Premium
Integration of computer vision and colorimetric sensor array for nondestructive detection of mango quality
Author(s) -
Huang Xingyi,
Lv Riqin,
Wang Sun,
Aheto Joshua H.,
Dai Chunxia
Publication year - 2018
Publication title -
journal of food process engineering
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.507
H-Index - 45
eISSN - 1745-4530
pISSN - 0145-8876
DOI - 10.1111/jfpe.12873
Subject(s) - mangifera , principal component analysis , climacteric , artificial intelligence , mathematics , support vector machine , pattern recognition (psychology) , computer science , ripening , horticulture , biology , genetics , menopause
A method of digital image and odor information processing has been proposed by integrating computer vision and colorimetric sensor array (CSA) for rapid and accurate evaluation of mango quality. Wholesome mango fruits, about 70–80% maturity were procured and stored in a constant temperature‐humidity chamber, 12 ± 0.5°C and 85–90%, respectively. Hardness and Total Soluble Solid (TSS) of the mango samples were measured by both conventional techniques and new nondestructive method developed combing computer vision and CSA. All data were analyzed using principal component analysis to reduce dimensionality. Support vector classification (SVC) models were established for qualitative discrimination of mango quality. Moreover, support vector regression (SVR) was applied to indicate the relationship between results got from nondestructive methods and conventional methods. SVC model was used to classify mango samples into three grades, the accuracy rates were 98.75 and 97.5% for the training and prediction sets, respectively. The SVR correlation coefficients for hardness were 0.9051 and 0.8897 for the training and prediction sets, respectively, and 0.9515 and 0.9241 for training set and prediction sets, respectively, in respect of TSS. Results showed that it is feasible to predict hardness and TSS of mango by the combination of computer vision and CSA. Practical applications Mango ( Mangifera indica L.) is one of the world's famous tropical fruits and enjoys the reputation of “tropical fruit king.” Mango is considered a climacteric fruit because during ripening it displays a surge of respiration and ethylene production which tends to hasten the ripening process. To keep the mango fruit fresh, it is very important to monitor the quality during transportation and storage. In this study, an innovative approach was developed, in which computer vision and colorimetric sensor array (CSA) were employed simultaneously to get more accurate result. This method simplified detection steps and shorten the detection time. The results showed that the integration of computer vision and CSA could serve as a rapid nondestructive testing method for mango quality detection. The method can be applied for rapid detection of mango products by both government department and food company.