Highlights A method that can separate homochromatic FM in mixed congee was proposed. Spectral features of FM and mixed congee were extracted to build recognition model. The SVM model achieved high identification rates (99.17%) for homochromatic FM. The proposed method is better than computer vision in separating homochromatic FM.

To fundamentally explore the correlation between crop nutritional quality and mineral elements, specific tests and accurate analysis were carried out on all organs and root soil of watermelon (Citrullus lanatus). The results showed that the distribution patterns of nitrogen and phosphorus at watermelon maturity were similar, and the average nitrogen and phosphorus contents were in the orders of leaf and seed > stem, peel, root > pulp > root soil, and peel > seed > root, stem, leaf > root soil > pulp, respectively. From the perspective of element geochemistry, biophile and lithophile elements had the strongest correlation with nitrogen and phosphorus, and watermelon did not antagonize soil nitrogen and phosphorus uptake. The prediction model of nitrogen translocation factor in watermelon organs to total acid was established by partial least squares with R2 = 0.81. Significantly, when the isometric log-ratio of nitrogen to phosphorus in watermelon leaves was 1.97 to 2.19, the watermelon pulp showed better quality with total acid > 0.5%, total sugar > 5% and soluble solids > 10%. Therefore, the characterization of nitrogen and phosphorus in watermelon leaves can serve as a non-destructive analysis to predict watermelon fruit quality.