Image Processing Algorithm for Tomato Detection and Ripeness Classification

Ong, Teng Guan (2018) Image Processing Algorithm for Tomato Detection and Ripeness Classification. Final Year Project (Bachelor), Tunku Abdul Rahman University College.

Text ONG TENG GUAN-FULL TEXT THESIS.pdf Restricted to Registered users only Download (1MB)

Abstract

Tomatoes are exportable items of Malaysia hence the strict quality of tomatoes is the most important factors that helps ensuring the marketing for both local and export market. The quality of the tomato was mainly determine by the ripeness. Hence, the determination of tomato ripeness stage is an industrial concern about to acquire the high quality of tomato. The proposed approach utilize both color features and texture features of the tomato for classifying tomato. Watershed transform algorithm is employ to de separate the background and touching tomato. However, watershed algorithm was not powerful enough to determine the ‘real’ tomato. Some elimination of false tomato had taken to ensure the dataset accuracy. Multiple class classification has employ by using the Support Vector Machines learning technique for the classification of the tomato condition. The project had conducted on a dataset of on-site tomato images, which has been use for both training and testing dataset.