A Reliable Method to Recognize Soybean Seed Maturation Stages Based on Autofluorescence-Spectral Imaging Combined With Machine Learning Algorithms

Abstract

In recent years, technological innovations have allowed significant advances in the diagnosis of seed quality. Seeds with superior physiological quality are those with the highest level of physiological maturity and the integration of rapid and precise methods to separate them contributes to better performance in the field. Autofluorescence-spectral imaging is an innovative technique based on fluorescence signals from fluorophores present in seed tissues, which have biological implications for seed quality. Thus, through this technique, it would be possible to classify seeds in different maturation stages. To test this, we produced plants of a commercial cultivar (MG/BR 46 “Conquista”) and collected the seeds at five reproductive (R) stages: R7.1 (beginning of maturity), R7.2 (mass maturity), R7.3 (seed disconnected from the mother plant), R8 (harvest point), and R9 (final maturity). Autofluorescence signals were extracted from images captured at different excitation/emission combinations. In parallel, we investigated physical parameters, germination, vigor and the dynamics of pigments in seeds from different maturation stages. To verify the accuracy in predicting the seed maturation stages based on autofluorescence-spectral imaging, we created machine learning models based on three algorithms: (i) random forest, (ii) neural network, and (iii) support vector machine. Here, we reported the unprecedented use of the autofluorescence-spectral technique to classify the maturation stages of soybean seeds, especially using the excitation/emission combination of chlorophyll a (660/700 nm) and b (405/600 nm). Taken together, the machine learning algorithms showed high performance segmenting the different stages of seed maturation. In summary, our results demonstrated that the maturation stages of soybean seeds have their autofluorescence-spectral identity in the wavelengths of chlorophylls, which allows the use of this technique as a marker of seed maturity and superior physiological quality.

Fungal identification in peanuts seeds through multispectral images: Technological advances to enhance sanitary quality

“…and A. flavus enhanced the brightness of the seeds (Figure 2C). This behavior contributed to the increase in reflectance (Fonseca de Oliveira et al, 2022); ii) the formation of mycelia and toxicogenic compounds alters the chemical composition of infected seeds (Zhang et al, 2022), which can reduce light absorption and favor reflection enhancement in the spectral range (Figure 3C); iii) fungal colonization intensifies the production of anthocyanins as a plant response to tissue deterioration (Liu et al, 2018;Fonseca de Oliveira et al, 2022). This pigment, is also produced by fungi (Bu et al, 2020;Sicilia et al, 2021) and its accumulation in the contaminated seeds (Figure 6) may have contributed to the lower reflectance found (Figure 3C).…”

“…In peanut seeds, information acquired at various wavelengths has made it possible to diagnose pathologies noninvasively (Qiao et al, 2017;Ziyaee et al, 2021). Some multispectral descriptors such as color (CIELab L*) and reflectance have high sensitivity for detecting the post-harvest decay status of peanut seeds (Fonseca de Oliveira et al, 2022). For instance, in a seed of an oleaginous species (J. curcas), textural parameters obtained from multispectral images make it possible to diagnose gray-level nuances associated with fungal contamination (Bianchini et al, 2021).…”

“…This information can be valuable for improving the characterization of the sanitary quality of the peanut seeds. In the case of the color and reflectance parameters, the brightness characteristics of the seed surface coat are indicative of aging of the embryonic tissues (Fonseca de Oliveira et al, 2022) or infection by pathogenic organisms (Boelt et al, 2018). The seed reflectance is altered in the presence of the pathogen mycelia, which makes it possible to distinguish fungal species with high accuracy (Franca-Silva et al, 2020).…”

“…Machine learning is a digital analytical resource for predicting patterns based on numerical data (Medeiros et al, 2020a;Medeiros et al, 2020b). It is a branch of artificial intelligence with wide application, including studies with seeds (Barboza da Silva et al, 2021b;Batista et al, 2022;Fonseca de Oliveira et al, 2022). The simplicity and high efficiency of these features can reduce analysis time (Elmasry et al, 2019a).…”

Integrating optical imaging techniques for a novel approach to evaluate Siberian wild rye seed maturity

“…However, with advances in spectroscopy and computational technologies, non-destructive identification of seed characteristics is now possible through X-ray analysis ( de Medeiros et al., 2020b ), multispectral and hyperspectral image analysis ( Xia et al., 2019 ), microtomography ( Gomes-Junior et al., 2019 ), magnetic resonance ( Melchinger et al., 2017 ), and other techniques. Recently, seed maturity analysis using multispectral imaging technology and ML methods has been applied to soybean ( Glycine max L.) seed harvesting ( Batista et al., 2022 ).…”