New computer vision solutions combined with artificial intelligence algorithms can help recognize patterns in biological images, reducing subjectivity and optimizing the analysis process. the aim of this study was to propose an approach based on interactive and traditional machine learning methods to classify soybean seeds and seedlings according to their appearance and physiological potential. in addition, we correlated the appearance of seeds to their physiological performance. images of soybean seeds and seedlings were used to develop models using low-cost approaches and free-access software. The models developed showed high performance, with overall accuracy reaching 0.94 for seeds and seedling classification. The high precision of the models that were developed based on interactive and traditional machine learning demonstrated that the method can easily be used to classify soybean seeds according to their appearance, as well as to classify soybean seedling vigor quickly and non-subjectively. the appearance of soybean seeds is strongly correlated with their physiological performance.

Automated analysis of seed vigor stands out by allowing greater accuracy, standardization, objectivity, and speed in evaluation of the physiological potential of seed lots. The objective of this study was to evaluate the efficiency of the Vigor-S® system in assessing the physiological quality of common bean seeds compared to the information provided by the traditional vigor tests recommended for this species. Four genotypes of common bean were used, each one represented by four seed lots. Characterization of the physiological potential of the lots was carried out by the following tests: germination, first count of germination, seedling emergence, accelerated aging, and electrical conductivity. The results of these tests were compared with the data obtained from the image analysis technique, specifically the Vigor-S® system, which was used to evaluate seedling growth at two, three, and four days after the beginning of the germination test. Shoot length, primary root length, and seedling length were measured, as well as the growth index, uniformity index, and vigor index were calculated. Computerized analysis of seedling images using the Vigor-S® software is a reliable alternative for evaluation the physiological potential of bean seeds, and it produces information similar to evaluations traditionally used for that purpose.

No abstract

No abstract

The application of desiccant herbicides in the bean crop is fundamental in the production of quality seeds, since it anticipates the harvesting season, which makes it possible to obtain seeds in the period of physiological maturation, when they are reported with maximum accumulation of dry matter, high vigor levels and germination percentage. However, there is little information about the ideal application time of these products to obtain beans of high physiological quality. Therefore, the objective of this study was to evaluate the physiological quality of bean seeds after application of desiccant herbicides at different times in preharvest. The experimental design was a randomized complete block design in a 2x3+1 factorial scheme, with four replications. Desiccant herbicides paraquat (400g ha-1) and paraquat + diuron (200 + 400g ha-1) were applied in three phenological stages (R8, R8/R9 and R9), plus one control (without herbicide application). Yield (Kg ha-1), mass of one thousand seeds (g), germination (%) and seed vigor were evaluated through tests of accelerated aging, electrical conductivity and primary root length. Application of paraquat molecule at the R8 stage and the paraquat + diuron mixture at the R8/R9 stage reduced the viability and vigor of the bean seeds, and compromised yield. Applications of the paraquat herbicide at the R8/R9 and R9 stages or of the mixture (paraquat + diuron) at the R9 stage preserved the physiological quality of the seeds without;however, affecting yield.