Relationship of Textures from Tomato Fruit Images Acquired Using a Digital Camera and Lycopene Content Determined by High-Performance Liquid Chromatography

This study aimed at correlating image features with the lycopene content of tomato fruit. Tomato cultivars with different fruit colors, such as ‘Ożarowski’ (yellow), ‘Marvel Striped’ (yellow-orange-pink), ‘Green Zebra’ (green), Sandoline F1 (red), Cupidissimo F1 (red), and Sacher F1 (brown) were selected for the study. The tomato fruits were imaged using a digital camera. The texture parameters were computed from the images converted to color channels <i>R</i>, <i>G</i>, <i>B</i>, <i>L</i>, <i>a</i>, <i>b</i>, <i>X</i>, <i>Y</i>, and <i>Z</i> based on the histogram, autoregressive model, gradient map, co-occurrence matrix, and run-length matrix. Lycopene content was determined using high-performance liquid chromatography (HPLC). Pearson’s correlation coefficients (R), regression equations, and coefficients of determination (R²) were determined. The lycopene content in fruit ranged from 0.31 mg 100 g⁻¹ for ‘Green Zebra’ to 11.83 mg 100 g⁻¹ for Sacher F1. The correlation coefficient (R) between lycopene content and selected image textures reached −0.99 for selected textures from color channels <i>G</i>, <i>b</i>, and <i>Y</i>. The highest positive correlation (R parameter equal to 0.98) was obtained for texture from color channel <i>Y</i>. Based on the individual color channel providing the highest results, one texture was selected for the determination of regression equations. Coefficients of determination (R²) of 0.99 were obtained for texture from color channel <i>G</i>. The regression equations may be used in practice for nondestructive, objective, and precise estimation of the lycopene content in tomato fruit.