The fruits of the nopal (Opuntia spp.), cactus pears, are considered functional foods due to their content of nutritive and bioactive substances. Its pulp is generated by numerous seeds that limit their consumption due to their size and hardness and detract from their quality. Other undesirable fruit characteristics are its inedible peel and its short shelf life. In the case of the cactus pear cactus (Opuntia spp.), no cultivar has been reported that produces quality parthenocarpic fruits, nor have they been obtained by artificially inducing parthenocarpy. The objectives of this research were to evaluate the response of 11 genotypes to the induction of parthenocarpy, to characterize the fruits of the genotypes with the best response, and to determine their postharvest life. To induce parthenocarpy, floral buds were used in the pre-anthesis stage, from which the style-stigma, stamens, and tepals were removed, generating a cavity in which a solution of growth regulators was applied (250 mg L⁻¹ AG₃ + 75 mg L ⁻¹ BA + 15 mg L⁻¹ of AIB). A similar number of buds was used as a control, without treatment, and in free pollination. Only two genotypes, MX CP-30 Red and MX CP-40 Yellow, produced parthenocarpic fruits since their empty integuments produced pulp, remaining small, empty, and soft. Their peel was sweet (10–14 °Brix) and edible, and they had a longer shelf life than the corresponding fruits with seeds.

Timely identification of phytosanitary problems in agricultural crops is essential to reduce production losses. Artificial intelligence algorithms facilitate their rapid and reliable identification. In this research, three learning classifiers, namely random forest (RF), support vector machine (SVM) and multilayer perceptron (MLP), were evaluated to identify three target classes (healthy fruit, anthracnose [Colletotrichum spp.] and scab [Sphaceloma perseae]) from digital fruit images. Two color descriptor extraction techniques (region selection and image subsampling) were compared with the RF classifier, and an overall classification accuracy (ACC) of 98±0.03 % with region selection and 84±0.08 % with subsampling was obtained. Subsequently, the classifiers were evaluated with color descriptors extracted with region selection. RF and MLP were superior to SVM, with an ACC of 98±0.03 %. Scab and anthracnose were identified with an F1 score of 98 %. %. The high performance of the classifiers shows the potential for applying artificial intelligence paradigms to identify phytosanitary problems in agricultural crops.

Las variedades de papa (Solanum tuberosum L.) producidas en México son susceptibles a Candidatus Liberibacter solanacearum (CaLso), causante de la enfermedad conocida como ‘manchado interno de la pulpa’, por lo que se requiere conocer la respuesta de genotipos experimentales a la bacteria. El presente estudio tuvo como objetivo evaluar el efecto de la infección de los haplotipos LsoA + LsoB de CaLso en el follaje, la biomasa seca y la calidad de tubérculo de papa, variedad Fianna, una colecta de Solanum demissum y los clones experimentales T90-1-63 y T05-13-21 de Solanum spp. El manchado interno de la pulpa del tubérculo se determinó mediante análisis de imágenes de tubérculos. Las plantas de Fianna mostraron la mayor severidad de daño foliar; en cambio, los clones experimentales presentaron 17 % menos daño foliar que Fianna y 8 % más daño foliar que S. demissum. Este último fue el genotipo con la mayor biomasa seca de hoja y produjo tubérculos de un tamaño pequeño; las plantas infectadas de S. demissum presentaron mayor número de tubérculos y mayor rendimiento de tubérculo fresco que las plantas sin inoculación, aunque tuvieron la menor proporción de superficie sana del tubérculo. El clon T90-1-63 presentó los porcentajes más altos de superficie sana de tubérculo (> 79 %) y las menores intensidades del manchado interno de la pulpa del tubérculo.

Objective: To develop a technique based on the partition method, to estimate the surface area and volume of mango fruit (Mangifera indica L. cv. “Ataulfo”) using two digital orthogonal images, obtained and processed on a mobile device with the Android® operating system, segmented by different discriminants. Design/Methodology/Approach: A technique was developed to estimate the surface area and volume of “Ataulfo” mango fruit based on digital images and was implemented directly on a smartphone with the Android® operating system. Three discriminants were evaluated for segmentation (red channel, green channel, and HIS intensity) and applied in n=30 fruits. The precision and accuracy of the technique was determined. Results: The surface area and volume of the fruit when estimated using the red channel presented a coefficient of variation of less than 2.0% in both variables and the estimation error was less than 3% for surface area and less than 5% for volume. Study Limitations/Implications: This technique was limited to fruit with an elliptical cross-section and smooth surface. Findings/Conclusions: The proposed technique is a non-destructive alternative to estimate the surface area and volume of “Ataulfo” mango fruit with a reasonable margin of error and can be implemented directly on an Android® device using only the red channel as a segmentation discriminant.