Person re-identification(RE-ID) has played a significant role in the fields of image processing and computer vision because of its potential value in practical applications. Researchers are striving to design new algo...
详细信息
ISBN:
(纸本)9781479970612
Person re-identification(RE-ID) has played a significant role in the fields of image processing and computer vision because of its potential value in practical applications. Researchers are striving to design new algorithms to improve the performance of RE-ID but ignore the advantages of existing approaches. In this paper, motivated by deep reinforcement learning, we propose a Deep Agent which can integrate existing algorithms and enable them to complement each other. Two Deep Agents are designed to integrate algorithms for data augmentation and feature extraction parts separately for RE-ID. Experiment results demonstrate that the integrated algorithms can achieve a better accuracy than using each one of them alone.
There are challenges for cloud detection over bright surfaces, especially for Sentinel-2 image without thermal band. In this paper, we developed a new cloud detection algorithm for Sentinel-2 data designed to tackle t...
详细信息
There are challenges for cloud detection over bright surfaces, especially for Sentinel-2 image without thermal band. In this paper, we developed a new cloud detection algorithm for Sentinel-2 data designed to tackle the challenges of cloud detection on bright targets. Our algorithm (called RF-NDTeI-SNIC) was built through the integration of Random Forest (RF) model, Normalized Difference Temporal Index (NDTeI) cloud index and Simple Non-Iterative Clustering (SNIC) image segmentation algorithm. First, RF and NDTeI were used to derive the initial and second stage's cloud detection results, respectively. Then, the RF- and NDTeI-based results were fused to remove the bright surfaces and land cover changes. Finally, the fusion results were further refined using SNIC for morphological processing. The proposed RF-NDTeI-SNIC algorithm was evaluated using nine challenging cloud-covered Sentinel-2 scenes with substantial bright surfaces. Results indicated that the average overall accuracy of our algorithm was 95.01%, with average commission rate of 10.19% for cloud. In addition, five commonly used Sentinel-2 cloud mask algorithms including s2cloudless, QA, Fmask, CDI and Tmask were selected for comparative analysis. Results suggested that our algorithm outperformed others with overall accuracies of 89.60-94.17% and cloud commission rate of 9.46-37.50%, and have significant advantages in terms of bright surfaces. In summary, the RF-NDTeI-SNIC algorithm we developed was capable of yielding accurate Sentinel-2 cloud masks, and the novel NDTeI cloud index we proposed gave a new and effective approach for improving the separability of clouds and bright surfaces.
Addressing the limitations of the ant colony algorithm in global path planning, such as its blind search tendencies, susceptibility to local optima, and slow convergence, as well as the Dynamic Window Approach (DWA) a...
详细信息
Addressing the limitations of the ant colony algorithm in global path planning, such as its blind search tendencies, susceptibility to local optima, and slow convergence, as well as the Dynamic Window Approach (DWA) algorithm's inadequacies in effectively avoiding dynamic obstacles in local path planning, this study proposes a hybrid strategy that integrates an improved ant colony algorithm with the DWA algorithm. This strategy enhances the heuristic function of the ant colony algorithm by combining the actual distance from the current node to adjacent nodes with the relative distance from adjacent nodes to the target point, making the heuristic information more aligned with the real-world search environment. The pheromone update strategy integrates local and global optimal path information with meticulous control over pheromone evaporation, balancing exploration and exploitation and preventing premature convergence of the algorithm. Furthermore, the DWA algorithm's local path planning capabilities are strengthened by incorporating target-oriented evaluation functions, speed evaluation functions, and obstacle avoidance evaluation functions, enabling the robot to more flexibly respond to changes in its environment. This study implemented the algorithm optimization through the following steps: initializing algorithm parameters and the environment map;employing the improved Ant Colony Optimization (ACO) for global path searching;utilizing DWA for local path decisions and obstacle avoidance;and continuously optimizing global and local avoidance strategies through iterative updates and feedback mechanisms. A series of simulation experiments demonstrate that the proposed integrated strategy significantly enhances path planning efficiency, reduces path length, and improves obstacle avoidance capabilities, providing an effective solution for path planning issues in complex environments.
Complex network models are frequently employed for simulating and studyingdiverse real-world complex *** these models,scale-free networks typically exhibit greater fragility to malicious ***,enhancing the robustness o...
详细信息
Complex network models are frequently employed for simulating and studyingdiverse real-world complex *** these models,scale-free networks typically exhibit greater fragility to malicious ***,enhancing the robustness of scale-free networks has become a pressing *** address this problem,this paper proposes a Multi-Granularity integrationalgorithm(MGIA),which aims to improve the robustness of scale-free networks while keeping the initial degree of each node unchanged,ensuring network connectivity and avoiding the generation of multiple *** algorithm generates a multi-granularity structure from the initial network to be optimized,then uses different optimization strategies to optimize the networks at various granular layers in this structure,and finally realizes the information exchange between different granular layers,thereby further enhancing the optimization *** propose new network refresh,crossover,and mutation operators to ensure that the optimized network satisfies the given ***,we propose new network similarity and network dissimilarity evaluation metrics to improve the effectiveness of the optimization operators in the *** the experiments,the MGIA enhances the robustness of the scale-free network by 67.6%.This improvement is approximately 17.2%higher than the optimization effects achieved by eight currently existing complex network robustness optimization algorithms.
The insufficient accuracy of GPS technology leads to the sampling trajectory data being away from the actual road. In order to improve the accuracy of GPS trajectory data for matching to map, a map matching integratio...
详细信息
ISBN:
(纸本)9781728163604
The insufficient accuracy of GPS technology leads to the sampling trajectory data being away from the actual road. In order to improve the accuracy of GPS trajectory data for matching to map, a map matching integrationalgorithm based on historical trajectory data is proposed. Firstly projection distance and hidden Markov model are used respectively to compare the matching results. Then the difference road segments are found and the DBSCAN algorithm is used to cluster the historical trajectory data to adjust the segments. Our experiment uses the truck trajectory data to test the algorithm. The results show that the map matching integrationalgorithm effectively improves the accuracy of map matching.
Protein interactions and cellular responses are fundamental pillars of molecular systems biology. Decoding these complex signaling pathways requires advanced computational methods. One promising direction of algorithm...
详细信息
ISBN:
(纸本)9798400704246
Protein interactions and cellular responses are fundamental pillars of molecular systems biology. Decoding these complex signaling pathways requires advanced computational methods. One promising direction of algorithm development is using graph algorithms to identify proteins involved in signaling pathways. Despite the availability of tools, many researchers grapple with software and user experience constraints. In response, we have developed the Signaling Pathway Reconstruction Analysis Streamliner (SPRAS), a robust containerized framework that enables users to easily reconstruct signaling pathways by connecting proteins of interest within molecular interaction networks. It seamlessly integrates graph algorithms designed for pathway reconstruction with downstream visualization and clustering analysis. We contribute and integrate three random-walk-based algorithms to SPRAS, including one algorithm we developed for large networks and two other algorithms that appear in the literature. Random walk approaches have been highly successful in predicting candidate proteins involved in a signaling pathway, and integrating them into SPRAS will greatly expand the framework's ability for pathway reconstruction. We illustrate their importance by using the random walk algorithms now available in SPRAS to explore potential proteins involved in cell-cell fusion in flies. In our computational experiments, five fly proteins appeared in multiple reconstructed pathways, suggesting a potential role for them in cell-cell fusion. With the addition of these new algorithms, SPRAS will become an essential tool for unraveling the mysteries of biological interactions.
Industrial Internet of Things (IIoT) applications are being used more and more frequently. Data collected by various sensors can be used to provide innovative digital services supporting increasing efficiency or cost ...
详细信息
Industrial Internet of Things (IIoT) applications are being used more and more frequently. Data collected by various sensors can be used to provide innovative digital services supporting increasing efficiency or cost reduction. The implementation of such applications requires the integration and analysis of heterogeneous data coming from a broad variety of sensors. To support these steps, this paper introduces OPAL, a software toolbox consolidating several software components for the semantically annotated integration and analysis of IoT-data. Data storage is realized in a standardized and INSPIRE-compliant way utilizing the SensorThings API. Supporting a broad variety of use cases, OPAL provides several import adapters to access data sources with various protocols (e.g., the OPC UA protocol, which is often used in industrial environments). In addition, a unified management and execution environment, called PERMA, is introduced to allow the programming language independent integration of algorithms.
The insufficient accuracy of GPS technology leads to the sampling trajectory data being away from the actual road. In order to improve the accuracy of GPS trajectory data for matching to map, a map matching integratio...
详细信息
ISBN:
(纸本)9781728163611
The insufficient accuracy of GPS technology leads to the sampling trajectory data being away from the actual road. In order to improve the accuracy of GPS trajectory data for matching to map, a map matching integrationalgorithm based on historical trajectory data is proposed. Firstly projection distance and hidden Markov model are used respectively to compare the matching results. Then the difference road segments are found and the DBSCAN algorithm is used to cluster the historical trajectory data to adjust the segments. Our experiment uses the truck trajectory data to test the algorithm. The results show that the map matching integrationalgorithm effectively improves the accuracy of map matching.
A parallel scheme based on Probabilistic Tensor Factorization which addresses the scalability problem of Collaborative Filtering (CF) is proposed for big data processing. Parallel algorithms for large scale recommenda...
详细信息
ISBN:
(纸本)9781450364195
A parallel scheme based on Probabilistic Tensor Factorization which addresses the scalability problem of Collaborative Filtering (CF) is proposed for big data processing. Parallel algorithms for large scale recommendation problems have witnessed advancements in the big data era in recent times. Matrix Factorization models have been enormously used to tackle such constraints, which we see as not scalable and does not converge easily unless numerous iterations making it computationally expensive. This study proposes a novel coordinate descent based probabilistic Tensor factorization method;Scalable Probabilistic Time Context Tensor Factorization (SPTTF) for collaborative recommendation. Our experiments with natural datasets show its efficiency.
The Ozone Mapping and Profiler Suite (OMPS) consisting of three instruments, a Nadir Mapper (OMPS-NM), a Nadir Profiler (OMPS-NP) and a limb Profiler (OMPS-LP) was launched aboard Suomi NPP satellite in 2011. The inst...
详细信息
ISBN:
(纸本)9781479979295
The Ozone Mapping and Profiler Suite (OMPS) consisting of three instruments, a Nadir Mapper (OMPS-NM), a Nadir Profiler (OMPS-NP) and a limb Profiler (OMPS-LP) was launched aboard Suomi NPP satellite in 2011. The instrument is also planned to be aboard JPSS satellite. The data from this instrument is processed at Interface Data Processing Segment (IDPS) built by Raytheon. The main products from OMPS include Total Column Ozone and Ozone Profile. algorithm Development Library (ADL) framework mimics IDPS system and is used to test, troubleshoot and integrate algorithm updates. The algorithm integration team (AIT) at NOAA STAR is tasked to integrate the algorithm updates in ADL that could be simply plugged in the operational system. In this paper we will present the process in detail and will discuss the results of the latest update of V8 algorithm from V6 algorithm for OMPS Nadir Profile instrument.
暂无评论