As the demand for enabling high-level autonomous driving has increased in recent years and visual perception is one of the critical features to enable fully autonomous driving, in this paper, we introduce an efficient approach for simultaneous object detection, depth estimation and pixel-level semantic segmentation using a shared convolutional architecture. The proposed network model, which we named Driving Scene Perception Network (DSPNet), uses multi-level feature maps and multi-task learning to improve the accuracy and efficiency of object detection, depth estimation and image segmentation tasks from a single input image. Hence, the resulting network model uses less than 850 MiB of GPU memory and achieves 14.0 fps on NVIDIA GeForce GTX 1080 with a 1024×512 input image, and both precision and efficiency have been improved over combination of single tasks.
With the rapid development and widespread applications of Internet of Things (IoT) systems, the corresponding security issues are getting more and more serious. This paper proposes a multistage asymmetric information attack and defense model (MAIAD) for IoT systems. Under the premise of asymmetric information, MAIAD extends the single-stage game model with dynamic and evolutionary game theory. By quantifying the benefits for both the attack and defense, MAIAD can determine the optimal defense strategy for IoT systems. Simulation results show that the model can select the optimal security defense strategy for various IoT systems.
The commercial coconut
shell-activated carbon was modified to change
the number of oxygen-containing functional groups. N2 adsorption/desorption
isotherms, Fourier transform infrared (FT-IR), and Boehm titration
were adopted to describe the physical and chemical properties of the
samples. The adsorption isotherms of CO2 and CH4 on both the unmodified and modified samples were measured. To better
understand the effects of surface oxygen-containing functional groups
on adsorption of CO2 and CH4, the overall adsorption
could be considered as the result of adsorption within the pores and
adsorption onto the oxygen-containing functional groups. Thus, a new
way to understand different adsorption mechanisms by calculation was
proposed. On the basis of the results, there is a significant correlation
between the saturation adsorption capacity of CO2 and the
number of oxygen-containing functional groups, especially carboxyl
and hydroxyl. According to the values of enthalpy (−12.2 to
−20 kJ/mol), it can be known that the adsorption caused by
oxygen-containing functional groups is exothermic and belongs to physisorption.
A semiempirical relationship between the variation of the surface
oxygen-functional groups and the variation of the adsorbed amount
was established. The method proposed in this paper provides a new
way to study the effects of surface functional groups on the adsorption
of CO2 and CH4 and can be even promoted in studying
the adsorption mechanism of other adsorbates.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.