An Approach Using Emerging Optical Technologies and Artificial Intelligence Brings New Markers to Evaluate Peanut Seed Quality

Abstract

Seeds of high physiological quality are defined by their superior germination capacity and uniform seedling establishment. Here, it was investigated whether multispectral images combined with machine learning models can efficiently categorize the quality of peanut seedlots. The seed quality from seven lots was assessed traditionally (seed weight, water content, germination, and vigor) and by multispectral images (area, length, width, brightness, chlorophyll fluorescence, anthocyanin, and reflectance: 365 to 970 nm). Seedlings from the seeds of each lot were evaluated for their photosynthetic capacity (fluorescence and chlorophyll index, F0, Fm, and Fv/Fm) and stress indices (anthocyanin and NDVI). Artificial intelligence features (QDA method) applied to the data extracted from the seed images categorized lots with high and low quality. Higher levels of anthocyanin were found in the leaves of seedlings from low quality seeds. Therefore, this information is promising since the initial behavior of the seedlings reflected the quality of the seeds. The existence of new markers that effectively screen peanut seed quality was confirmed. The combination of physical properties (area, length, width, and coat brightness), pigments (chlorophyll fluorescence and anthocyanin), and light reflectance (660, 690, and 780 nm), is highly efficient to identify peanut seedlots with superior quality (98% accuracy).

Transcripts Expressed during Germination Sensu Stricto Are Associated with Vigor in Soybean Seeds

“…Thus, obtaining soybean seeds with high physiological quality which germinate uniformly and rapidly is paramount for the success of the production of this oilseed. Seeds with superior physiological quality are the input for agriculture [ 3 ] because they are responsible for the successful establishment of the desired stand, which is essential to achieve high productivity [ 4 ].…”

“…Vigor is an important component of the physiological quality [ 5 ] and is responsible for the seed’s resilience [ 6 ]. Thus, vigorous seeds present better physiological performance under non-ideal conditions for their germination (capacity for the new individual to grow from the embryo), and/or germinate faster according to Fonseca de Oliveira et al [ 3 ] and Ebone et al [ 4 ], which leaves the seed exposed to adverse conditions for a shorter time. The most robust and accepted definition of vigor is ‘the sum total of those properties of the seed that determine the potential level of activity and performance of the seed during germination and seedling emergence’ [ 7 , 8 ].…”

“…Therefore, the screening of this property is commonly performed using measurements of germination speed, seedling growth and tolerance to stress, since these characteristics are more pronounced in seeds with a higher level of vigor. In parallel, to help in the diagnosis of more vigorous seeds, other tests can be used, such as membrane integrity, lipid peroxidation level and more recently the use of spectral images that map seed constituents [ 3 , 10 ].…”

Multiomics Analyses Reveal the Dual Role of Flavonoids in Pigmentation and Abiotic Stress Tolerance of Soybean Seeds

“…6 The content of anthocyanins in peanut seeds can reflect the seed quality for germination capacity and uniform seedling establishment. 32 Moreover, the flavonoid content was significantly increased during germination in soybean. 21 Based on our results, the BLK seeds had significantly higher germination potency when imbibed with solutions containing Basta, MV, NaCl, and NaHCO 3 , whereas the germination rate between YL and BLK seeds showed no difference when imbibed in water (Figure 2A,B).…”

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

“…The information generated from multispectral images can be interpreted with artificial intelligence ( Fonseca de Oliveiran et al, 2022 ). This is possible using machine learning algorithms [e.g., random forest (RF), neural network (NN), and support vector machine (SVM)] on the data extracted from the multispectral images.…”