Search

Deep learning frameworks(such as deep convolutional networks) require data to have a regular shape. However, discrete features extracted from heterogeneous data cannot be collected in a regular shape to convolute. In this article, a Two-Dimensional Discrete Feature Based Spatial Attention CapsNet(TDACAPS) is proposed to convert one-dimensional discrete features into two-dimensional structured data through Cartesian Product for surface electromyogram(sEMG) signal recognition. sEMG signal varies from person to person is the main signal source of prosthetic control. Our model transforms multi-angle discrete features into structured data to find the inherent law of sEMG signal. Due to uneven information distribution of structured data, this model combines capsule network with attention mechanism to place emphasis on abundant information regions and reduce ancillary information loss. Extensive experiments show our model yields an improvement for sEMG signal recognition of almost 3% than capsule network and other neural networks under different conditions. Our attention mechanism that employs overlapping pooling to search feature map weight is preferable to the squeeze-and-excitation module, convolutional block attention module and others. Moreover, we validate that our model has great expansibility on Wine Quality Dataset and Breast Cancer Wisconsin.

Intelligent manufacturing is a focus of current manufacturing research, and, in combination with the Internet, it enables accurate real-time control of intelligent equipment. Highly accurate real-time prosthesis control has very important applications in therapeutics, intelligent prosthesis, and other fields. However, the applicability of the current electromyogram signal recognition method is not strong because of multiple factors. These include considering one objective (correctness) only and the inability to consider differences of recognition accuracy between actions, to recognize the number of channels, or to recognize computational complexity. In this article, we propose a multi-objective evolutionary algorithm based on a decomposition-based multi-objective differential evolution framework to construct a multi-objective model for electromyogram signals with multiple features and channels. Such channels and features are balanced and selected by using a support vector machine as an electromyogram signal classifier. Results of substantial experiment analyses indicate that the multi-objective electromyogram signal recognition method is superior to the single-objective ant colony algorithm and that the decomposition-based multiobjective evolutionary algorithms with Angle-based updating and global margin ranking is better than the decomposition-based multi-objective evolutionary algorithm and decomposition-based multiobjective evolutionary algorithms with angle-based updating strategy in handling multi-objective models for electromyogram signals.

Phleum paniculatum Huds., commonly known as British timothy, is an increasingly serious weed in wheat field of China. However, the biology of its seed germination and seedling emergence remain unclear. In the present study, the effects of environmental factors on seed germination and seedling emergence of P. paniculatum were explored. Phleum paniculatum seeds had a shallow dormancy (20-30 d) when stored at room temperature (25 ± 5 °C). Seeds could germinate at constant temperatures between 10 and 25 °C, except for 5 or 30 °C. Light was not essential for seed germination, and pH values from 4 to 10 did not inhibit germination. Seeds were moderately adaptable to water potential and NaCl concentration, and germination rates would be decreased 50% when water potential was -0.4 MPa or NaCl concentration was 130 mM. Increased soil burial depth decreased the seedling emergence, and no seeds emerged when the burial depth was more than 4 cm. Taken together, our results provide useful information of the germination and emergence of P. paniculatum seed, and strategies such as proper drainage systems managed and deep plowing are recommended to limit its detrimental effects on agricultural production.

Glyphosate has been used worldwide for nearly 40 years, and 30 types of resistant weeds have been reported. Glyphosate is mass-produced and widely used in China, but few studies and reports on glyphosate-resistant weeds and resistance mechanisms exist. Previous studies found a goosegrass species with high glyphosate resistance from orchards in South China and its glyphosate resistant mechanism was described in this study. The cDNA of 5-enolpyruvylshikimate-3-phosphate synthase (EPSPS, EC 2.5.1.19), the target enzyme of glyphosate, was cloned from the glyphosate-resistant and -susceptible goosegrass, respectively, and referred as EPSPS-R and EPSPS-S. The Pro106 residue was known to be involved in the glyphosate resistance in most goosegrass populations. However, sequence analysis did not find the mutation at the Pro106 residue in the R biotype EPSPS amino acid sequence. The residue 133 and 382 was mutated in the R biotype EPSPS amino acid sequence instead, but it did not affect the EPSPS -S and EPSPS -R genes sensitivities to glyphosate. RT-PCR and Western blot analyses suggested that EPSPS mRNA and protein are mainly present in the shoot tissues both in the R and S goosegrass biotypes. The EPSPS -R rapidly responds to the glyphosate in R-biotype goosegrass and the induced expression was detected at 12 h post glyphosate treatment. The mRNA and protein expression of EPSPS-R increased constantly as the increasing concentration of glyphosate. However, the expression of the EPSPS -S was not induced significantly by glyphosate in the S goosegrass biotype. Quantification of real-time PCR results showed that the copy number of the EPSPS in R-biotype goosegrass was 4.7 times higher than that in the S goosegrass biotype. All the results implied that EPSPS gene amplification might mainly caused the glyphosate resistance of a goosegrass population collected from orchards in South China.