Lacto-fermentation can be used to preserve chili pepper after harvesting. It extends the shelf life of fruit. Additionally, lacto-fermented products can be characterized by health-promoting properties.
The main objective of this study was to assess the usefulness of image parameters of the flesh of chili pepper for the estimation of the changes in the content of chemical compounds in samples during spontaneous fermentation. Therefore, this study aimed at assessing changes in the image parameters of the flesh and the content of chemical compounds of chili pepper after specified fermentation periods, determining the correlations between the image parameters and chemical properties of samples, and setting the regression equations to assess the chemical properties of chili pepper using image textures.
The raw material consisted of the chili pepper belonging to two cultivars, ‘Cyklon’ and ‘Picasso’. The spontaneous lacto-fermentation of whole fruit was carried out using mustard seeds, black pepper, bay leaves, allspice, and 3.5% brine. For raw material and samples after 7, 14, 21, 28, and 56 days of the process, image parameters from the images acquired using a flatbed scanner and chemical properties analyzed by HPLC (high-performance liquid chromatography) were determined. Image processing was performed using MaZda 4.7 software (Łódź University of Technology, Institute of Electronics, Łódź, Poland). Models to classify the raw material and samples after specific periods of lacto-fermentation were developed using MATLABR2024a (MathWorks, Inc., Natick, MA, USA). The differences in means of the analyzed parameters between fresh and lacto-fermented samples, linear correlations, and regression equations were determined using STATISTICA 13.1 (Dell Inc., Tulsa, OK, USA, StatSoft Polska, Kraków, Poland).
The developed models and mean comparison indicated significant changes in the values of selected image textures after the first 7 days of the process, and after the following periods, smaller changes occurred. In the case of samples of both cultivars, fermentation influenced a statistically significant decrease in the content of L-ascorbic acid, glucose, fructose, and total sugars as the processing time increased. For the ‘Cyklon’ cultivar, the strongest correlations were found between total sugars and image texture RS5SV5SumAverg (R=0.995) and fructose and RS5SV5SumAverg (R=0.994). In the case of ‘Picasso’, the strongest correlations were between fructose and GS5SN1SumEntrp (R=-0.997), total sugars and GS5SN1SumEntrp (R=-0.993), and L-ascorbic acid and SSGKurtosis (R=0.993). For both chili pepper cultivars, regression equations and high coefficients of determination were determined.