Accurate detection of small objects is critically important in high-stakes applications such as military reconnaissance and emergency rescue.However,low resolution,occlusion,and background interference make small obje...Accurate detection of small objects is critically important in high-stakes applications such as military reconnaissance and emergency rescue.However,low resolution,occlusion,and background interference make small object detection a complex and demanding task.One effective approach to overcome these issues is the integration of multimodal image data to enhance detection capabilities.This paper proposes a novel small object detection method that utilizes three types of multimodal image combinations,such as Hyperspectral-Multispectral(HSMS),Hyperspectral-Synthetic Aperture Radar(HS-SAR),and HS-SAR-Digital Surface Model(HS-SAR-DSM).The detection process is done by the proposed Jaccard Deep Q-Net(JDQN),which integrates the Jaccard similarity measure with a Deep Q-Network(DQN)using regression modeling.To produce the final output,a Deep Maxout Network(DMN)is employed to fuse the detection results obtained from each modality.The effectiveness of the proposed JDQN is validated using performance metrics,such as accuracy,Mean Squared Error(MSE),precision,and Root Mean Squared Error(RMSE).Experimental results demonstrate that the proposed JDQN method outperforms existing approaches,achieving the highest accuracy of 0.907,a precision of 0.904,the lowest normalized MSE of 0.279,and a normalized RMSE of 0.528.展开更多
随着个性化推荐系统在各类应用中的广泛应用,电影推荐作为其中的典型场景,受到了越来越多的关注。推荐系统的核心任务是根据用户的历史行为数据,特别是评分数据,来为用户提供个性化的推荐。推荐效果的好坏与相似度计算方法的选择密切相...随着个性化推荐系统在各类应用中的广泛应用,电影推荐作为其中的典型场景,受到了越来越多的关注。推荐系统的核心任务是根据用户的历史行为数据,特别是评分数据,来为用户提供个性化的推荐。推荐效果的好坏与相似度计算方法的选择密切相关。常见的相似度计算方法包括余弦相似度、皮尔逊相似度、欧几里得距离和Jaccard相似度。每种方法有其独特的特点和适用场景,但单一使用某种相似度方法往往会受到数据特性和环境的限制,导致推荐性能的下降。本文通过系统地比较和分析这四种相似度计算方法在不同环境下的表现,探讨了它们在电影推荐中的应用效果。研究表明,在不同数据场景下(如稀疏数据、新用户、活跃用户等),合理组合不同相似度计算方法的比例,能够克服单一方法的局限性,显著提高推荐系统的准确性和质量。通过实验验证,我们发现基于加权组合的多维度推荐方法,相较于单一相似度方法,能够在不同推荐场景中提升推荐系统的综合表现。With the widespread adoption of personalized recommendation systems in various applications, movie recommendation as a typical scenario has attracted increasing attention. The core task of recommendation systems lies in providing personalized suggestions based on users’ historical behavioral data, particularly rating data. The effectiveness of recommendations is closely tied to the selection of similarity computation methods. Common approaches include cosine similarity, Pearson correlation, Euclidean distance, and Jaccard similarity. While each method has unique characteristics and applicable scenarios, relying solely on a single similarity measure often leads to performance degradation due to data-specific limitations and environmental constraints. This study systematically compares and analyzes the performance of these four similarity computation methods under different environmental conditions, exploring their application effectiveness in movie recommendations. The research demonstrates that rationally combining multiple similarity measures with weighted proportions can overcome the limitations of individual methods and significantly enhance recommendation accuracy and quality across diverse data scenarios (e.g., sparse data, new users, and active users). Experimental results verify that the proposed multi-dimensional recommendation method based on weighted combinations outperforms single similarity approaches in improving comprehensive system performance across various recommendation scenarios.展开更多
文摘Accurate detection of small objects is critically important in high-stakes applications such as military reconnaissance and emergency rescue.However,low resolution,occlusion,and background interference make small object detection a complex and demanding task.One effective approach to overcome these issues is the integration of multimodal image data to enhance detection capabilities.This paper proposes a novel small object detection method that utilizes three types of multimodal image combinations,such as Hyperspectral-Multispectral(HSMS),Hyperspectral-Synthetic Aperture Radar(HS-SAR),and HS-SAR-Digital Surface Model(HS-SAR-DSM).The detection process is done by the proposed Jaccard Deep Q-Net(JDQN),which integrates the Jaccard similarity measure with a Deep Q-Network(DQN)using regression modeling.To produce the final output,a Deep Maxout Network(DMN)is employed to fuse the detection results obtained from each modality.The effectiveness of the proposed JDQN is validated using performance metrics,such as accuracy,Mean Squared Error(MSE),precision,and Root Mean Squared Error(RMSE).Experimental results demonstrate that the proposed JDQN method outperforms existing approaches,achieving the highest accuracy of 0.907,a precision of 0.904,the lowest normalized MSE of 0.279,and a normalized RMSE of 0.528.
文摘随着个性化推荐系统在各类应用中的广泛应用,电影推荐作为其中的典型场景,受到了越来越多的关注。推荐系统的核心任务是根据用户的历史行为数据,特别是评分数据,来为用户提供个性化的推荐。推荐效果的好坏与相似度计算方法的选择密切相关。常见的相似度计算方法包括余弦相似度、皮尔逊相似度、欧几里得距离和Jaccard相似度。每种方法有其独特的特点和适用场景,但单一使用某种相似度方法往往会受到数据特性和环境的限制,导致推荐性能的下降。本文通过系统地比较和分析这四种相似度计算方法在不同环境下的表现,探讨了它们在电影推荐中的应用效果。研究表明,在不同数据场景下(如稀疏数据、新用户、活跃用户等),合理组合不同相似度计算方法的比例,能够克服单一方法的局限性,显著提高推荐系统的准确性和质量。通过实验验证,我们发现基于加权组合的多维度推荐方法,相较于单一相似度方法,能够在不同推荐场景中提升推荐系统的综合表现。With the widespread adoption of personalized recommendation systems in various applications, movie recommendation as a typical scenario has attracted increasing attention. The core task of recommendation systems lies in providing personalized suggestions based on users’ historical behavioral data, particularly rating data. The effectiveness of recommendations is closely tied to the selection of similarity computation methods. Common approaches include cosine similarity, Pearson correlation, Euclidean distance, and Jaccard similarity. While each method has unique characteristics and applicable scenarios, relying solely on a single similarity measure often leads to performance degradation due to data-specific limitations and environmental constraints. This study systematically compares and analyzes the performance of these four similarity computation methods under different environmental conditions, exploring their application effectiveness in movie recommendations. The research demonstrates that rationally combining multiple similarity measures with weighted proportions can overcome the limitations of individual methods and significantly enhance recommendation accuracy and quality across diverse data scenarios (e.g., sparse data, new users, and active users). Experimental results verify that the proposed multi-dimensional recommendation method based on weighted combinations outperforms single similarity approaches in improving comprehensive system performance across various recommendation scenarios.
