Fruit Sorting By Machine Vision Technique

Khor, Chin We (2016) Fruit Sorting By Machine Vision Technique. Project Report. UTeM, Melaka, Malaysia . (Submitted)

Text (24 Pages) Fruit Sorting By Machine Vision Technique.pdf - Submitted Version Download (277kB)

Abstract

Nowadays, automation system such as machine vision system is widely used in many industrial. Fruit sorting by machine vision technique is a potential automation system which is able to obtain their target in term of productivity and quality. However, there are several problems such as camera resolution, illumination and performance that need to be considered for using the machine vision technique. Hence, the objectives of this project is to design a high accuracy algorithm that is able to classify apples based on color and size. Furthermore, the performance of fruit sorting system is needed to be evaluated and optimized by manipulating the illumination, thresholding value and camera distance. MATLAB is an analysis tool that has been used in this research project. There are three experiments such as HSV value comparison, thresholding value performance comparison and the different camera distance performance comparison. The experiments are aim to define the tolerance limits of hue value and number of pixels. The tolerance limits are used for the purpose of categorizing the type of apple. At the end of the experiment, the analysis and discussion explained about the performances of each variation of parameters. The problems such as low illumination condition and the limitation of different thresholding value and distance of camera are described as well in the analysis and discussion. As conclusion, the performance in term of accuracy and precision of the designed algorithm has been evaluated and optimized at the end of this project.

Actions (login required)

View Item