Banana Plant Disease Detection Using Image Processing

Pakistan is one of the major banana cultivated countries . Banana crop is cultivated upto 32.2 thousands hectare, which is almost 92% of the area under banana in Pakistan, producing 126 thousands tones 80% of total bananas produced in the country. however crops are affected by different disease, and

Project Title

Banana Plant Disease Detection Using Image Processing

Project Area of Specialization

Artificial Intelligence

Project Summary

Pakistan is one of the major banana cultivated countries . Banana crop is cultivated upto 32.2 thousands hectare, which is almost 92% of the area under banana in Pakistan, producing 126 thousands tones 80% of total bananas produced in the country. however crops are affected by different disease, and banana is also hotspot for diseases , most common and very affected diseases in banana cultivation is black sigatoka(black leaf streak disease ) caused by fungus named Mycosphaerella fijiensis.This disease affect the banana production level and also destroy the banana tree, that is why it is necessary to take prevention against these type of disease. It is very difficult to monitor these type of plant diseases manually b/c they require high botanical knowledge, well trained and skilled person too. To
overcome these types of diseases we must set up a technological system which not only monitors it continuously but also diagnoses the type of disease .This project outlines the development of a support system for the detection of black sigatoka diseases using digital images.

Advantages:
? With very less time and less computational efforts, the disease can be detected and an optimal result will be provided.
? It reduces the loss as diseases can be detected, identified and procured at initial stages.
? It reduces the economical loss to the farmers.
? Increases the chances fruit quality to be exported.
? Increase economic growth.

Applications:
? It is very important research content in the field of machine vision.
? Widely applied to agriculture science
? It has immense perception in the plant protection field.

Our aim is to propose a plant disease detection system for banana plants which can help us to grow our major fruit crop of good quality by early detection of a powerful banana leaf disease named “black sigatoka” so the plants can be saved, and all the required treatment and cured techniques
will be applied to ensure its quality and health at the correct time. This system saves a lot of money at the local level as it reduces farmers' economical loss, efforts, and time whereas at the global level it over all saves the loss of 30% to 50% of banana crop which can be consumed effectively and exported to grow the country’s economy.

The work that is proposed in this system aims at classifying the black sigatoka disease of banana plants.

Diseases disintegrate the normal structure, growth and other activities of the plant. In this proposed system black Sigatoka disease is mentioned which is commonly found in banana plants. This disease is detected through this system and can be classified as well as suggestions will also be provided which includes cultural techniques and chemicals to control
and overcome the banana plant disease.

Project Objectives

Our aim is to propose a plant disease detection system for banana plants which can help us to grow our major fruit crop of good quality by early detection of a powerful banana leaf disease named “black sigatoka” so the plants can be saved, and all the required treatment and cured techniques will be applied to ensure its quality and health at the correct time. This system saves a lot of money at the local level as it reduces farmers' economical loss, efforts, and time whereas at the global level it over all saves the loss of 30% to 50% of banana crop which can be consumed effectively and exported to grow the country’s economy.
? We will use image processing techniques to detect the disease in banana plants by means of leaf’s color, texture and morphology.
? The image classification is done by using ANN which takes the query image as input and classifies images based on the learning database.
? The outcome specifies that whether the plant is affected by black sigatoka or the plant is healthy, it also determines the probability of disease in the plant.
? This system is capable of providing suggestions based on the stages of disease as your affected plant requires which chemicals are to be used for cured as well as provides alternative suggestions of cultural techniques for the plant health and treatment.
? We will design an android application of this system to make it more usable and portable.

Project Implementation Method

The work that is proposed in this system aims at classifying the black sigatoka disease of banana plants.

Image Acquisition: Image acquisition is the first step of image processing technique.The images for black sigatoka disease of the leaf are captured by a digital camera of resolution of 16 mega pixels.

Image pre-processing: The enhancement of image or data image to convert to a proper image prior to computational processing.
Pre-processing of images includes cropping, resizing and color conversion. The images that are obtained by pre-processing are heterogeneous in dimensions. For the efficient processing we can resize the images.

Image Segmentation: Image Segmentation is referred to as partitioning of images into various parts of the same features or having any kind of similarity. The segmentation can be done by using multiple methods such as, otsu’ method, k-means clustering, converting RGB image into HIS
model etc.

i) Segmentation using Boundary and spot detection algorithm: The RGB image is converted into the HIS model for segmenting. Boundary detection and spot detection helps to find the infected part of the leaf.

ii)K-means clustering: The K-means clustering is used for classification of objects based on a set of features into K number of classes. The classification of objects is done by minimizing the sum of the squares of the distance between the object and the corresponding cluster.

iii)Otsu Threshold Algorithm: Thresholding creates binary images from gray-level images by setting all pixels below some threshold to zero and all pixels above that threshold to one.

Feature Extraction: Feature extraction is very useful for the identification of an object. It is done after the image segmentation.According to the segmented information and predefined dataset some features of the image should be extracted. Feature extraction is useful in many applications of
image processing techniques.

Classification: Classification is done after the feature extraction.

Using machine learning: After feature extraction is done, the learning database images are classified by using a neural network. These feature vectors are considered as neurons in machine learning techniques.

Benefits of the Project

Advantages:
? With very less time and less computational efforts, the disease can be detected and an optimal result will be provided.
? It reduces the loss as diseases can be detected, identified and procured at initial stages.
? It reduces the economical loss to the farmers.
? Increases the chances fruit quality to be exported.
? Increase economic growth.

Technical Details of Final Deliverable

This project will detect the infected plant(banana ) leave . The projetc also calculate the percentage of  damage done by the disease ,after all the calculation done project also suggests the techniques i.e cultural as well as chemical to to overcome the disease . project is very easy to operate , farmers just take the picture project scan that picture and performs all the calculation , after the calculation a result window show on the screen which display details of result. this project is very easy to handle as simple as  calculator app.

Final Deliverable of the Project

Software System

Core Industry

Agriculture

Other Industries

Food

Core Technology

Artificial Intelligence(AI)

Other Technologies

Sustainable Development Goals

No Poverty, Zero Hunger, Clean Water and Sanitation, Decent Work and Economic Growth

Required Resources

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