An Intelligent Noninvasive Taste Detection App for Watermelons

• Published in: 2017 5th Intl Conf on Applied Computing and Information Technology/4th Intl Conf on Computational Science/Intelligence and Applied Informatics/2nd Intl Conf on Big Data, Cloud Computing, Data Science (ACIT-CSII-BCD) pp. 90 - 94
• Main Authors: Cheng, Yu-Shan, Wang, Shun-Chung, Liu, Yu-Hua, Peng, Bo-Ruei
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.07.2017
• Subjects: Android; Application software (app); artificial neural network; Eigenvalues and eigenfunctions; Gray-scale; Image color analysis; Indexes; non-invasive taste detection; Smart phones; Training; watermelon
• DOI: 10.1109/ACIT-CSII-BCD.2017.74

In this paper, a convenient, low-cost, noninvasive and smart watermelon taste detection system with an app is developed based on mobile devices for helping people to pick tasty watermelons. By patting and photoing the watermelon using smart phones, the emitted sound and image are recorded. These data are sent to the system cloud for training and computation. The fast Fourier transform (FFT) and digital image processing technology are used to extract the eigenvalue from voice signal and image for judging the watermelon taste. Taste indices are transferred and displayed from cloud back to the consumers' smartphones. In order to achieve intelligent watermelon taste judgment, a questionnaire about the sweetness, pulp color and water is carried out to build the taste samples and, a learning based on the database is conducted with the artificial neural network. Various mobile devices equipped with Android or iOSx operating software can be used to install the developed WatermelonSweet app. The Xiaoyu watermelon is taken as the test target in this paper. Without additional overhead, everyone can become a fruit expert through the developed app. The proposed non-invasive taste detection system can also be applied to the other fruits which can emit sound by patting them.