Food Maturity and Disease Detection Using Digital Image Processing

Abstract

We have developed an automated system that detects the maturity of the fruit as well as detects if the plant is diseased. In this project we worked on the fruits whose maturity can be determined on the basis of the colour. We have mainly concentrated our project on tomatoes as their maturity can be detected on the basis of the colour as well as they are one of the most eaten fruits around the globe. Further we went on to determine if the tomato plant is diseased or not on the basis of the symptoms observed on the plant leaves. Detection of the diseased plant at correct time is very important as ignorance of it can lead to the spread of disease in other plants too. This could cause destruction to the plantation on large scale. The main purpose of our project is to improve the crop yield by detection of the disease on time and the proper use of the crop on the basis of its maturity. We have worked on two methods of digital image processing in our project. Initially we performed thresholding algorithm to determine the maturity of the fruit as well as to detect the diseased plant. To make the system more generalised and self-adapting we shifted to k-means clustering. Finally we did a comparative analysis of both the methods to analyse which method is more suitable in different conditions .