World Health Organization (WHO) recommends a daily intake of at least 400 grams of fruits and vegetables, to prevent diet related chronic diseases and micronutrient deficiencies. It is essential to ensure the quality of foods consumed by the population on a daily basis. Quality of fruits and vegetables is governed by nutritional level, appearance, flavor and climate. Quality assessment methods should be environment friendly and should also benefit both consumers and farmers by enhancing taste and increasing yield, respectively. Conventionally employed quality assessment methods like bio-chemical analysis, imaging etc. are destructive, inefficient and time consuming. Bio-tissues are made up of cells with selectively permeable cell membranes and this makes them equivalent to resistive-capacitive network. Such a network impedes an alternating current (AC) excitation signal applied to it. This bio-impedance (BI) is measured through LCR meters, impedance analysers and off-the shelf chip based boards. In this work we have developed an accurate, smart and non destructive bio-impedance based quality evaluation technique. BI is measured as magnitude and phase for 7 days for each variety of fruit and vegetable followed by exhaustive frequency (5 kHz-200 kHz), ripening and rot analysis. BI magnitude increases as number of days advance i.e. with ripening and the phase undergoes considerable decrease with rotting. The system is made smart by incorporating an expert system. 178 samples of bio-impedance data are used to train the expert system and supervised classification is done through Random Forest classifier. Any fruit or vegetable can be classified as ’Good’ or ’Bad’ immediately and accurately with a maximum accuracy of 98.57%
New Application of Nondestructive Methods for Quality Evaluation of Fruits and Vegetables in Japan.
