This paper presents a novel autonomous drone-based smoke plume tracking system capable of navigating and tracking plumes in highly unsteady atmospheric conditions. The system integrates advanced hardware and software and a comprehensive simulation environment to ensure robust performance in controlled and real-world settings. The quadrotor, equipped with a high-resolution imaging system and an advanced onboard computing unit, performs precise maneuvers while accurately detecting and tracking dynamic smoke plumes under fluctuating conditions. Our software implements a two-phase flight operation: descending into the smoke plume upon detection and continuously monitoring the smoke's movement during in-plume tracking. Leveraging Proportional Integral-Derivative (PID) control and a Proximal Policy Optimization (PPO) based Deep Reinforcement Learning (DRL) controller enables adaptation to plume dynamics. Unreal Engine simulation evaluates performance under various smoke-wind scenarios, from steady flow to complex, unsteady fluctuations, showing that while the PID controller performs adequately in simpler scenarios, the DRL-based controller excels in more challenging environments. Field tests corroborate these findings. This system opens new possibilities for drone-based monitoring in areas like wildfire management and air quality assessment. The successful integration of DRL for real-time decision-making advances autonomous drone control for dynamic environments.

This article investigates the use of deep knockoff, a modern statistical variable selection methodology, to uncover the spectral signatures of marine debris. This method uses a generative model by leveraging deep neural networks to learn the high dimensional distribution of reflectance in visible to near infrared wavelengths. To that end, a public dataset obtained from ground-based labeling of the observations by the Multispectral Instrument (MSI) on-board Sentinel-2 satellite is used. Through controlling the false discovery rate, consistent with the known physical causalities, the results indicate that the near infrared (band 8, 833 nm) and red (band 4, 665 nm) are the most important bands, respectively for discrimination of marine (plastic) debris from the background water. In the presence of denseSargassummacroalgae, the deep knockoff isolates the green (band 3, 560 nm) and the narrow NIR (band 8a, 864.7 nm) as another important band.

We present a method for Chinese character component recognition. We use an attributed relational graph model to describe a component. This model allows us to express the knowledge of the component shape and can include stable features of the component in various writing styles and various characters. We then describe a method to extract graph representation of an input character. The graph models are used to recognize components from a whole character by subgraph isomorphism. Our method need not segment component from character and can tolerate links and overlaps between a component and the other part. Experiments for handwritten Chinese characters show its efficiency.

Now all the block based skeletonizing algorithms only use the compressed line adjacency graph scanned in one direction. For lines approximately parallel to the scan direction, there is difficulty extracting the skeleton because such a line may be separated into several graph nodes or mixed with some pixels of other lines in the intersection point. In this paper, we propose a new skeletonizing method by combining c-LAG of horizontal and vertical direction. The main idea of the method is a rule for the skeletons in horizontal and vertical direction c-LAG and knowledge based node validation. The validation makes full use of global information in the image. It has been tested on a large amount of characters and high quality achieved.

It is argued that some failure in the existing expert systems mainly results from the ill-organized knowledge bases. The authors propose an effective structure for a knowledge base called a solution-oriented structure, which is different from various existing knowledge base structures. Using several strategies of machine learning, an expert system tool called THOUGHT can reorganize the systematic structure into a solution-oriented structure. On the basis of the solution-oriented structure, THOUGHT can perform the three-level inference for solving different kinds of problems. The solution-oriented structure of the knowledge base is efficient and effective on large, complicated knowledge bases and time critical situations.>