Image Processing System for Automatic Segmentation and Yield Prediction of Fruits using Open CV

In fruit harvesting systems, automatic yield counting of fruits becomes a big issue. Image processing techniques minimizes the manual task of recognizing and counting the fruits. In this paper, an image processing system for automatic segmentation and yield prediction of fruits is proposed on the basis of color and shape features is being performed. Initially the preprocessing is done on input fruit tree images. Then it is converted from RGB to HSV color space to detect the fruit region from its background. Color thresholding is used to mask the desired colors. Gaussian filter is used to remove noise. The contour of the image is taken. Then these images are processed by image processing algorithm. Color and shape based counting of fruit is presented at the output. The edge detection and combination of a circular fitting algorithm is applied for the automatic segmentation and automatic counting of fruits in the image. Different types of fruits (orange/tangerine, pomegranate, apple, lemon, mango, cherry) are used for automatic counting. Open CV Python software is used to perform the required image processing operations.