, a gantry crane is used in order to move the suction pipe across the conveyor belt to the exact position of the defective pepper. Furthermore, a camera is attached to the machine to identify defective pepper based on colour of the defective pepper. The main purpose of this design is to automate the sorting procedure for pepper industry and to optimize its work efficiency.
R.Arun,S.Vijay Rajpurohit and V.B .Nargund (2017), have proposed a neural network assisted machine vision system for sorting pomegranate fruits was proposed. Sorting is an important step in processing and packing lines of pomegranate fruits. Currently pomegranates are sorted into quality categories manually. But manual sorting poses problems such as tediousness, low accuracy, subjectivity etc. Moreover, manual sorting is not recommended for export quality fruits. Hence a machine vision system is required in order to sort the pomegranate fruits.
This paper is aimed at developing a robust non-destructive method to sort pomegranates using wavelet features and Artificial Neural Network (ANN) training. Pomegranates are sorted into `diseased’ or `healthy’ class. Initially, images of the diseased and healthy pomegranates are acquired from a local fruit market. As part of pre-processing, histogram equalization is applied followed by wavelet de-noising. The pre-processed images are then fed to a feature extraction module where 15 spatial domain features and 252 wavelet features are extracted.
Experiments were conducted to train ANN and calculate the performance based on spatial and wavelet features separately. Network performance is analyzed based on the parameters such as Sensitivity, specificity, accuracy, mean square error and Receiver Operating Characteristic (ROC) curve. The results of experimentation showed that the performance of ANN was high when wavelet features were used for training as compared to the spatial domain features
Rija Hasan,Syed Mohammed And G.Monir (2017) have proposed a Fruit maturity estimation based on fuzzy classification sign in or purchase is used to detect whether the fruit is raw and overripe. In this paper an efficient approach of fruit maturity classification based on apparent colour of the specimen is implemented by the aid of Fuzzy Inference System (FIS). Heuristically acquired hue and its corresponding saturation and lightness are the attributes of choice, which are utilized to classify the sample into three classes; Raw, Ripe, and Overripe.
The membership functions and fuzzy rules required by the Mamdani FIS are estimated by the approach of classification tree. The experimentation is performed upon 200 guava samples. The fuzzy system is trained upon 60% of the dataset, yielding 93.4% classification accuracy.
Harish S Gujjar, Dr. M. Siddappa (2015), have propose a method for an identification of basmati rice grain of India and its quality using pattern classification. Here the image warping and image analysis algorithm is used which is use to analysis the texture of the basmati rice and it is also used for identifying the give rice whether it is defective or non defective one.
This chapter discussed about the research in the design of various modules in the FPGA implementation of rice / grain sorting machines. Their pros and cons are analyzed for the implementation of the proposed modules in the rice sorting machine
MODULES IN RICE SORTING MACHINE
The different modules in the proposed rice sorting machine are listed below.
Calibration input output
The CALIBRATION INPUT OUTPUT MODULE is abbreviated as CALIB_IO Module. This is the first process to be done and is used for calculating the reference value for the chute process. The Number of chutes in the grain sorting machine are 16, 20, 32, 40 and 63. The second is the wave generation module, which is used for positioning the on and off time period of the clock. The third module is the clock generation module, and it is an instantiation process.
The ADC process compares the ADC value and calib_io value, for automatic ejection and testing process.In the ADC setting process two clocks will be produced from the input clock signal and the ADC configuration process configures the AD9826 to work properly.
This project aims to design and implement the following modules of the rice sorting machine.
Calibration input output
4.2 RICE SORTING MACHINE CALIB_IO MODULE
CALIB_IO takes place here
Fig 4.1 Rice sorting machine calibration input output module
Fig 4.1 shows the process of rice sorting machine, first the rice is inserted in to the hopper, hopper is a container for a loose bulk material such as rice or rock typically one that tapers downwards and is able to discharge its contents at the bottom. Then again the rice flow through the vibrator, the vibrator tray which is used for getting high quality sorting of rice, the flow of rice is maintained uniformly. The vibrator works consistently when the voltage is high or low.
The purpose of Chute is to separate the grains and provide a controlled distribution at the bottom of the chute the grains are examined optically and contaminants or defective grains are removed from the steam by gets of air. The grain passing through the inspection area is very closely scanned by the camera for any impurity.
The tolerance value is set , Rice enters the machine through the input hopper and is vibrated towards the anodised aluminium chute (16, 20, 32, 40 and 63) along a tray. The rice drops off the edge of the tray onto the chute and slides down. The chute has the effect of separating the grains so they arrive at the end in a continuous stream. The calibration process starts it is used for calculating the reference value at the end of the chute there is a detector head consisting of a number of cameras. An image is taken of each grain as it passes the head.
The image is quickly processed and compared to a reference standard that is used to accept or reject the grain. A series of air jets controlled by high speed poppet valves are used to blow the defective grains or contaminants out of the stream.
4.3 FLOW CHART OF CHUTE PROCESS
PIPE = PIPE+1
BIT_1 SET = 1
Fig 4.2 Flow chart of the chute process
Fig 4.2 shows the chute process of the rice sorting machine, here the tolerance value is set, when BIT_SET=1, the TADC_C TRIG_S=1 the signal becomes high and then the chute process starts it fills the chute till 63, the pipe here denotes the chute, once the chute is filled with the rice the calibration process started, it calculates the reference value, if TADC_C TRIG_S=0 the signal becomes low and the process will not run.
4.4 FLOW CHART OF CALIBRATION PROCESS
BIT_1 SET = 1
CALIBRATE TO SET THE REFERENCE VALUE
Fig 4.3 Flow chart of calibration process
Fig 4.3 Shows the flow diagram of calibration process, If the BIT_1 SET =1 the value is 1 then the signal becomes high the calibration process starts to calculate the reference value and if the BIT_1 SET= 0 the value is 0 then the signal becomes low the calibration process will not takes place it stops the process.
4.5 CHUTE VALUE CALCULATION
INPUT CHUTE VALUE
Fig 4.4 Chute value calculation
Fig 4.4 Shows the chute value calculation, the input chute value is given (totally five chute values are present namely as 16,20,32,40 and 63). The average value can be calculated as total pixels (i.e. 2048) divided by the chute value.
Average value = pixels/chute value
For example take chute value as 16 the average value can be calculated as
4.6 WAVE GENERATION MODULE
The wave generation module has the following components that are to be present in it
The counter is a digital device and the output of the counter includes a predefined state based on the clock pulse application. The output of the counter can be used to count the number of pulses
The comparator is a device that compares two voltage or current and output a digital signal indicating which is larger.
4.9 COMPARATOR PROCESS
Given, in data sheet;
Total time period = 268?s
On-time is 256?s
Off time is 12?s
Given, Frequency = 50 MHz
Time period for one cycle = 1/ 50MHz
= 20 ns
On-time period = 256 µs /20ns = 12800
Total time period = 268 µs /20ns = 13400 ,Off-time period = 13400 -12800 = 600
4.10 ON AND OFF TIME PERIOD OF THE CLOCK SIGNAL
12 µs µs268 µs
ON TIME PERIOD
OFF TIME PERIOD
TOTAL TIME PERIOD
Fig 4.5 Generated clock signal
Fig 4.5 shows the generated clock signal, the positive edge of the clock is called as ON time period and the negative edge of the clock is called as OFF time period of the clock signal. The ON time period ranges from 0 to 12800 and the OFF time period ranges from 12800 to 13400.
4.11 CALCULATING ON AND OFF TIME PERIOD
The wave generation module is use for calculating the on and off time period of the clock signal
Total time period is to be taken is 268 µs, on time period of the clock is 256 µs and then the off time period of the clock is 12 µs. The given frequency is 50MHZ. Time period for one cycle is calculated as 1 divided by 50MHZ so the value is 20ns.
4.12 ON TIME PERIOD CALCULATION
The on time period can be as calculated as follows
On time period of the clock is 256, it can be given as
On time period = 256 µs/20ns = 12800
So the on time period ranges from 0 to 12800.
4.13 OFF TIME PERIOD CALCULATION
The off period can be as calculated as follows.
Off time period of the clock is 12 µs, it can be given as
Off time period = 12 µs/20ns =13400
So the off time period ranges from 12800 to 13400
4.14 COMPARATOR PROCESS
The comparator process consists of three types it can be as given below:
The value of comparator A is assigned as 0
The value of comparator B is assigned as 12800
The value of comparator C is assigned as 13400
4.15 AND OPERATION
The wave generation module is done using the AND operation.
The AND operation is used for denoting the position 1 and position 2 of the clock signal.
Position 1 256?s (0 to 12800)
Position 2 12?s (12800 to 13400)
The Position 1 denotes the on time (256 ?s) period of the clock signal and it ranges from 0 to 12800. The position 2 denotes the off time (12 ?s) period of the clock signal and it ranges from 12800 to 13400. This is the use of AND operation.
4.16 COMPARATOR A PROCESS
A ? B (A is the value of counter output i.e. 0 to 13400), Assign B as 0
COUNTER = A VALUE (0 TO 13400)
CHECK IF A?B
Fig 4.6 Comparator A process
Fig 4.6 Shows the comparator A process the counter output value ranges from 0 to 13400. Assign B value as 0 compare A ; B if the value of A is greater than or equal to B then the process is true and it is denoted as CA1, if the value of A is not greater than or equal to B then the process is ignored.
4.17 COMPARATOR B PROCESS
Checks if A ? B or A ? B; (A is the value of counter output ie: 0 to 13400), Assign B as 0 to 12800
COUN TER = A VALUE O TO 13400)
COMPARE A ; B
CHECK IF A? B OR A?B
TRUE SAVE AS CB2
TRUE SAVE AS CB1
Fig 4.7 Comparator B Process
Fig 4.7 shows the comparator B process the counter output value ranges from 0 to 13400 the value of B is assigned as 12800. Now the value of A;B is compared and it checks the condition A? B OR A?B, if A is greater than or equal to B the process becomes true and the value is saved as CB1 and if the value of A is less than or equal to B again the process becomes true and it is saved as CB2.
4.18 COMPARATOR C PROCESS
In comparator C process Assign B = 13400
SET B= 13400
COMPARE A ; B
COUNTER VALUE A=13400
TRUE SAVE AS CC1
Fig 4.8 Comparator C process
Fig 4.8 Shows the comparator C process here the value of A is assigned as counter output value and it ranges from 0 to 13400, the value of B is set as 13400 compare A ; B if the value of A is less than or equal to B then it is true and saved as CC1 if it is not true then the value gets ignored.
4.19 AND PROCESS FOR POSITION 1
The AND process for the position 1 process is described in the block below, which denotes the ON time period.
INPUTS CA1 AND CB1
Fig 4.9 AND operation for POSITION 1
Fig 4.9 Shows the AND operation takes the input as CA1 and CB1 and it checks for the conditions, if the output value becomes 1 then it denotes the position 1 and it ranges from 0 to 12800 and if the output value is not equal to 1 then it gets ignored.
4.20 AND PROCESS FOR POSITION 2
The AND process for the position 2 process is described in the block below, which denotes the OFF time period.
INPUT CB2 AND CC1
Fig 4.10 AND operation for POSITION 2
Fig 4.10 Shows the AND operation takes the input as CB2 and CC1 and it checks for the conditions, if the output value becomes 1 then it denotes the position 2 and it ranges from 12800 to 13400 and if the output value is not equal to 1 then it gets ignored.
RESUTS AND DISCUSSION
5.1 RTL SCHEMATIC OF OVER ALL RICE SORTING MACHINE
Fig 5.1 RTL schematic of the overall rice sorting machine
Fig 5.1 shows the RTL schematic of overall rice sorting machine, it consists of all the six types of modules such as calibration input output, wave generation, clock generation. ADC settings, ADC process and ADC configurations that are to be as present in it.
5.2 RTL SCHEMATIC OF CALIB_IO PROCES
Fig 5.2 RTL schematic of the calib_io process
Fig 5.2 Shows the RTL schematic of the calibration input output process for rice sorting process, the calibration input output process is used for calculating the reference value.
5.3 RTL SCHEMATIC OF WAVE GENERATION PROCESS
Fig 5.3 RTL schematic of the wave generation process
Fig 5.3 shows the represents of RTL schematic of the wave generation process, in this it consists of comparator A, comparator B, comparator C and AND operation, which is used for positioning the ON and OFF position of the clock signal.
5.4 HARDWARE IMPLEMENTATION OF THE RICE SORTING MACHINE
Fig 5.4 Hardware implentation of the sorting machine
Fig 5.4 Shows the Hardware implentation of the sorting machine, in this it consists of Cyclone IV fpga, DSO(Digital Storage Oscillation) and power supply that are to be present in it, a digital storage oscilloscope is an oscilloscope which stores and analyses the signal digitally rather than using analog technique.
5.5 CLOCK GENERATION OF THE RICE SORTING MACHINE
The below images shows the 8MHz clock generation for the rice sorting machine
Fig 5.5 Clock Generation of the Rice Sorting Machine
Fig 5.5 Shows the Clock Generation of the Rice Sorting Machine, here the input is given as the 8MHz.
5.6 OUTPUT WAVEFORM OF THE RICE SORTING MACHINE
Fig 5.6 Output waveform of the rice sorting machine
Here the chute value is set as 40 then the rice gets flows through the machine the calibration input output process starts which is used to calculate the reference value and it also denotes the maximum and minimum tolerance value, once the calibration input output module stops the ADC module get start.
CONCLUSION AND FUTURE WORK
This project has described about the design and development of calibration input-output module, wave generation and clock generation modules for an automated rice sorting machine. The rice sorting machine can be used for categorizing the rice as defective ( fully black), partially defective ( black and white ) and non defective ( white). These three modules has been integrated with other modules of the rice sorting machine in the industry and has been found to be functioning successfully. This machine process can be extended for various grains, based on colour sorting techniques.
Md. Hazrat aii, N. Mir-Nasiri , (2017), “Design of an Automated Pepper Sorting Machine”, ELSEVIER-International Journal of Signal Processing: Image Communication, vol. 47, pp.263-270.
LeiYan, Sang-Ryong Lee, Seung-Han Yang, Choon-Young Leeka (2017), “CCDRice Grain Positioning Algorithm for Colour Sorter with Line CCD CSamera”, International Journal on Power and Energy Engineering, vol.4, pp. 263-270.
Rija hasan,Syed Mohammed and G.Monir (2017), “Fruit maturity estimation based on fuzzy classification Sign In or Purchase “, IEEE Conference on Fuzzy Logic Systems, vol.45, pp. 512-537.
Harish S Gujjar, (2016), “A Method for Identification of Basmati Rice grain of India and Its Quality Using Pattern Classification”, International Journal of Engineering Research and Applications (IJERA), vol.53, pp.268-273.
Xiai chen, Shuang ke, Ling wang, Hong xu and Wenquan chen (2012) , “Classification of rice appearance quality based on LS-SVM using machine vision”– Springer, vol.62, pp.104-109.
Mebatsion, H.K. Paliwal J., Jayas D.S. (2013),”Automatic classification of non-touching cereal grains in digital images using limited morphological and color features”, Elsevier Publications, Journal on Computers and Electronics in Agriculture, vol.64.pp.67-72.
Megha R. Siddagangappa1, A. H. Kulkarni (2014), “Classification and Quality Analysis of Food Grains”, International Conference on Computer Engineering, vol.16, pp.01-10.
Savakar .D,(2012),” Recognition And Classification Of Similar Looking Grain Images Using Artificial Neural Networks”, Journal of Applied Computer Science and Mathematics, vol.41, pp56-57.
Neelamma K. Patil, Ravi M. Yadahalli (2011),”Classification of Food Grains Using HSI Color Model by Combining Color and Texture”, Third International Journal on Computer Vision , Pattern recognition, and graphics, vol.23, pp.32-41.
Guo, P., Wang, X., Han, Y (2010), “The enhanced genetic algorithms for the optimization design”, International Conference on Biomedical Engineering and Informatics, vol.4, pp.2990-2994.
Dai, X., Xiong, D (2010), “Conjoint grains of rice based on machine vision”, International Conference on Signal Processing Systems (ICSPS), vol.23, pp.760-763.
Golpira H and Hoseinzadeh B (2010), “Design,development and investigation of a weight sorting automatic for apple”, Journal of Food Research, vol.19, pp. 38-47.
Anami B.S, Savakar D. G (2009), “Improved Method for Identification and Classification of Foreign Bodies in Mixed Food Grains Image Samples”, ICGST-AIML Journal on Image Analysis, vol.9, pp.106-110.
Juan,Sanz R, Uribe,Paula J., Ramos-Giralda (2008), “Algorithm to Identify Maturation Stages of Coffee Fruits”, International Conference on Image Processing, vol.67, pp. 329-332.