Real-Time Surveillance for Identification of Fruits Ripening Stages and Vegetables Maturation Stages with Infection Detection

For every fruit and vegetable, there are various ripening stages and maturation stages. For example, according to US standards for greenhouse tomatoes, there are six stages in tomato ripening. But the useful stages are only last two. Improper supervision in harvest methods and fertilization leads to the decrease in the post-harvest life of fruit and vegetables. Monitoring the ripening stages and maturation stages require proper knowledge of color and size of fruits and vegetables. In order to decrease excessive labor charges and wastages of fruits and vegetables, autonomous farming methods can be introduced. In autonomous farming, the main objective is to detect a vegetable or a fruit is ripe or not and it’s infected or not. If this objective is achieved then farmers can easily harvest fruits and remove the infected fruits too. The infected fruits need to be removed so that it doesn’t infect others. And proper fertilizers can be given to there to fruits. For example, if mango is infected by a deficiency disease then proper vitamin fertilizer can be given to the fruit. So by using image processing in python using an efficient HSL color space algorithm and OpenCV library,a real time monitoring system is made to identify the maturation and ripening stages of a vegetable or fruit and to check whether a fruit is infected or not. Based on the comparison of existing systems for various fruits and vegetables, there is a very scarce availability for some fruits and vegetables detection of stages, which specifically in real time image processing. By using images there are lot of disadvantages where we need to take a lot of pictures, the lighting, and in real time in farming it is tedious. So live detection of image processing is done in the proposed system using Open CV library in python. In existing system RGB color spaces are used. Here an efficient algorithm using HSL color space is used to identify a fruit vegetable ripening and maturation stages. The proposed system also includes the detection of diseases in fruit or vegetable using the above method. This is done by setting the specific color ranges, brightness and saturation ranges in algorithm and metrics for shapes and size too. By an efficient setting of these metrics an efficient algorithm is found out.