MCSNet+: Enhanced Convolutional Neural Network for Detection and Classification of Tribolium and Sitophilus Sibling Species in Actual Wheat Storage Environments.

Insect pests like Tribolium and Sitophilus siblings are major threats to grain storage and processing, causing quality and quantity losses that endanger food security. These closely related species, having very similar morphological and biological characteristics, often exhibit variations in biology and pesticide resistance, complicating control efforts. Accurate pest species identification is essential for effective control, but workplace safety in the grain bin associated with grain deterioration, clumping, fumigator hazards, and air quality create challenges. Therefore, there is a pressing need for an online automated detection system. In this work, we enriched the stored-grain pest sibling image dataset, which includes 25,032 annotated Tribolium samples of two species and five geographical strains from real warehouse and another 1774 from the lab. As previously demonstrated on the Sitophilus family, Convolutional Neural Networks demonstrate distinct advantages over other model architectures in detecting Tribolium. Our CNN model, MCSNet+, integrates Soft-NMS for better recall in dense object detection, a Position-Sensitive Prediction Model to handle translation issues, and anchor parameter fine-tuning for improved matching and speed. This approach significantly enhances mean Average Precision (mAP) for Sitophilus and Tribolium, reaching a minimum of 92.67 ± 1.74% and 94.27 ± 1.02%, respectively. Moreover, MCSNet+ exhibits significant improvements in prediction speed, advancing from 0.055 s/img to 0.133 s/img, and elevates the recognition rates of moving insect sibling species in real wheat storage and visible light, rising from 2.32% to 2.53%. The detection performance of the model on laboratory-captured images surpasses that of real storage facilities, with better results for Tribolium compared to Sitophilus. Although inter-strain variances are less pronounced, the model achieves acceptable detection results across different Tribolium geographical strains, with a minimum recognition rate of 82.64 ± 1.27%. In real-time monitoring videos of grain storage facilities with wheat backgrounds, the enhanced deep learning model based on Convolutional Neural Networks successfully detects and identifies closely related stored-grain pest images. This achievement provides a viable solution for establishing an online pest management system in real storage facilities.

Jet fuel and hydrogen produced from waste plastics catalytic pyrolysis with activated carbon and MgO.

Catalytic pyrolysis of waste plastics to produce jet fuel and hydrogen using activated carbon and MgO as catalysts was studied. The effects of catalyst to waste plastics ratio experimental temperature, catalyst placement and activated carbon to MgO ratio on the yields and distributions of pyrolysis products were studied. The placement of catalysts played an important role on the catalytic pyrolysis of LDPE, and the pyrolytic volatiles first flowing through MgO and then biomass-derived activated carbon (BAC) could obtain an excellent result to produce H2 and jet fuel-rich products. The higher pyrolysis temperature converted diesel range alkanes into jet fuel range alkanes and promoted the aromatization of alkanes to generate aromatic hydrocarbons. BAC and MgO as catalysts had excellent performance in catalytic conversion of LDPE to produce hydrogen and jet fuel. 100 area.% jet fuel range products can be obtained in LDPE catalytic pyrolysis under desired experimental conditions. The combination of BAC and MgO as catalysts had a synergy effect on the gaseous product distribution and promoted the production of hydrogen, and up to 94.8 vol% of the obtained gaseous components belonged to hydrogen. This work provided an effective, convenient and economical pathway to produce jet fuel and hydrogen from waste plastics.