Analyzing the Accuracy of KNN-based Cacao Bean Grading System
Authors:
- 1 November 2019
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-5
Abstract
Cacao is one of the major crops of the tropical world and is known worldwide for its beans used for manufacturing of products which are highly popular and widely consumed around the world such as chocolate and cocoa powder. Grading the cacao beans is a method utilized by the cacao experts and farmers to ensure valuable and good supply of cacao beans in the market. However, manual grading of the cacao beans using the naked eye observation is more laborious, time consuming and less accurate. To automate the said process, a computer-based cacao beans grading system was developed using image processing and KNN algorithm. One hundred ninety (190) samples were consumed as training examples and sixty samples (60) for classification. The system was designed and developed using C# Windows Form Application as its programming language, XAMPP as its server scripting language, and MySQL as its database. The accuracy calculation of the system resulted to 93.33%, which implies that the KNN model was able to effectively grade cacao beans.Keywords
This publication has 7 references indexed in Scilit:
- Sensing fermentation degree of cocoa (Theobroma cacao L.) beans by machine learning classification models based electronic nose systemJournal of Food Process Engineering, 2019
- Prediction of fermentation index of cocoa beans (Theobroma cacao L.) based on color measurement and artificial neural networksTalanta, 2016
- Corn Seed Varieties Classification Based on Mixed Morphological and Color Features Using Artificial Neural NetworksResearch Journal of Applied Sciences, Engineering and Technology, 2013
- Grading of Fermented and Dried Cocoa Beans Using Fungal Contamination, Ergosterol Index and Ochratoxin a ProductionMycobiology, 2009
- Identification of rice seed varieties using neural networkJournal of Zhejiang University-SCIENCE B, 2005
- AE—Automation and Emerging TechnologiesJournal of Agricultural Engineering Research, 2001
- Determination of wheat kernel morphological variation by digital image analysis: I. Variation in Eastern Canadian milling quality wheatsJournal of Cereal Science, 1988