Automatic wheat ear counting using machine learning based on RGB UAV imagery.

In wheat (Triticum aestivum L) and other cereals, the number of ears per unit area is one of the main yield-determining components. An automatic evaluation of this parameter may contribute to the advance of wheat phenotyping and monitoring. There is no standard protocol for wheat ear counting in the field, and moreover it is time consuming. An automatic ear-counting system is proposed using machine learning techniques based on RGB (red, green, blue) images acquired from an unmanned aerial vehicle (UAV). Evaluation was performed on a set of 12 winter wheat cultivars with three nitrogen treatments during the 2017-2018 crop season. The automatic system uses a frequency filter, segmentation and feature extraction, with different classification techniques, to discriminate wheat ears in micro-plot images. The relationship between the image-based manual counting and the algorithm counting exhibited high levels of accuracy and efficiency. In addition, manual ear counting was conducted in the field for secondary validation. The correlations between the automatic and the manual in-situ ear counting with grain yield were also compared. Correlations between the automatic ear counting and grain yield were stronger than those between manual in-situ counting and GY, particularly for the lower nitrogen treatment. Methodological requirements and limitations are discussed.

Storage, fertilization and cost properties highlight the potential of dried microbial biomass as organic fertilizer.

The transition to sustainable agriculture and horticulture is a societal challenge of global importance. Fertilization with a minimum impact on the environment can facilitate this. Organic fertilizers can play an important role, given their typical release pattern and production through resource recovery. Microbial fertilizers (MFs) constitute an emerging class of organic fertilizers and consist of dried microbial biomass, for instance produced on effluents from the food and beverage industry. In this study, three groups of organisms were tested as MFs: a high-rate consortium aerobic bacteria (CAB), the microalga Arthrospira platensis ('Spirulina') and a purple non-sulfur bacterium (PNSB) Rhodobacter sp. During storage as dry products, the MFs showed light hygroscopic activity, but the mineral and organic fractions remained stable over a storage period of 91 days. For biological tests, a reference organic fertilizer (ROF) was used as positive control, and a commercial organic growing medium (GM) as substrate. The mineralization patterns without and with plants were similar for all MFs and ROF, with more than 70% of the organic nitrogen mineralized in 77 days. In a first fertilization trial with parsley, all MFs showed equal performance compared to ROF, and the plant fresh weight was even higher with CAB fertilization. CAB was subsequently used in a follow-up trial with petunia and resulted in elevated plant height, comparable chlorophyll content and a higher amount of flowers compared to ROF. Finally, a cost estimation for packed GM with supplemented fertilizer indicated that CAB and a blend of CAB/PNSB (85%/15%) were most cost competitive, with an increase of 6% and 7% in cost compared to ROF. In conclusion, as bio-based fertilizers, MFs have the potential to contribute to sustainable plant nutrition, performing as good as a commercially available organic fertilizer, and to a circular economy.

Presence of amylose crystallites in parboiled rice.

Mildly, intermediately, and severely parboiled Jacinto [16% free amylose (FAM) content] and Puntal (26% FAM content) rice samples were submitted to differential scanning calorimetry (DSC) and wide-angle X-ray diffraction (WAXD). DSC thermograms revealed ungelatinized starch only in mildly parboiled rices and retrograded amylopectin in all parboiled samples. Amylose crystallites were present in intermediately and severely parboiled samples but could not be detected due to their high melting temperature. Nonparboiled and parboiled rice DSC profiles showed only type I and type II amylose-lipid complexes, respectively. Intermediately and severely parboiled rice showed a clear V(h)-type (crystalline amylose-lipid complexes) with a superimposed B-type (retrograded amylopectin and/or amylose crystallites) pattern. The mildly parboiled samples showed a mix of A- (native starch crystallites) and V(h)-type patterns (Puntal) and A-, V(h)-, and B-type patterns (Jacinto). Mild acid hydrolysis destroyed the acid labile retrograded amylopectin crystallites and increased the relative abundance of amylose crystallites. Indeed, acid-hydrolyzed intermediately and severely parboiled samples of both cultivars showed a clear B-type diffraction pattern conclusively showing, for the first time, the presence of amylose crystallites. The melting temperature of the amylose crystallites was ca. 135 degrees C, and melting peaks were visible in the DSC thermograms of the intermediately and severely parboiled samples. Their levels depended on the degree of parboiling and FAM content.

Gelatinization of starch in excess water: beyond the melting of lamellar crystallites. A combined wide- and small-angle X-ray scattering study.

The gelatinization of waxy rice, regular rice, and potato starch suspensions (66% w/w moisture) was investigated by real-time small-angle X-ray scattering (SAXS) and wide-angle X-ray diffraction (WAXD) during heating and by fast ramp differential scanning calorimetry (DSC). The high-angle tail of the SAXS patterns suggested the transition from surface to mass fractal structures in the DSC gelatinization range. Amylose plays a major role in determining the dimensions of the self-similar structures that develop during this process as the characteristic power-law scattering behavior extends to lower scattering angles for regular than for waxy starches. Crystallinity of A-type starches is lost in the temperature region roughly corresponding to the DSC gelatinization range. At the end of the gelatinization endotherm, the B-type potato starch showed residual crystallinity (WAXD), while SAXS-patterns exhibited features of remaining lamellar stacks. Results indicate that the melting of amylopectin crystallites during gelatinization is accompanied by the (exothermic) formation of amorphous networks.

Structural transformations during gelatinization of starches in limited water: combined wide- and small-angle X-ray scattering study.

Rice flour (18-25% moisture) and potato starch (20% moisture) were heated with continuous recording of the X-ray scattering during gelatinization. Rice flours displayed A-type crystallinity, which gradually decreased during gelatinization. The development of the characteristic 9 nm small-angle X-ray scattering (SAXS) peak during heating at sub-gelatinization temperatures indicated the gradual evolution into a stacked lamellar system. At higher temperatures, the crystalline and lamellar order was progressively lost. For potato starch (B-type crystallinity), no 9 nm SAXS peak was observed at ambient temperatures. Following the development of lamellar structures at sub-gelatinization temperatures, B-type crystallinity and lamellar order was lost during gelatinization. On cooling of partially gelatinized potato starch, A-type crystallinity steadily increased, but no formation of stacked lamellar structures was observed. Results were interpreted in terms of a high-temperature B- to A-type recrystallization, in which the lateral movement of double helices was accompanied by a shift along their helical axis. The latter is responsible for the inherent frustration of the lamellar stacks.