Characterization of optical parameters in seeded and seedless watermelon using laser-induced backscattering imaging

Monitoring of watermelon fruit quality is essential in order to regulate proper postharvest handling and yield production. Problems arise in forecasting the quality parameters of watermelon during storage as the shelf-life of the fruit is only lasts for three weeks. The non-climacteric nature of the fruit is also related to high perishability which does not undergo a continuous process to ripen after being harvested. In the present study, laser-induced backscattering imaging was used to determine the firmness, soluble solids content (SSC), pH, moisture content (MC), and colour changes of watermelons in seven interval days starting from storage day 0 after harvesting (day 0, day 4, day 8, day 12, day 15, day 18, and day 21). Two types of watermelon cultivars were used in this study; seeded watermelon (Black Beauty) and seedless watermelon (Red Seedless). The backscattered images of the fruit surface were obtained at six different locations using laser diodes emitting at 658 nm wavelength. The backscattered images were analysed and the feature information was extracted based on the backscattering image parameters which are minor length, major length, perimeter, maximum intensity, minimum intensity, and mean intensity. The standard reference methods were carried out after the image acquisition process in order to determine the quality parameter measurements. The multivariate analysis was used to optimise the classification between seeded and seedless watermelons, as well as the regression models between backscattering data and quality parameters, were also discussed. Principal component analysis (PCA) was used to classify the seeded and seedless watermelons into two different classes according to the physicochemical changes. Partial least squares (PLS) regression with full cross-validation method was used to establish the regression models between the backscattering data and quality parameters. The firmness values obtained were 5.07 to 3.24 kg/cm2 and 4.78 to 3.03 kg/cm2, whereas the SSC values achieved 9.06 to 5.66 Brix and 8.52 to 6.41 Brix for the seeded and seedless watermelons, respectively. The pH values were 5.27 to 7.10 pH and 5.60 to 6.42 pH while the moisture content values revealed 95.46 to 82.79 % w.b. and 94.53 to 86.43 % w.b. for the seeded and seedless watermelons, respectively. For colour parameters, the L* values ranged from 24.08 to 51.06, whereas the b* values ranged from 11.52 to 36.84. The chroma and hue values were also increased from 12.41 to 38.95 and 104.55º to 111.60º, respectively. The a* value reduced significantly (P<0.05) from -4.32 to -13.06 for both seeded and seedless watermelons. For seedless watermelon, the colour prediction (L*, a*, b*, chroma, and hue) gave the highest coefficient of determination (R2) with all of them above 0.90. Meanwhile, the firmness prediction gave the highest R2 of 0.92 for seeded watermelon. On the whole, it is concluded that the application of laser-induced backscattering imaging is a promising technique for assessing quality parameters of watermelons during storage. The proposed approach has significant potential as a well-controlled system for developing automated sorting and grading system in the future.

Preview Text FK 2017 31 - IR.pdf Download (2MB) | Preview

Actions (login required)

View Item