Computer Vision Based Smart Fruit Sorter

Project Title

Computer Vision Based Smart Fruit Sorter

Project Area of Specialization Electrical/Electronic EngineeringProject Summary

Fruit based agriculture not only supports the economy of Pakistan, it is a beneficial enterprise for employment and is favorable geographically. Although yearly agricultural yield is high quality, the lack in marketing strategies, standardization procedures limit the export as well as the criteria set to fit the demand of the competitors. The most important thing for the exporter is standardization of agricultural yield. The grading of export quality fruit in terms of weight and quality as well as the delivery protocols up to a standardized level are necessary for the export market and it is done manually in Pakistan. Manually grading fruits is a tedious task and is prone to human error and certain inconsistencies in the color and weight of the fruit. An automated system has precision, takes lesser time, and eliminates the cost of manual workers, therefore proving to be more cost-effective and impervious to tiredness.

Computer Vision is a necessity in implementing an automated fruit grading system. Computer Vision is able to learn and make decisions by keeping in regard physical parameters of objects, reducing costs while making sure accurate and precise recognition in real-time. This technology acquires and analyzes an image of a real scene by computers to control machines or to process it. It includes capturing, processing, and analyzing images to facilitate the assessment of visual quality characteristics in agricultural and food products. The proposed system consists of a conveyer belt with an imaging mechanism that will capture an image of tomato and classify it into Raw, Ripe, and Overripe state.

Keywords: Fruit Sorting, Computer Vision, Fruit Grading, Image processing.

Project Objectives

1. To develop an automated system over manual system for sorting the fruits.
2. To develop a system that will take the snap of the fruit of all sides according to the given requirement.
3. The objective of the project is to reduce the manpower required and increase the productivity by developing a mechanism for automatic sorting of the fruits.
4. To reject the product if the sample does not meet the requirement.
5. To meet the standards of industry requirement.

6. To design a project with less errors.

Project Implementation Method

First of all, we will make wheel type conveyer belts. The conveyer belts will be operated by motors and after the complete checking of these motors we will apply it on the conveyer belt. After that we will design the bottom knockout mechanism using servo motors. The knockout mechanism would be made on a metal bar on which it will move push and pull and this mechanism will control by Arduino.

The connection circuit of whole project would be through controller programming. We shall also design the motor controlling circuit for the controlling of conveyer belt and for the grab mechanism as well. In the end we will compile all the circuits along with their mechanism and after that we will test the project.

Benefits of the Project

Agricultural exportation regarding fruit exports is beneficial for a country’s economy. Pakistan’s economy is based majority on agriculture and annually it exports tons of agricultural exports. The results revealed that t?m?t? ex??rts from ??kist?n fetched lower ?ri?es as ??m??red to that of the world. On an ?ver?ge, ??kist?n is ex??rting 9949t which is
very less as ??m??red to world’s ?ver?ge ex??rts of 37.7 mt But the market suffers from a lack of standardized methods of grading exportation material inflicting a toll upon the cost of
high-quality export material. The grading of export quality fruit in terms of weight and quality as well as the delivery protocols up to a standardized level are necessary for the export market and it is done manually in Pakistan. Manually grading fruits is a tedious task and is prone to human error and certain inconsistencies in the color and weight of the fruit. An automated system has precision, takes lesser time, and eliminates the cost of manual workers, therefore proving to be more cost-effective and impervious to tiredness.

Technical Details of Final Deliverable

The project will have consisted of wheel-type conveyor belts connected through sensors and a camera. There will be a manual mechanism to put the fruits on the conveyor belt on the top. The fruit will reach the image capturing chamber then camera takes the snaps with all the sides and then will reach the grading classification, this part decides the quality of the fruit according to the parameters (i.e., raw, ripe overripe, size, color, and texture) and then distribute in class A, class B, and class C. The whole process would be controlled by microcontroller.

Final Deliverable of the Project Hardware SystemCore Industry AgricultureOther Industries Agriculture , Food , Manufacturing Core Technology Internet of Things (IoT)Other TechnologiesSustainable Development Goals Decent Work and Economic Growth, Industry, Innovation and Infrastructure, Reduced Inequality, Responsible Consumption and ProductionRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	77800
Mechanical Structure	Equipment	1	10000	10000
Stepper Motor NEMA43	Equipment	2	10000	20000
Wheel Type Conveyor	Equipment	1	5000	5000
Sensor and Camera	Equipment	2	2200	4400
Raspberry Pi 3	Equipment	2	10000	20000
Servo Motors Mg-977t	Equipment	4	1200	4800
Motor Driver Circuit	Equipment	2	700	1400
ON/OFF Panel Box	Equipment	1	1000	1000
Breaker and Indication Lights	Equipment	2	1000	2000
Wires, screws, nut bolds, metal plates.	Miscellaneous	2	1000	2000
Wheels	Miscellaneous	6	500	3000
Powder Coating	Miscellaneous	1	2000	2000
Motor Mounts	Miscellaneous	4	300	1200
Camera Mount	Miscellaneous	1	1000	1000