Towards complex scenes: A deep learning-based camouflaged people detection method for snapshot multispectral images

Camouflaged people are extremely expert in actively concealing themselves by effectively utilizing cover and the surrounding environment. Despite advancements in optical detection capabilities through imaging systems, including spectral, polarization, and infrared technologies, there is still a lack of effective real-time method for accurately detecting small-size and high-efficient camouflaged people in complex real-world scenes. Here, this study proposes a snapshot multispectral image-based camouflaged detection model, multispectral YOLO(MS-YOLO), which utilizes the SPD-Conv and Sim AM modules to effectively represent targets and suppress background interference by exploiting the spatial-spectral target information. Besides, the study constructs the first real-shot multispectral camouflaged people dataset(MSCPD), which encompasses diverse scenes, target scales, and attitudes. To minimize information redundancy, MS-YOLO selects an optimal subset of 12 bands with strong feature representation and minimal inter-band correlation as input. Through experiments on the MSCPD, MS-YOLO achieves a mean Average Precision of 94.31% and real-time detection at 65 frames per second, which confirms the effectiveness and efficiency of our method in detecting camouflaged people in various typical desert and forest scenes. Our approach offers valuable support to improve the perception capabilities of unmanned aerial vehicles in detecting enemy forces and rescuing personnel in battlefield.

CFD Investigation of Diffusion Law and Harmful Boundary of Buried Natural Gas Pipeline in the Mountainous Environment

The leakage gas from a buried natural gas pipelines has the great potential to cause economic losses and environmental pollution owing to the complexity of the mountainous environment.In this study,computational fluid dynamics(CFD)method was applied to investigate the diffusion law and hazard range of buried natural gas pipeline leakage in mountainous environment.Based on cloud chart,concentration at the monitoring site and hazard range of lower explosion limit(LEL)and upper explosion limit(UEL),the influences of leakage hole direction and shape,soil property,burial depth,obstacle type on the diffusion law and hazard range are analyzed.Results show that the leakage gas is not radially diffused until it reaches the ground,and the velocity of gas diffusion to the ground and the hazard range decrease as the angle between the leaking direction and the buoyancy direction increases.Triangular and square leak holes have a faster diffusion rate and a wider hazard range than circular.The diffusion rate of leakage gas in soil rises as soil granularity and porosity increase.The time of leakage gas diffusion to the ground increases significantly with the increase of burial depth,and the hazard range reduces as burial depth increases.Boulder-type obstacles will alter the diffusion path of the leakage gas and accelerate the expansion of the hazard distance,while trench-type obstacles will cause the natural gas to accumulate in the trench and form a high concentration region slowing the expansion of the surface gas concentration.

Impact Damage Testing Study of Shanxi-Beijing Natural Gas Pipeline Based on Decision Tree Rotary Tiller Operation

The North China Plain and the agricultural region are crossed by the Shanxi-Beijing natural gas pipeline.Resi-dents in the area use rototillers for planting and harvesting;however,the depth of the rototillers into the ground is greater than the depth of the pipeline,posing a significant threat to the safe operation of the pipeline.Therefore,it is of great significance to study the dynamic response of rotary tillers impacting pipelines to ensure the safe opera-tion of pipelines.This article focuses on the Shanxi-Beijing natural gas pipeline,utilizingfinite element simulation software to establish afinite element model for the interaction among the machinery,pipeline,and soil,and ana-lyzing the dynamic response of the pipeline.At the same time,a decision tree model is introduced to classify the damage of pipelines under different working conditions,and the boundary value and importance of each influen-cing factor on pipeline damage are derived.Considering the actual conditions in the hemp yam planting area,targeted management measures have been proposed to ensure the operational safety of the Shanxi-Beijing natural gas pipeline in this region.