Classification of Fluorescently Labelled Maize Kernels Using Convolutional Neural Networks.

Accurate real-time classification of fluorescently labelled maize kernels is important for the industrial application of its advanced breeding techniques. Therefore, it is necessary to develop a real-time classification device and recognition algorithm for fluorescently labelled maize kernels. In this study, a machine vision (MV) system capable of identifying fluorescent maize kernels in real time was designed using a fluorescent protein excitation light source and a filter to achieve optimal detection. A high-precision method for identifying fluorescent maize kernels based on a YOLOv5s convolutional neural network (CNN) was developed. The kernel sorting effects of the improved YOLOv5s model, as well as other YOLO models, were analysed and compared. The results show that using a yellow LED light as an excitation light source combined with an industrial camera filter with a central wavelength of 645 nm achieves the best recognition effect for fluorescent maize kernels. Using the improved YOLOv5s algorithm can increase the recognition accuracy of fluorescent maize kernels to 96%. This study provides a feasible technical solution for the high-precision, real-time classification of fluorescent maize kernels and has universal technical value for the efficient identification and classification of various fluorescently labelled plant seeds.

Cold-Rolled Strip Steel Stress Detection Technology Based on a Magnetoresistance Sensor and the Magnetoelastic Effect.

Driven by the demands for contactless stress detection, technologies are being used for shape control when producing cold-rolled strips. This paper presents a novel contactless stress detection technology based on a magnetoresistance sensor and the magnetoelastic effect, enabling the detection of internal stress in manufactured cold-rolled strips. An experimental device was designed and produced. Characteristics of this detection technology were investigated through experiments assisted by theoretical analysis. Theoretically, a linear correlation exists between the internal stress of strip steel and the voltage output of a magneto-resistive sensor. Therefore, for this stress detection system, the sensitivity of the stress detection was adjusted by adjusting the supply voltage of the magnetoresistance sensor, detection distance, and other relevant parameters. The stress detection experimental results showed that this detection system has good repeatability and linearity. The detection error was controlled within 1.5%. Moreover, the intrinsic factors of the detected strip steel, including thickness, carbon percentage, and crystal orientation, also affected the sensitivity of the detection system. The detection technology proposed in this research enables online contactless detection and meets the requirements for cold-rolled steel strips.

Research on the Hot Straightening Process of Medium-Thick Plates Based on Elastic-Viscoplastic Material Modeling.

With the continuous improvement in the strength of medium-thick plate materials, the hot straightening of plates at high temperatures is increasingly influencing the final defect characteristics of products. In the high-temperature hot straightening process, the temperature and straightening speed of the plate significantly influence its intrinsic material properties, which, in turn, affect the straightening characteristics of the plate. However, most current material models used in the straightening process do not consider the relationship between temperature and strain rate, which leads to an inaccurate characterization of the actual material structure. Additionally, the continuous reverse bending mechanics model for straightening does not account for the impact of different bending strain rates on the bending characteristics of the plate in the thickness direction. In this study, a numerical calculation method was employed to investigate the evolution process of stress and curvature in the roll-type hot straightening process of medium-thick plates. Experimental data and mathematical methods were utilized to develop a viscous plastic material model that accounted for temperature and strain rate. Furthermore, a cross-sectional continuous reverse bending model was established, taking into account the temperature and straightening speed, enabling a reasonable interpretation of the mechanical parameter behaviors of medium-thick plates during high-temperature straightening.

Eco-Friendly Phenol-Urea-Formaldehyde Co-condensed Resin Adhesives Accelerated by Resorcinol for Plywood Manufacturing.

Eco-friendly and good-performance resin adhesives are needed for wood manufacturing. In this study, phenol-urea-formaldehyde (PUF) resin adhesives were modified by adding various ratios of resorcinol. The properties of PUF, phenol-resorcinol-urea-formaldehyde (PRUF) resin adhesives, and the performances of the prepared plywood were tested. The curing behaviors and the structural features of the PUF and PRUF resin adhesives were investigated via dynamic scanning calorimetry, thermal gravimetric analysis, 13C NMR, and cross polarization magic angle spinning 13C NMR. The results indicated that 1.3% resorcinol (based on resin, w/w) could decrease the curing temperature and accelerate the curing process after PUF resin modification. The PRUF resin adhesives demonstrated a lower activation energy during the curing process, with up to 28.8% less energy than that of PUF resin adhesive without any curing agent. The plywood demonstrated low formaldehyde emissions (<0.1 mg L-1) and acceptably high bonding strengths (>1.00 MPa). This work provided a method for preparing an easy-cured and high-performance phenolic resin for wood manufacturing.

Principal curvature effects on the early evolution of three-dimensional single-mode Richtmyer-Meshkov instabilities.

The Richtmyer-Meshkov instability (RMI) of single-mode air-SF(6) interfaces is studied numerically and the emphasis is placed on the effect of the principal curvature on the early evolution of the shocked interface. Two three-dimensional initial interfaces with opposite (3D-) and identical (3D+) principal curvatures and a traditional two-dimensional interface (2D) are considered. The weighted essentially nonoscillatory scheme and the Level-Set method combined with the real ghost fluid method are adopted. For comparison, perturbations on the initial interfaces with the same wavelength and amplitude in the symmetry plane are employed. The numerical results confirm the experimental finding that the growth rate of perturbations in the symmetry plane at the linear stage in the 3D- case is much smaller than that in the 2D and 3D+ cases. The difference among them can be ascribed to the different pressure and vorticity distributions associated with the principal curvatures of the initial interface. On the one hand, the high-pressure zones in the vicinity of the deformed interface are significantly different for three cases especially in the very beginning. The shock convergence and divergence at the interface are more severe in the 3D+ case than those in the 2D case, while the wave pattern in the 3D- case is more complex. On the other hand, the baroclinic vorticity distribution plays a leading role in the interface deformation of the 3D RMI after the passage of the planar shock. The accumulated vorticity changes the movement of the deformed interface and makes the local growth of perturbations different among three cases.

Richtmyer-Meshkov instability of a three-dimensional SF_{6}-air interface with a minimum-surface feature.

The Richmyer-Meshkov instability of a three-dimensional (3D) SF_{6}-air single-mode interface with a minimum-surface feature is investigated experimentally. The interface produced by the soap film technique is subjected to a planar shock and the evolution of the shocked interface is captured by time-resolved schlieren photography. Different from the light-heavy single-mode case, a phase inversion occurs in the shock-interface interaction and a bubblelike structure is observed behind the shocked interface, which may be ascribed to the difference in pressure perturbation at different planes. The superimposition of spikelike forward-moving jets forms a complex structure, indicating a distinctly 3D effect. Quantitatively, it is also found that the instability at the symmetry plane grows much slower than the prediction of two-dimensional linear model, but matches the extended 3D linear and nonlinear models accounting for the curvature effects. Therefore, the opposite curvatures of the 3D interface are beneficial for suppressing the growth of the instability.