Asymmetric Convolution: An Efficient and Generalized Method to Fuse Feature Maps in Multiple Vision Tasks.

Fusing features from different sources is a critical aspect of many computer vision tasks. Existing approaches can be roughly categorized as parameter-free or learnable operations. However, parameter-free modules are limited in their ability to benefit from offline learning, leading to poor performance in some challenging situations. Learnable fusing methods are often space-consuming and timeconsuming, particularly when fusing features with different shapes. To address these shortcomings, we conducted an in-depth analysis of the limitations associated with both fusion methods. Based on our findings, we propose a generalized module named Asymmetric Convolution Module (ACM). This module can learn to encode effective priors during offline training and efficiently fuse feature maps with different shapes in specific tasks. Specifically, we propose a mathematically equivalent method for replacing costly convolutions on concatenated features. This method can be widely applied to fuse feature maps across different shapes. Furthermore, distinguished from parameter-free operations that can only fuse two features of the same type, our ACM is general, flexible, and can fuse multiple features of different types. To demonstrate the generality and efficiency of ACM, we integrate it into several state-of-the-art models on three representative vision tasks: visual object tracking, referring video object segmentation, and monocular 3D object detection. Extensive experimental results on three tasks and several datasets demonstrate that our new module can bring significant improvements and noteworthy efficiency.

[Potential Source Identification and Ecological Risk Assessment of Heavy Metals in Surface Soil of Heze Oil Peony Planting Area Based on PMF-PCA/APCS and PERI].

To study the sources and potential risks of heavy metals in soils of characteristic agricultural product producing areas is of great significance for the scientific management and safe utilization of soil and crop resources. The contents of heavy metals As, Cd, Cr, Cu, Hg, Ni, Pb, and Zn in the 254 surface soil samples collected from the Heze oil peony planting area were determined. The content characteristics and correlation of heavy metals were analyzed using multivariate statistical methods. The sources of heavy metals in topsoil were analyzed using Igeo, PMF, and PCA/APCS. The ecological risks of the eight heavy metals were assessed through the potential ecological risk index (PERI). The results showed that the average contents of seven heavy metals in the soil were basically consistent with the background values of soil elements in Heze City, except that the average value of Cd was 1.44 times higher than the background value in Heze City. Correlation analysis and cluster analysis revealed that Pb, Hg, and Cd elements in the soil were greatly affected by human activities in the later period. The sources of eight heavy metals in the study area were natural sources, agricultural fertilizer sources, industrial coal sources, and domestic transportation sources, with the contribution rates of 81.31%, 15.45%, 2.74%, and 0.50%, respectively; 84.25% of the sites in the study area were at slight ecological risk, whereas the moderate risk and strong risk sites accounted for 14.96% and 0.79%, respectively. Among them, Cd and Hg were the dominant elements of ecological risk in the study area.

Deoxygenation of oximes for the synthesis of pyrrolines via hydroimination cyclization.

Herein, we report the generation of iminyl radicals through photocatalytic deoxygenation of oximes with trivalent phosphine. The hydroimination reaction proceeded via ß-scission of a phosphoranyl radical, followed by 5-exo-trig cyclization of the resulting iminyl radical. This protocol transforms oximes, including alkyl oximes, into a variety of pyrrolines in moderate to good yields. A radical clock experiment confirmed the formation of a cyclic radical and supported our reaction design.

Photoredox Catalytic Phosphine-Mediated Deoxygenation of Hydroxylamines Enables the Construction of N-Acyliminophosphoranes.

The chemistry of phosphoranyl radicals has received increasing attention in recent years. Here, we report the generation of amidyl radicals through photocatalytic deoxygenation of hydroxylamines with triphenylphosphine. This methodology offers a novel and convenient route to a diverse range of N-acyliminophosphoranes in moderate to good yields under visible-light photoredox conditions. Fluorescence quenching experiments suggest that the excited-state of the organic photocatalyst (4CzIPN) was oxidatively quenched by a Cu(II) salt.

Synergetic Effect of Water, Temperature, and Pressure on Methane Adsorption in Shale Gas Reservoirs.

Adsorption is one of the most important forms of storage of gas in shale reservoirs. Shale gas adsorption in the actual reservoir is not only affected by individual factors such as water content, temperature, and pressure but also by the synergetic effect of these factors. In this study, we conducted laboratory experiments on methane adsorption in dry and wet shale at different pressures and temperatures. The synergetic effect of water content, temperature, and pressure on shale gas adsorption is explored. The results show that increasing temperature weakens the interaction between methane and shale and reduces adsorption capacity due to the exothermic nature of adsorption. Water reduces methane adsorption capacity by occupying adsorption sites and blocking pores in the shale system. Although temperature and water reduce methane adsorption individually, the effect of these two factors weakens each other. Temperature has a more significant effect on methane adsorption in shales with low water content, while water has a more remarkable impact on methane adsorption at a low temperature. Furthermore, the increase in pressure reduces the negative influence of water and temperature on methane adsorption. By quantitatively analyzing the relationship between methane adsorption in dry and wet shales, a predictive adsorption model for wet shale considering the influence of in situ conditions is proposed and validated. Validation shows that the proposed model has high accuracy and broad applicability to shales with different properties.