Disease Detection of Mango Leaves

Project Title

Disease Detection of Mango Leaves

Project Area of Specialization Artificial IntelligenceProject Summary

Mangifera(Disease Detection of Mango Leaves) is an application which will help the layman planting at home or farmers in the fields to accurately detect different types of diseases in Mango plants by imaging leaves through their Mobile phone. The application will also assist in the proper treatment of a particular disease by prescribing suitable pesticides. Disease detection and recognition will be based on machine learning and image processing techniques. The application would be developed using state of the art machine learning algorithms. Our project will involve research and development for better strategies and developing efficient object detection methods.

Project Objectives

The Project Objectives are:

• To reduce the risk, time, cost and increase accuracy and the efficiency of the mango leaves disease detection process.
• To prescribe the most suitable pesticide for the detected disease.

Project Implementation Method

The proposed method consist of Six phases:

1. Image acquisition
2. Pre-processing of the image (smoothing, histogram equalization, resizing etc.).
3. Thresholding of the image: if the pixel intensity of the green component is less than the pre-computed threshold value, then zero value is assigned to the red, green and blue components of this pixel.
4. Segmentation: Kmean Clustering is used to extract the leaf from the image.
5. Implementation of different Classification techniques to get the best results.
   1. Extract features using AlexNet or googLeNet separately (fully connected layer) from the segmented images then Classify using multiclass Support Vector Machine (SVM)
   2. Implement different Convolutional Neural Network directly apply to segmented images.

Benefits of the Project

Our application will revolutionize the disease detection process of mango leaves and the most suitable pesticide recommendation. Diseases on plants are mostly detected through the naked eye, and it takes a lot of time and experience to cover large areas. It is practically impossible for a relatively new person to detect the disease accurately and timely; that's where the machine comes. Our application will help farmers across the globe in the detection of diseases?.

Technical Details of Final Deliverable

The final deliverable will be an application that will be supported for both Andriod and IOS devices. The application will be developed using the following technologies. 

• React Native
• Firebase
• Django rest framework for API development

Final Deliverable of the Project Software SystemCore Industry AgricultureOther Industries IT Core Technology Artificial Intelligence(AI)Other TechnologiesSustainable Development Goals Good Health and Well-Being for People, Responsible Consumption and Production, Life on LandRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	10000
Travel for Data Collection	Miscellaneous	1	10000	10000