Enhanced CenterTrack for Robust Underwater Multi-Fish Tracking.

Most multi-object tracking methods have achieved good results in tracking multiple pedestrians with Kalman filter, but their tracking performance in crowded scenes is still poor due to pedestrian avoidance and frequent occlusion. In crowded scenes, the pedestrian trajectory prediction with Kalman filter alone is unreliable. In this paper, a two-dimensional field-of-view avoidance force model (AFM) is proposed to assist the Kalman filter prediction by sensing the avoidance force and then complete pedestrian tracking. In the model, each pedestrian has a two-dimensional field of view to perceive the avoidance force, which determines the next predicted trajectory. In real scenes, pedestrians tend to be more concerned about the surrounding area, so different areas are set to simulate the attention mechanisms of pedestrians in real scenes. In the FairMOT model, AFM is used to optimize the pedestrian state values of Kalman filter prediction and the optimized model is trained on the MOT16 public dataset. The experimental results on the MOT20 benchmark dataset show that compared with the mainstream tracking model FairMOT, our method respectively improves MOTA by 2.7% and IDF1 by 2.2%. Our method also achieves the good performance on MOT15, MOT16, and MOT17 tracking benchmarks.

In practice, many autonomous vehicle developers put a tremendous amount of time and efforts in tuning and calibrating the path tracking controllers in order to achieve robust tracking performance and smooth steering actions over a wide range of vehicle speed and road curvature changes. This design process becomes tiresome when the target vehicle changes frequently. In this study, a model-based Linear Quadratic Gaussian (LQG) Control with adaptive Q-matrix is proposed to efficiently and systematically design the path tracking controller for any given target vehicle while effectively handling the noise and error problems arise from the localization and path planning algorithms. The regulator, in turn, is automatically designed, without additional efforts for tuning at various speeds. The performance of the proposed algorithm is validated based on KAIST autonomous vehicle. The experimental results show that the proposed LQG with adaptive Q-matrix has tracking performance in both low (15kph) and high (45kph) speed driving conditions better than those of other conventional tracking methods like the Stanley and Pure-pursuit methods.

Multiple object tracking (MOT) constitutes a critical research area within the field of computer vision. The creation of robust and efficient systems, which can approximate the mechanisms of human vision, is essential to enhance the efficacy of multiple object-tracking techniques. However, obstacles such as repetitive target appearances and frequent occlusions cause considerable inaccuracies or omissions in detection. Following the updating of these inaccurate observations into the tracklet, the effectiveness of the tracking model, employing appearance features, declines significantly. This paper introduces a novel method of multiple object tracking, employing graph attention networks and track management (GATM). Utilizing a graph attention network, an attention mechanism is employed to capture the relationships of nodes within the graph as well as node-to-node correlations across graphs. This mechanism allows selective focus on the features of advantageous nodes and enhances discriminability between node features, subsequently improving the performance and robustness of multiple object tracking. Simultaneously, we categorize distinct tracklet states and introduce an efficient track management method, which employs varying processing techniques for tracklets in diverse states. This method can manage occluded tracks in crowded scenes and improves tracking accuracy. Experiments conducted on three challenging public datasets (MOT16, MOT17, and MOT20) demonstrate that our method could deliver competitive performance.

Convolutional neural networks (CNN) based tracking approaches have shown favorable performance in recent benchmarks. Nonetheless, the chosen CNN features are always pre-trained in different task and individual components in tracking systems are learned separately, thus the achieved tracking performance may be suboptimal. Besides, most of these trackers are not designed towards realtime applications because of their time-consuming feature extraction and complex optimization details. In this paper, we propose an end-to-end framework to learn the convolutional features and perform the tracking process simultaneously, namely, a unified convolutional tracker (UCT). Specifically, The UCT treats feature extractor and tracking process (ridge regression) both as convolution operation and trains them jointly, enabling learned CNN features are tightly coupled to tracking process. In online tracking, an efficient updating method is proposed by introducing peak-versus-noise ratio (PNR) criterion, and scale changes are handled efficiently by incorporating a scale branch into network. The proposed approach results in superior tracking performance, while maintaining real-time speed. The standard UCT and UCT-Lite can track generic objects at 41 FPS and 154 FPS without further optimization, respectively. Experiments are performed on four challenging benchmark tracking datasets: OTB2013, OTB2015, VOT2014 and VOT2015, and our method achieves state-of-the-art results on these benchmarks compared with other real-time trackers.

Population dynamics in group-living species can be strongly affected both by features of sociality per se and by resultant population structure. To develop a mechanistic understanding of population dynamics in highly social species we need to investigate how processes within groups, processes linking groups, and external drivers act and interact to produce observed patterns. We model social group dynamics in cooperatively breeding meerkats, Suricata suricatta, paying attention to local demographic as well as dispersal processes. We use generalized additive models to describe the influence of group size, population density, and environmental conditions on demographic rates for each sex and stage, and we combine these models into predictive and individual-based simulation models of group dynamics. Short-term predictions of expected group size and simulated group trajectories over the longer term agree well with observations. Group dynamics are characterized by slow increases during the breeding season and relatively sharp declines during the pre-breeding season, particularly after dry years. We examine the demographic mechanisms responsible for environmental dependence. While individuals appear more prone to emigrate after dry years, seasons of low rainfall also cause reductions in reproductive output that produce adult-biased age distributions in the following dispersal season. Adult subordinates are much more likely to disperse or be evicted than immature individuals, and demographic structure thus contributes to crashes in group size. Our results demonstrate the role of social structure in characterizing a population's response to environmental variation. We discuss the implications of our findings for the population dynamics of cooperative breeders and population dynamics generally.

This study presents a comprehensive geospatial framework for assessing coastal vulnerability and ecosystem service distribution along the Greek coastline, one of the longest and most diverse in Europe. The framework integrates two complementary components: a Coastal Erosion Vulnerability Index applied to all identified beach units, and Coastal Flood Risk Indexes focused on low-lying and urbanized coastal segments. Both indices draw on harmonized, open-access European datasets to represent environmental, geomorphological, and socio-economic dimensions of risk. The Coastal Erosion Vulnerability Index is developed through a multi-criteria approach that combines indicators of physical erodibility, such as historical shoreline retreat, projected erosion under climate change, offshore wave power, and the cover of seagrass meadows, with socio-economic exposure metrics, including land use composition, population density, and beach-based recreational values. Inclusive accessibility for wheelchair users is also integrated to highlight equity-relevant aspects of coastal services. The Coastal Flood Risk Indexes identify flood-prone areas by simulating inundation through a novel point-based, computationally efficient geospatial method, which propagates water inland from coastal entry points using Extreme Sea Level (ESL) projections for future scenarios, overcoming the limitations of static ‘bathtub’ approaches. Together, the indices offer a spatially explicit, scalable framework to inform coastal zone management, climate adaptation planning, and the prioritization of nature-based solutions. By integrating vulnerability mapping with ecosystem service valuation, the framework supports evidence-based decision-making while aligning with key European policy goals for resilience and sustainable coastal development.

Seeing how great the contribution and potential of the aquaculture sector to the fulfillment of food for the people of Indonesia, the development of aquaculture including the Recirculating Aquaculture System (RAS) is very important to be carried out intensively. An important aspect that should be considered for the development of RAS is maintaining water quality as the main medium of cultivation. Optimization of the aquaculture tank is an effort that can be conducted to achieve this goal. A transient numerical study to determine the effect of inlet angle variation on flow conditions inside a circular aquaculture tank was conducted to assess the optimum geometry for the application. Single-phase CFD simulations using the k-ω SST turbulent model were carried out to evaluate the effect of inlet nozzle angle variations. Three angle variations 0°, 45°, and 90° were selected and the resultant fluid flow conditions were analyzed. From the CFD results, it can be known that the optimum condition for the circular aquaculture tank is achieved for the angle 0° among the tested angle in this present study.

Sedimentation efficiency evaluation of an aquaculture tank through experimental floc characterization and CFD simulation

Heavy metals may interact with ecological factors such as temperature, food level and salinity, causing both mortality and reduced reproduction in organisms. Among different heavy metals, copper compounds are commonly used for eliminating algal blooms in aquaculture tanks. At certain concentrations, copper is toxic to rotifers. In the present work, we evaluated the combined effects of salt concentrations (2.5 and 5.0 g l−1 NaCl), copper levels (0, 0.03125, 0.0625, 0.125 and 0.25 mg l−1 as CuCl2) and two temperatures (20 and 25 °C) on the population growth of B. rotundiformis using Chlorella as the algal food (at 0.5 × 106 cells ml−1 for every 24 h). Regardless of salinity and temperature, copper at concentrations as low as 0.03 mg l−1 had an adverse effect on the population growth of rotifers and above 0.125 mg l−1, the populations did not grow. The effect of the toxicant on B. rotundiformis was more severe at 25° than at 20 °C at lower salinity. In general, we observed peak densities of rotifers around day 12 at 20 °C but 6–8 days earlier at 25 °C. Peak population densities of B. rotundiformis in the controls at the salinity of 2.5 g l−1 ranged from 90 to 180 ind. ml−1, depending on temperature; at a salinity of 5.0 g l−1, these were lower. The population growth rates, r, in our study varied from +0.31 to −0.12 depending on the test conditions. There was a significant impact of temperature, salinity and toxicity level on the population growth rate of B. rotundiformis. Our results suggested that even narrow changes in salinity could negatively influence the toxicity of heavy metal on the population growth rates of B. rotundiformis.

A new framework for on-line object tracking based on SURF

Occlusion is one of the most difficult challenges in the area of visual tracking. We propose an occlusion handling framework to improve the performance of local tracking in a smart camera view in a multicamera network. We formulate an extensible energy function to quantify the quality of a camera’s observation of a particular target by taking into account both person–person and object–person occlusion. Using this energy function, a smart camera assesses the quality of observations over all targets being tracked. When it cannot adequately observe of a target, a smart camera estimates the quality of observation of the target from view points of other assisting cameras. If a camera with better observation of the target is found, the tracking task of the target is carried out with the assistance of that camera. In our framework, only positions of persons being tracked are exchanged between smart cameras. Thus, communication bandwidth requirement is very low. Performance evaluation of our method on challenging video sequences with frequent and severe occlusions shows that the accuracy of a baseline tracker is considerably improved. We also report the performance comparison to the state-of-the-art trackers in which our method outperforms.

This paper presents a real-time surveillance system for detecting and tracking people, which takes full advantage of local texture patterns, under a stationary monocular camera. A novel center-symmetric scale invariant local ternary pattern feature is put forward to combine with pattern kernel density estimation for building a pixel-level-based background model. The background model is then used to detect moving foreground objects on every newly captured frame. A variant of a fast human detector that utilizes local texture patterns is adopted to look for human objects from the foreground regions, and it is assisted by a head detector, which is proposed to find in advance the candidate locations of human, to reduce computational costs. Each human object is given a unique identity and is represented by a spatio-color-texture object model. The real-time performance of tracking is achieved by a fast mean-shift algorithm coupled with several efficient occlusion-handling techniques. Experiments on challenging video sequences show that the proposed surveillance system can run in real-time and is quite robust in segmenting and tracking people in complex environments that include appearance changes, abrupt motion, occlusions, illumination variations and clutter.

In a general Adaptive Robust Control (ARC) framework, the emphasis is always on the guaranteed transient performance and accurate trajectory tracking in the presence of uncertain nonlinearity and parametric uncertainties. However, when secondary purposes such as system health monitoring and prognosis are of equal importance, intelligent integration of output tracking performance oriented direct adaptive robust control (DARC) and the recently proposed accurate parameter estimation-based indirect adaptive robust control (IARC) is required. In this paper, we will consider such a seamless integration for a hydraulic robotic arm. The newly developed IARC design is first applied to the trajectory tracking for the robotic arm but with an improved estimation model, in which accurate parameter estimates are obtained through a parameter estimation algorithm that is based on physical dynamics rather than the tracking error dynamics. An integrated direct/indirect adaptive robust controller (DIARC) is then presented that preserves the excellent transient tracking performance of the direct ARC designs as well as the better parameter estimation process of the IARC design. The proposed Integrated Direct/Indirect Adaptive Robust Controller (DIARC) achieves the controller-identifier separation, thus enabling certain modularity in the controller design.

Product architecture represents components grouped into modules that can be assembled later to constitute a specific variant. Literature provides Methods of clustering components into weakly related modules with strong interconnections between components within modules. The number of modules and their hierarchical relationships shape product architecture and determine the balance between modular design and components integration. A novel hierarchical clustering approach, based on the biological Cladistics analysis, has been developed to cluster Design Structure Matrix (DSM) widely used to promote modularity. It evaluates different granularity levels of the resulting hierarchy and finds the best granularity level for maximum modularity. An automotive Body-in-White of 38 different components is used as a case study. Results showed the superiority of the recommended modularity pattern and synthesized product architecture over other clustering techniques.

Aiming at the problems of ID switching and tracking performance degradation caused by frequent occlusion and similar appearance of the tracked objects in dense scenes, a multi-object tracking method named TPFairMOT based on trajectory prediction and FairMOT is proposed in this paper. In the trajectory prediction branch, the object position of the future frame is predicted by using the object bounding box of the past frame and the velocity information learning network parameters, which overcomes the prediction failure caused by the uncertain motion state after the object is occluded in the tracking process. Secondly, the joint learning framework is used to combine the trajectory prediction branch with the detection and re-identification branch, and the tracking error caused by the high similarity between multiple objects in the tracking process is solved by integrating the appearance features and motion features of the tracked objects. Finally, MOTChallenge benchmarks (IDF1, IDs, MOTA, MT, and ML) are introduced to evaluate TPFairMOT, and different trajectory prediction strategies (FairMOT_KF and TPFairMOT_RNN) are used on FairMOT and TPFairMOT for comparative analysis. It is proved that the accuracy and ID switching times of trajectory prediction in this paper are better than other strategies. In addition, TPFairMOT, TPFairMOT_RNN, and FairMOT were compared on the public data sets MOT16, MOT17, and MOT20. The results show that TPFairMOT reduces the number of ID switching when the object is occluded, maintains the long-term validity of the identity information, and demonstrate good anti-occlusion performance.