Evolving topological colour landscape unravels the final stages of pistachio nut development and the incidence of blank nuts.

Pistachio is a major nut crop worldwide; however, there is a lack of standardized non-destructive methods to effectively evaluate maturity and kernel filling for improved management and harvest timing. This study presents an image-based approach to determine pistachio nut maturation and blank kernel incidence by analysing the surface colour patterns of individual nuts at three time points during late development. We identified eight major hull colours to represent the full colour spectrum and applied principal component analysis to divide each nut into seven spatial sections. Within each section, we constructed eight colour-based feature variables (covariates) and associated them with a binary response variable indicating kernel presence or absence. We explored the specific response-covariate relationships at each developmental time point using a data-driven method called categorical exploratory data analysis, which identified key first-order and second-order feature-categories that link hull colour patterns with kernel status. These relationships were visualized using block-structured heatmaps, revealing consistent distinctions between filled and blank nuts. Based on these findings, we developed an algorithm with two main functions: (i) identifying a nut's growth stage from its image for optimal harvest timing and (ii) estimating blank nut incidence for quality assessment and economic decision-making.