Development of an Algorithm for Size and Weight Measurement of Intact Pineapples

Abstract

In the past, in order to classify pineapples according to their size and weight, farmers usually measured the size of each pineapple manually, it takes a lot of time and human resources to obtain pineapple information. The objective of this project is to develop an algorithm for size and weight measurement for intact pineapples by using OpenCV and Python as image processing algorithms. In this project, the webcam is used to capture the appearance of pineapples and the computer is used to execute the algorithm by providing reliable measure results. The developed system was able to capture a valid image of the pineapple and accurately estimate the size and predict the weight of the pineapple. The regression analysis will decide the reliability of the prediction. The size and weight measured by the proposed algorithm were strongly correlated with manually measured values. The final phase of the project is to provide a rapid, accurate, and non-destructive method for agriculture and improve work efficiency for farmers.  The result shows the highest percentage error of the diameter is 2.77% and the length is 2.84%. For the weight prediction, there is a 94.56% of strong correlation in between the size and weight prediction. Next, hope that this project can reduce the physical strength and time required for farmers to work and thereby reduce the dependence on human resources for agriculture.