Development and validation of software for estimating physical parameters of tomato fruits using digital images

Iranian Journal of Science and Technology, Transactions of Civil Engineering · Springer Nature · 2026-11-19 · DOI: 10.1007/s42452-026-09831-3

Аннотация

Non-destructive tools for measuring the physical parameters of fruits are increasingly required to support quality control processes throughout the agricultural production chain. The objective of this study was to develop software capable of estimating the surface area, sphericity, and volume of tomato fruits from digital images acquired using conventional RGB cameras. The software, named Tomato Image Metrics (TIM4desk), operated independently of the image acquisition system and was based on image processing techniques and geometric modelling to quantify physical attributes from two-dimensional images. The performance of the software was evaluated using tests conducted with idealised samples, commercially acquired tomatoes, and tomatoes obtained after mechanical harvesting. The parameters estimated by the software were compared with reference values obtained using conventional measurement methods. The comparison between the reference values and those estimated by TIM4desk was performed using linear regression and Bland–Altman analyses. The results demonstrated excellent performance for the idealised samples across all evaluated physical parameters, confirming the effectiveness of the adopted geometric approach under controlled conditions. Among the evaluated parameters, volume estimation showed the highest level of agreement with the reference values. Surface area estimation was more sensitive to irregularities in fruit contours and surface deformations. Furthermore, the physical parameter estimates obtained for commercially sourced tomatoes showed greater agreement with the reference methods than those obtained for mechanically harvested fruits. TIM4desk proved to be an efficient tool for estimation of physical parameters of tomato fruits from digital images through a computational routine that operates independently of the image acquisition system.

Читать у издателя →

Похожие статьи