Detection of Fruit Fly using Intelligent Traps

Agriculture has been the source of economic wealth and the vital source of the population?s food supply, which is extensively affected by the infestation of harmful insects. Thus, it requires an effort to control the infestation of such pests. Fruit Fly is one of the largest threats to the

Project Title

Detection of Fruit Fly using Intelligent Traps

Project Area of Specialization

Artificial Intelligence

Project Summary

Agriculture has been the source of economic wealth and the vital source of the population’s food supply, which is extensively affected by the infestation of harmful insects. Thus, it requires an effort to control the infestation of such pests. Fruit Fly is one of the largest threats to the fruit farms which spoils a large number of fields every year. Therefore, the objective of this project is to prevent and predict the presence of the fruit fly using intelligent traps.

Fruit Fly has been a serious threat to Pakistan’s agriculture and caused approximately 200$ million in losses to the farms, traders, retailers, and exporters annually. The traditional strategies have significant downsides, which we can feature: the serious, undesirable, and expensive work of dislodging a human by the estate, region and you can't get continuous and synchronized data from all traps as they are being visited consistently.

This solution proposes a modern idea upon the traditional ways of tackling the harms of the pests more accurately and precisely. Dataset as the images of the fruit fly, collected from the farm fields of the University of Agriculture, Faisalabad, and areas where fruit fly propagates, are used for building the deep learning model. Using the proposed solution, the propagation and the infestation of fruit flies can be taken down exponentially which would result in better farms and more wealth.

Project Objectives

The main goal is to design and develop and, an intelligent trap for the early detection of fruit flies. Furthermore, using deep learning techniques to build such an automated system to detect the fruit fly presence in the intelligent traps. The force behind the idea of the proposed solution for the detection of pests particularly fruit flies-, is to decrease the ratio of loss that is accumulated every year.

Project Implementation Method

Traps that are equipped with cameras capture the image of the fruit fly and use Computer Vision techniques to detect the fruit fly. After detecting the presence of fruit flies, the system alerts the farmer about a particular location of fruit fly infestation so that necessary measurements can be carried out.

• Collection of the dataset for training the model.
• Preprocessing the collected data for training.
• Training the model with suitable parameters.
• Applying optimizing techniques to enhance the accuracy and precision of the model.
• Testing and validation of the model with the help of domain expert Dr Khuram Zia (PhD Entolomogy).
• Integration with the graphical dashboard.
• Building a graphical dashboard application for farm experiments.

Benefits of the Project

By implementing the proposed solution to agriculture, the stakeholders can spend their time and money on the things that would help in the growth of the fields rather than the protection and security of the farm fields.

• Less number of farmers and the labor will be required for the maintenance of the farms and prevention of the pests.
• This proposed solution idea is a modern technique regarding the traditional ways of eradicating the harmful effects of pests with more precision and effortlessly.
• The value regarding the financial capital that will be generated in terms of less spending on pest infestation control and prevention would preferably be used for the growth and cultivation of healthier and quality fruits.

Technical Details of Final Deliverable

Final deliverable consists of an automated dashboard that alerts the user about the presence of fruit fly in a particular area of the farm. The intelligent trap would notify on the dashboard about the particular location of the trap where fruit fly presence has detected.

Python programming language will be used as core development tool along with artificial intelligence frameworks such as, TensorFlow, jupyter notebook, and HTML, Bootstrap.

The hardware system will consist of a high-resolution camera, Raspberry Pi 2 Model B Module, and with mobile internet dongle, powered by a high-capacity battery, fixed inside the fruit fly trap.

Final Deliverable of the Project

HW/SW integrated system

Core Industry

IT

Other Industries

Agriculture

Core Technology

Artificial Intelligence(AI)

Other Technologies

Others

Sustainable Development Goals

Industry, Innovation and Infrastructure

Required Resources

If you need this project, please contact me on contact@adikhanofficial.com