Categorization of watermelon maturity level based on rind features

• Main Authors: Ahmad Syazwan, N. (Author), Nooritawati, M.T (Author), Shah Rizam, M.S.B (Author)
• Format: Article
• Language: English
• Published: Elsevier Ltd 2012
• Subjects: Artificial neural network; Image processing; RGB; Watermelon
• Online Access: View Fulltext in Publisher; View in Scopus

 This paper proposed a method to determine the best capture position of the watermelon's rind for categorization of maturity level. A total of forty five watermelon samples are used specifically fifteen samples are in 60% ripeness, another fifteen samples in 70% ripeness and remaining fifteen samples in 80% ripeness. Five different positions of the watermelon which are from the top, bottom, lean, yellowish spot and best side are captured using a digital camera. The images are then analyzed using feature extraction based on RGB color components value. Each capturing positions is analyzed and comparisons are made between all five capturing position to identify the optimum capturing angle that provides significant ripeness stage. Initial results attained showed that capturing the yellowish spot of the watermelon as the best categorization features. © 2012 The Authors.