Demosaicking DoFP images using edge compensation method based on correlation.

With the development of nanotechnology, the division of focal plane (DoFP) infrared polarization imaging system with real-time imaging has matured. Meanwhile, the demand for real-time acquisition of polarization information is growing, but the super-pixel structure of the DoFP polarimeter will bring instantaneous field of view (IFoV) errors. Existing polarization demosaicking methods cannot satisfy both accuracy and speed in terms of efficiency and performance. According to the characteristics of DoFP, this paper proposes an edge compensation demosaicking method by analyzing the channel correlations of polarized images. The method performs demosaicing in the differential domain, and the proposed method's performance is verified by comparison experiments using synthetic and authentic polarized images in the near-infrared (NIR) band. The proposed method outperforms the state-of-the-art methods in terms of accuracy and efficiency. It achieves an average peak signal-to-noise ratio (PSNR) improvement of 2 db on public datasets compared to current state-of-the-art methods. A typical 768 × 1024 specification short-wave infrared (SWIR) polarized image can be processed in 0.293s on the Intel Core i7-10870 H CPU, and the technique significantly outperforms various existing demosaicking methods.

Multiexperience-Assisted Efficient Multiagent Reinforcement Learning.

Recently, multiagent reinforcement learning (MARL) has shown great potential for learning cooperative policies in multiagent systems (MASs). However, a noticeable drawback of current MARL is the low sample efficiency, which causes a huge amount of interactions with environment. Such amount of interactions greatly hinders the real-world application of MARL. Fortunately, effectively incorporating experience knowledge can assist MARL to quickly find effective solutions, which can significantly alleviate the drawback. In this article, a novel multiexperience-assisted reinforcement learning (MEARL) method is proposed to improve the learning efficiency of MASs. Specifically, monotonicity-constrained reward shaping is innovatively designed using expert experience to provide additional individual rewards to guide multiagent learning efficiently, with the invariance guarantee of the team optimization objective. Furthermore, a reward distribution estimator is specially developed to model an implicated reward distribution of environment by using transition experience from environment, containing collected samples (state-action pair, reward, and next state). This estimator can predict the expectation reward of each agent for the taken action to accurately estimate the state value function and accelerate its convergence. Besides, the performance of MEARL is evaluated on two multiagent environment platforms: our designed unmanned aerial vehicle combat (UAV-C) and StarCraft II Micromanagement (SCII-M). Simulation results demonstrate that the proposed MEARL can greatly improve the learning efficiency and performance of MASs and is superior to the state-of-the-art methods in multiagent tasks.

Attention Enhanced Reinforcement Learning for Multi agent Cooperation.

In this article, a novel method, called attention enhanced reinforcement learning (AERL), is proposed to address issues including complex interaction, limited communication range, and time-varying communication topology for multi agent cooperation. AERL includes a communication enhanced network (CEN), a graph spatiotemporal long short-term memory network (GST-LSTM), and parameters sharing multi-pseudo critic proximal policy optimization (PS-MPC-PPO). Specifically, CEN based on graph attention mechanism is designed to enlarge the agents' communication range and to deal with complex interaction among the agents. GST-LSTM, which replaces the standard fully connected (FC) operator in LSTM with graph attention operator, is designed to capture the temporal dependence while maintaining the spatial structure learned by CEN. PS-MPC-PPO, which extends proximal policy optimization (PPO) in multi agent systems with parameters' sharing to scale to environments with a large number of agents in training, is designed with multi-pseudo critics to mitigate the bias problem in training and accelerate the convergence process. Simulation results for three groups of representative scenarios including formation control, group containment, and predator-prey games demonstrate the effectiveness and robustness of AERL.

Formation Control With Collision Avoidance Through Deep Reinforcement Learning Using Model-Guided Demonstration.

Generating collision-free, time-efficient paths in an uncertain dynamic environment poses huge challenges for the formation control with collision avoidance (FCCA) problem in a leader-follower structure. In particular, the followers have to take both formation maintenance and collision avoidance into account simultaneously. Unfortunately, most of the existing works are simple combinations of methods dealing with the two problems separately. In this article, a new method based on deep reinforcement learning (RL) is proposed to solve the problem of FCCA. Especially, the learning-based policy is extended to the field of formation control, which involves a two-stage training framework: an imitation learning (IL) and later an RL. In the IL stage, a model-guided method consisting of a consensus theory-based formation controller and an optimal reciprocal collision avoidance strategy is designed to speed up training and increase efficiency. In the RL stage, a compound reward function is presented to guide the training. In addition, we design a formation-oriented network structure to perceive the environment. Long short-term memory is adopted to enable the network structure to perceive the information of obstacles of an uncertain number, and a transfer training approach is adopted to improve the generalization of the network in different scenarios. Numerous representative simulations are conducted, and our method is further deployed to an experimental platform based on a multiomnidirectional-wheeled car system. The effectiveness and practicability of our proposed method are validated through both the simulation and experiment results.

[Clinical efficacy of porcine pulmonary surfactant combined with budesonide suspension intratracheal instillation in the treatment of neonatal meconium aspiration syndrome].

OBJECTIVE: To study the clinical efficacy of porcine pulmonary surfactant (PS) combined with budesonide suspension intratracheal instillation in the treatment of neonatal meconium aspiration syndrome (MAS). METHODS: Seventy neonates with MAS were enrolled for a prospective study. The neonates were randomly assigned to PS alone treatment group and PS+budesonide treatment group (n=35 each). The PS alone treatment group was given PS (100 mg/kg) by intratracheal instillation. The treatment group was given budesonide suspension (0.25 mg/kg) combined with PS (100 mg/kg). RESULTS: The rate of repeated use of PS in the PS+ budesonide group was significantly lower than that in the PS alone group 12 hours after treatment (p<0.05). The improvement of PaO2/FiO2, TcSaO2, PaO2, and PaCO2 in the PS+ budesonide group was significantly greater than that in the PS alone group 6, 12, and 24 hours after treatment (p<0.05). The chest X-ray examination showed that the pulmonary inflammation absorption in the PS+ budesonide group was significantly better than that in the PS alone group 48 hours after treatment (p<0.05). The incidence of complications in the PS+budesonide group was significantly lower than that in the PS alone group (p<0.05), and the average hospitalization duration was significantly shorter than that in the PS alone group (p<0.01). CONCLUSIONS: PS combined with budesonide suspension intratracheal instillation for the treatment of neonatal MAS is effective and superior to PS alone treatment.

[Risk factors for congenital anal atresia].

OBJECTIVE: To investigate the risk factors for the development of congenital anal atresia in neonates. METHODS: A total of 70 neonates who were admitted to 17 hospitals in Foshan, China from January 2011 to December 2014 were enrolled as case group, and another 70 neonates who were hospitalized during the same period and had no anal atresia or other severe deformities were enrolled as control group. Univariate and multivariate logistic regression analyses were used to investigate the risk factors for the development of congenital anal atresia. RESULTS: The univariate analysis revealed that the age of mothers, presence of oral administration of folic acid, infection during early pregnancy, and polyhydramnios, and sex of neonates showed significant differences between the case and control groups (P<0.05). The multivariate logistic regression analysis revealed that infection during early pregnancy (OR=18.776) and male neonates (OR=9.304) were risk factors for congenital anal atresia, and oral administration of folic acid during early pregnancy was the protective factor (OR=0.086). CONCLUSIONS: Infection during early pregnancy is the risk factor for congenital anal atresia, and male neonates are more likely to develop congenital anal atresia than female neonates. Supplementation of folic acid during early pregnancy can reduce the risk of congenital anal atresia.