Weed Detection Using AI

Project Title

Weed Detection Using AI

Project Area of Specialization Artificial IntelligenceProject Summary

These days a general opinion has been developed about fields such as artificial intelligence and machine learning that they are the future and a time would come that most of human decision making would be automated to a great extent. Artificial intelligence which was a mere speculation around two to three decades ago, now has emerged as the new big thing in the IT industry worldwide. Sub-fields such as machine learning and computer vision are branches of artificial intelligence which work together in order to form an artificial neural network. In order to formulate an efficient neural network we take aid from techniques such as semantic segmentation and deep convolutional neural networks. Segmentation is a process in which we classify an image at pixel level by labeling the different parts of the image with specific names. This technique is used so that the process of understanding an image for an artificial neural network becomes easy.

Project Objectives

In the recent past, machine learning and computer vision have transpired as aspiring fields and we as computer engineers plan to capitalize on it by creating a system which employs deep convolutional neural networks to detect wanted plants and unwanted plants in a crop field. We plan to do so by training our model with specific data which is feasible for us to collect and then implementing it on specific practical scenarios.

The main purpose of choosing this project was to provide an efficient and healthy way of eliminating weeds (unwanted plants). By being part of a progressive society it is our utmost duty to contribute for the betterment of our society. Our system can successfully replace all existing systems (different weedicides causing pollution and health problems) being used for the removal of weeds in the agricultural sector of Pakistan.

Project Implementation Method

project implementation will be carried out through deep learning (artificial intelligence). we are using specific convouloutional neural network which is known as fully convuloutional neural network for our implementation. we will use semantic segementation which will help us in identifying unwanted plants in fields of crops.

Benefits of the Project

1. Encourages automation dependent systems over labor dependent systems.
2. Cost effective in a bigger perspective.
3. Discourages pollution and supports green energy.
4. A revolution for the agricultural industry in Pakistan.
5. Time efficient.
6. The overall health, vigor and yield of crops would enhance.

Technical Details of Final Deliverable

it is graphical unit interface which has to option of detecting weeds in field of potatoes and maize. user will have to select the type of crop then he will be be able to see crop,weed and soil in the given picture. computer is doing this automatically so it will help in automation of agriculture side.

Final Deliverable of the Project Software SystemCore Industry AgricultureOther Industries Food Core Technology Artificial Intelligence(AI)Other Technologies Artificial Intelligence(AI)Sustainable Development Goals Good Health and Well-Being for People, Decent Work and Economic GrowthRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	53000
graphic card	Equipment	1	35000	35000
camera	Equipment	1	15000	15000
printing	Miscellaneous	1	3000	3000