The establishment of a sound science and technology ethics governance system is an inevitable requirement for national modernization.Faced with the development of human gene technology and the chaos in research activi...The establishment of a sound science and technology ethics governance system is an inevitable requirement for national modernization.Faced with the development of human gene technology and the chaos in research activities,the ethical standards and legal positioning of human gene research activities urgently need to be clarified.The human rights ethics view has value inclusiveness and value fundamentality,and includes three levels of connotations:content dimension,relationship dimension,and obligation dimension.It should serve as the ethical standard for human gene research activities.Based on the provisions of China’s Constitution,the human rights ethics view on human gene research,as a constitutional ethics view,can elucidate different levels of rights content,such as human dignity,life and health,and research freedom.It also addresses the weighing of basic rights conflicts and the dual obligation subjects of public and private nature.Relying on the constitutional value embedding of the research ethics view to form ethical consensus,improving ethical review through framework legislation for human rights interests,and implementing ethical responsibility through the human rights-oriented interpretation of ethical legal norms are the three pathways to realizing the human rights ethics view on human gene research.展开更多
In recent years,with the rapid development of software systems,the continuous expansion of software scale and the increasing complexity of systems have led to the emergence of a growing number of software metrics.Defe...In recent years,with the rapid development of software systems,the continuous expansion of software scale and the increasing complexity of systems have led to the emergence of a growing number of software metrics.Defect prediction methods based on software metric elements highly rely on software metric data.However,redundant software metric data is not conducive to efficient defect prediction,posing severe challenges to current software defect prediction tasks.To address these issues,this paper focuses on the rational clustering of software metric data.Firstly,multiple software projects are evaluated to determine the preset number of clusters for software metrics,and various clustering methods are employed to cluster the metric elements.Subsequently,a co-occurrence matrix is designed to comprehensively quantify the number of times that metrics appear in the same category.Based on the comprehensive results,the software metric data are divided into two semantic views containing different metrics,thereby analyzing the semantic information behind the software metrics.On this basis,this paper also conducts an in-depth analysis of the impact of different semantic view of metrics on defect prediction results,as well as the performance of various classification models under these semantic views.Experiments show that the joint use of the two semantic views can significantly improve the performance of models in software defect prediction,providing a new understanding and approach at the semantic view level for defect prediction research based on software metrics.展开更多
The increasing prevalence of multi-view data has made multi-view clustering a crucial technique for discovering latent structures from heterogeneous representations.However,traditional fuzzy clustering algorithms show...The increasing prevalence of multi-view data has made multi-view clustering a crucial technique for discovering latent structures from heterogeneous representations.However,traditional fuzzy clustering algorithms show limitations with the inherent uncertainty and imprecision of such data,as they rely on a single-dimensional membership value.To overcome these limitations,we propose an auto-weighted multi-view neutrosophic fuzzy clustering(AW-MVNFC)algorithm.Our method leverages the neutrosophic framework,an extension of fuzzy sets,to explicitly model imprecision and ambiguity through three membership degrees.The core novelty of AWMVNFC lies in a hierarchical weighting strategy that adaptively learns the contributions of both individual data views and the importance of each feature within a view.Through a unified objective function,AW-MVNFC jointly optimizes the neutrosophic membership assignments,cluster centers,and the distributions of view and feature weights.Comprehensive experiments conducted on synthetic and real-world datasets demonstrate that our algorithm achieves more accurate and stable clustering than existing methods,demonstrating its effectiveness in handling the complexities of multi-view data.展开更多
Trends Traveler Issue6,2025 The Young Traveler The"Grand Tour,"a form of long distance travel that allows young adults to gain insights and broeden their view of the world,began to emerge around the world du...Trends Traveler Issue6,2025 The Young Traveler The"Grand Tour,"a form of long distance travel that allows young adults to gain insights and broeden their view of the world,began to emerge around the world during the Renaissance in Burope and the Tang Dynasty in China.展开更多
Urban environments offer a wealth of opportunities for residents to respite from their hectic life.Outdoor running or jogging becomes increasingly popular of an option.Impacts of urban environments on outdoor running,...Urban environments offer a wealth of opportunities for residents to respite from their hectic life.Outdoor running or jogging becomes increasingly popular of an option.Impacts of urban environments on outdoor running,despite some initial studies,remain underexplored.This study aims to establish an analytical framework that can holistically assess the urban environment on the healthy vitality of running.The proposed framework is applied to two modern Chinese cities,i.e.,Guangzhou and Shenzhen.We construct three interpretable random forest models to explore the non-linear relationship between environmental variables and running intensity(RI)through analyzing the runners'trajectories and integrating with multi-source urban big data(e.g.,street view imagery,remote sensing,and socio-economic data)across the built,natural,and social dimensions,The findings uncover that road density has the greatest impact on RI,and social variables(e.g.,population density and housing price)and natural variables(e.g.,slope and humidity)all make notable impact on outdoor running.Despite these findings,the impact of environmental variables likely change across different regions due to disparate regional construction and micro-environments,and those specific impacts as well as optimal thresholds also alter.Therefore,construction of healthy cities should take the whole urban environment into account and adapt to local conditions.This study provides a comprehensive evaluation on the influencing variables of healthy vitality and guides sustainable urban planning for creating running-friendly cities.展开更多
Volumetric additive manufacturing(VAM) transforms traditional 2D light pattern projection into spatial light field energy superposition,maximizing the utilization of radiated light and allowing for ultra-fast,support-...Volumetric additive manufacturing(VAM) transforms traditional 2D light pattern projection into spatial light field energy superposition,maximizing the utilization of radiated light and allowing for ultra-fast,support-free printing,which has specific applications in fields such as life sciences and optics.However,traditional VAM processes require numerous projections and extensive computational preparation,limiting practical applications due to low projection efficiency and prolonged calculation times.In this study,we developed sparse-view irradiation processing VAM(SVIP-VAM),employing an optimized odd-even(OE) irradiation strategy inspired by sparse-view computed tomography.Theoretically,we demonstrated structural contour reconstruction feasibility with as few as 8 projections.Using this sparse-view approach,we achieved high-quality fabrication with only 15 projections,enhancing each projection efficiency by over 60 times and reducing projection set computational time by nearly 10-fold.Ultimately,this efficient sparse-view method significantly expands VAM applications into fields requiring rapid manufacturing,such as tissue engineering,medical implants,and aerospace manufacturing.展开更多
This paper is dedicated to constructing a theoretical framework for the identification and treatment of affective disorders in traditional Chinese medicine based on the“five-organ view”.Through in-depth analysis of ...This paper is dedicated to constructing a theoretical framework for the identification and treatment of affective disorders in traditional Chinese medicine based on the“five-organ view”.Through in-depth analysis of the theoretical connotation of the“five-organ concept”,we discussed the characteristics of the five-organ mechanism of affective-philosophical disorders in detail,systematically constructed a system of identification based on the association of the five organs,and proposed a comprehensive and holistic treatment strategy.The results of the study clearly show that the theoretical framework can provide systematic theoretical guidance for the clinical diagnosis and treatment of affective-philosophical disorders in Chinese medicine,help to improve the diagnostic and therapeutic effects of affective-philosophical disorders,and provide new ideas and methods for the theoretical development and clinical practice of affective-philosophical disorders in Chinese medicine,which is of important theoretical and practical significance,and can further promote the modernization of the development of affective-philosophical disorders in Chinese medicine.展开更多
In order to enhance the efficiency of visual inspection and effectively carry out 3D visual coverage tasks,this paper focuses on the 3D view planning problem concerning the visual coverage of an airplane's surface...In order to enhance the efficiency of visual inspection and effectively carry out 3D visual coverage tasks,this paper focuses on the 3D view planning problem concerning the visual coverage of an airplane's surface using unmanned aerial vehicles(UAv).Our objective is to attain a sufficiently high coverage rate with the least number of viewpoints.The contributions of this work are enumerated as follows.Firstly,the 3D model of the target aircraft is spatially extended in accordance with the depth range of the camera mounted on the drone,thereby confining the sampling range of 3D viewpoints.Next,a candidate set of viewpoints is generated through random sampling and the probabilistic potential field technique.Subsequently,we propose a novel hyper-heuristic algorithm.In this algorithm,a genetic algorithm serves as a high-level heuristic strategy,in tandem with multiple low-level heuristic operators devised for combinatorial optimization.This not only augments the capacity to seek the global optimal solution but also expedites the convergence rate,aiming to ascertain the optimal subset of viewpoints.Moreover,we devise a new fitness function for appraising candidate solution vectors in the set covering problem(ScP),strengthening the evolutionary guidance for genetic algorithms.Eventually,experimental findings on the simulated and real airplanes corroborate the efficacy of the proposed method,i.e.,it markedly diminishes the requisite number of viewpoints and augments inspection efficiency.展开更多
Human dignity is widely regarded as the foundation of modern human rights concepts and norms.The doctrine of human dignity in Chinese culture enjoys a long and profound history,and the pre-Qin assertion that“humans a...Human dignity is widely regarded as the foundation of modern human rights concepts and norms.The doctrine of human dignity in Chinese culture enjoys a long and profound history,and the pre-Qin assertion that“humans are the most precious”is the most representative expression of human dignity.Ancient Chinese scholars’elaboration on dignity was ethically oriented;they affirmed that humans have the freedom to make moral choices in spirit and required them to assume moral responsibilities towards others and society.Since modern times,with the changes of the times and the introduction of Western liberalism,the traditional view of moral dignity has seen a significant expansion of its scope,incorporating freedom in economic,political,and social life into the category of human dignity and establishing a closer connection with human rights.In contemporary China,under the guidance of Marxism,the view of dignity regards the free,comprehensive,and common development of human beings as the intrinsic requirement and external manifestation of human dignity,takes the rights to subsistence and development as the primary and fundamental human rights,and comprehensively safeguards the dignity of every individual through the coordinated protection of economic,political,social,and cultural rights.展开更多
The relationship between heaven and humanity is one of the fundamental philosophical foundations of ecological ethics in ancient Chinese Confucian thought.As a master of Confucian philosophy of mind,Wang Yangming inte...The relationship between heaven and humanity is one of the fundamental philosophical foundations of ecological ethics in ancient Chinese Confucian thought.As a master of Confucian philosophy of mind,Wang Yangming integrated the traditional Confucian discourse on the relationship between heaven and humanity into the principles of the philosophy of mind.Building on the traditional doctrine of benevolence centered on moral concern,he further developed an ecological view of‘benevolence as the unity of heaven and earth’.In his work Inquiry on the Great Learning,Wang Yangming systematically elaborated on this notion,emphasizing the philosophical expression of the relationship between humans and nature within an ethical framework and outlining the new implications of traditional Confucian ecological thought.This paper aims to analyze Wang Yangming’s ecological view of‘benevolence as the unity of heaven and earth’by examining the ecological ideas in his Inquiry on the Great Learning.On this basis,it seeks to refine the valuable achievements of traditional Chinese ecological civilization thought and strengthen the theoretical foundation of contemporary ecological ideas with Chinese characteristics.展开更多
Newspapers and magazines were primary media for Sino-Western cultural exchange in the early 19th century.The Chinese Repository,a monthly periodical founded by American missionary Elijah Bridgman in 1832,reported on C...Newspapers and magazines were primary media for Sino-Western cultural exchange in the early 19th century.The Chinese Repository,a monthly periodical founded by American missionary Elijah Bridgman in 1832,reported on Chinese social life as well as legal cases,system,and penalty.These reports served as a crucial window for the West to observe 19th-century Chinese criminal practices and significantly shaped Western perceptions of Chinese penal systems.Analysis of the periodical’s reports and reviews reveals a predominantly critical and negative Western view on China,arising from the collective Western impressions of China at the time and the identities and experiences of The Chinese Repository’s main contributors.The Western impressions of China formed during this period of time have had lasting negative impacts on Western legal demands towards China.Studies on the negative Western impressions of the Chinese view of legal penalty with a case study of The Chinese Repository offer valuable insights into the current Sino-Western legal cultural exchanges and dissemination.展开更多
Drone-based small object detection is of great significance in practical applications such as military actions, disaster rescue, transportation, etc. However, the severe scale differences in objects captured by drones...Drone-based small object detection is of great significance in practical applications such as military actions, disaster rescue, transportation, etc. However, the severe scale differences in objects captured by drones and lack of detail information for small-scale objects make drone-based small object detection a formidable challenge. To address these issues, we first develop a mathematical model to explore how changing receptive fields impacts the polynomial fitting results. Subsequently, based on the obtained conclusions, we propose a simple but effective Hybrid Receptive Field Network (HRFNet), whose modules include Hybrid Feature Augmentation (HFA), Hybrid Feature Pyramid (HFP) and Dual Scale Head (DSH). Specifically, HFA employs parallel dilated convolution kernels of different sizes to extend shallow features with different receptive fields, committed to improving the multi-scale adaptability of the network;HFP enhances the perception of small objects by capturing contextual information across layers, while DSH reconstructs the original prediction head utilizing a set of high-resolution features and ultrahigh-resolution features. In addition, in order to train HRFNet, the corresponding dual-scale loss function is designed. Finally, comprehensive evaluation results on public benchmarks such as VisDrone-DET and TinyPerson demonstrate the robustness of the proposed method. Most impressively, the proposed HRFNet achieves a mAP of 51.0 on VisDrone-DET with 29.3 M parameters, which outperforms the extant state-of-the-art detectors. HRFNet also performs excellently in complex scenarios captured by drones, achieving the best performance on the CS-Drone dataset we built.展开更多
基金This paper is an interim result of“Constitutional Boundaries of the Application of Human Gene Editing Technology,”a Youth Project of the National Social Science Fund of China(Project Approval Number 23CFX040)supported by the“National Funded Programs for Postdoctoral Researchers”(GZC20230937).
文摘The establishment of a sound science and technology ethics governance system is an inevitable requirement for national modernization.Faced with the development of human gene technology and the chaos in research activities,the ethical standards and legal positioning of human gene research activities urgently need to be clarified.The human rights ethics view has value inclusiveness and value fundamentality,and includes three levels of connotations:content dimension,relationship dimension,and obligation dimension.It should serve as the ethical standard for human gene research activities.Based on the provisions of China’s Constitution,the human rights ethics view on human gene research,as a constitutional ethics view,can elucidate different levels of rights content,such as human dignity,life and health,and research freedom.It also addresses the weighing of basic rights conflicts and the dual obligation subjects of public and private nature.Relying on the constitutional value embedding of the research ethics view to form ethical consensus,improving ethical review through framework legislation for human rights interests,and implementing ethical responsibility through the human rights-oriented interpretation of ethical legal norms are the three pathways to realizing the human rights ethics view on human gene research.
基金supported by the CCF-NSFOCUS‘Kunpeng’Research Fund(CCF-NSFOCUS2024012).
文摘In recent years,with the rapid development of software systems,the continuous expansion of software scale and the increasing complexity of systems have led to the emergence of a growing number of software metrics.Defect prediction methods based on software metric elements highly rely on software metric data.However,redundant software metric data is not conducive to efficient defect prediction,posing severe challenges to current software defect prediction tasks.To address these issues,this paper focuses on the rational clustering of software metric data.Firstly,multiple software projects are evaluated to determine the preset number of clusters for software metrics,and various clustering methods are employed to cluster the metric elements.Subsequently,a co-occurrence matrix is designed to comprehensively quantify the number of times that metrics appear in the same category.Based on the comprehensive results,the software metric data are divided into two semantic views containing different metrics,thereby analyzing the semantic information behind the software metrics.On this basis,this paper also conducts an in-depth analysis of the impact of different semantic view of metrics on defect prediction results,as well as the performance of various classification models under these semantic views.Experiments show that the joint use of the two semantic views can significantly improve the performance of models in software defect prediction,providing a new understanding and approach at the semantic view level for defect prediction research based on software metrics.
文摘The increasing prevalence of multi-view data has made multi-view clustering a crucial technique for discovering latent structures from heterogeneous representations.However,traditional fuzzy clustering algorithms show limitations with the inherent uncertainty and imprecision of such data,as they rely on a single-dimensional membership value.To overcome these limitations,we propose an auto-weighted multi-view neutrosophic fuzzy clustering(AW-MVNFC)algorithm.Our method leverages the neutrosophic framework,an extension of fuzzy sets,to explicitly model imprecision and ambiguity through three membership degrees.The core novelty of AWMVNFC lies in a hierarchical weighting strategy that adaptively learns the contributions of both individual data views and the importance of each feature within a view.Through a unified objective function,AW-MVNFC jointly optimizes the neutrosophic membership assignments,cluster centers,and the distributions of view and feature weights.Comprehensive experiments conducted on synthetic and real-world datasets demonstrate that our algorithm achieves more accurate and stable clustering than existing methods,demonstrating its effectiveness in handling the complexities of multi-view data.
文摘Trends Traveler Issue6,2025 The Young Traveler The"Grand Tour,"a form of long distance travel that allows young adults to gain insights and broeden their view of the world,began to emerge around the world during the Renaissance in Burope and the Tang Dynasty in China.
基金National Natural Science Foundation of China,No.42171455The Hong Kong RGC Research Impact Fund,No.R5011-23The Hong Kong General Research Fund,No.15204121。
文摘Urban environments offer a wealth of opportunities for residents to respite from their hectic life.Outdoor running or jogging becomes increasingly popular of an option.Impacts of urban environments on outdoor running,despite some initial studies,remain underexplored.This study aims to establish an analytical framework that can holistically assess the urban environment on the healthy vitality of running.The proposed framework is applied to two modern Chinese cities,i.e.,Guangzhou and Shenzhen.We construct three interpretable random forest models to explore the non-linear relationship between environmental variables and running intensity(RI)through analyzing the runners'trajectories and integrating with multi-source urban big data(e.g.,street view imagery,remote sensing,and socio-economic data)across the built,natural,and social dimensions,The findings uncover that road density has the greatest impact on RI,and social variables(e.g.,population density and housing price)and natural variables(e.g.,slope and humidity)all make notable impact on outdoor running.Despite these findings,the impact of environmental variables likely change across different regions due to disparate regional construction and micro-environments,and those specific impacts as well as optimal thresholds also alter.Therefore,construction of healthy cities should take the whole urban environment into account and adapt to local conditions.This study provides a comprehensive evaluation on the influencing variables of healthy vitality and guides sustainable urban planning for creating running-friendly cities.
基金supported financially by the Beijing Municipal Natural Science Foundation (L232109)National Natural Science Foundation of China (22073003)Fundamental Research Funds for the Central Universities (YWF-22-K-101)。
文摘Volumetric additive manufacturing(VAM) transforms traditional 2D light pattern projection into spatial light field energy superposition,maximizing the utilization of radiated light and allowing for ultra-fast,support-free printing,which has specific applications in fields such as life sciences and optics.However,traditional VAM processes require numerous projections and extensive computational preparation,limiting practical applications due to low projection efficiency and prolonged calculation times.In this study,we developed sparse-view irradiation processing VAM(SVIP-VAM),employing an optimized odd-even(OE) irradiation strategy inspired by sparse-view computed tomography.Theoretically,we demonstrated structural contour reconstruction feasibility with as few as 8 projections.Using this sparse-view approach,we achieved high-quality fabrication with only 15 projections,enhancing each projection efficiency by over 60 times and reducing projection set computational time by nearly 10-fold.Ultimately,this efficient sparse-view method significantly expands VAM applications into fields requiring rapid manufacturing,such as tissue engineering,medical implants,and aerospace manufacturing.
文摘This paper is dedicated to constructing a theoretical framework for the identification and treatment of affective disorders in traditional Chinese medicine based on the“five-organ view”.Through in-depth analysis of the theoretical connotation of the“five-organ concept”,we discussed the characteristics of the five-organ mechanism of affective-philosophical disorders in detail,systematically constructed a system of identification based on the association of the five organs,and proposed a comprehensive and holistic treatment strategy.The results of the study clearly show that the theoretical framework can provide systematic theoretical guidance for the clinical diagnosis and treatment of affective-philosophical disorders in Chinese medicine,help to improve the diagnostic and therapeutic effects of affective-philosophical disorders,and provide new ideas and methods for the theoretical development and clinical practice of affective-philosophical disorders in Chinese medicine,which is of important theoretical and practical significance,and can further promote the modernization of the development of affective-philosophical disorders in Chinese medicine.
基金supported by the Trade Union Employee Innovation Foundation of China(2022270024).
文摘In order to enhance the efficiency of visual inspection and effectively carry out 3D visual coverage tasks,this paper focuses on the 3D view planning problem concerning the visual coverage of an airplane's surface using unmanned aerial vehicles(UAv).Our objective is to attain a sufficiently high coverage rate with the least number of viewpoints.The contributions of this work are enumerated as follows.Firstly,the 3D model of the target aircraft is spatially extended in accordance with the depth range of the camera mounted on the drone,thereby confining the sampling range of 3D viewpoints.Next,a candidate set of viewpoints is generated through random sampling and the probabilistic potential field technique.Subsequently,we propose a novel hyper-heuristic algorithm.In this algorithm,a genetic algorithm serves as a high-level heuristic strategy,in tandem with multiple low-level heuristic operators devised for combinatorial optimization.This not only augments the capacity to seek the global optimal solution but also expedites the convergence rate,aiming to ascertain the optimal subset of viewpoints.Moreover,we devise a new fitness function for appraising candidate solution vectors in the set covering problem(ScP),strengthening the evolutionary guidance for genetic algorithms.Eventually,experimental findings on the simulated and real airplanes corroborate the efficacy of the proposed method,i.e.,it markedly diminishes the requisite number of viewpoints and augments inspection efficiency.
基金“Research on the Content and Realization Methods of Citizens’Participation Rights,”a major project(Project Number 21JJD820003)funded by the National Human Rights Education and Training Base of the Ministry of Education of China.
文摘Human dignity is widely regarded as the foundation of modern human rights concepts and norms.The doctrine of human dignity in Chinese culture enjoys a long and profound history,and the pre-Qin assertion that“humans are the most precious”is the most representative expression of human dignity.Ancient Chinese scholars’elaboration on dignity was ethically oriented;they affirmed that humans have the freedom to make moral choices in spirit and required them to assume moral responsibilities towards others and society.Since modern times,with the changes of the times and the introduction of Western liberalism,the traditional view of moral dignity has seen a significant expansion of its scope,incorporating freedom in economic,political,and social life into the category of human dignity and establishing a closer connection with human rights.In contemporary China,under the guidance of Marxism,the view of dignity regards the free,comprehensive,and common development of human beings as the intrinsic requirement and external manifestation of human dignity,takes the rights to subsistence and development as the primary and fundamental human rights,and comprehensively safeguards the dignity of every individual through the coordinated protection of economic,political,social,and cultural rights.
文摘The relationship between heaven and humanity is one of the fundamental philosophical foundations of ecological ethics in ancient Chinese Confucian thought.As a master of Confucian philosophy of mind,Wang Yangming integrated the traditional Confucian discourse on the relationship between heaven and humanity into the principles of the philosophy of mind.Building on the traditional doctrine of benevolence centered on moral concern,he further developed an ecological view of‘benevolence as the unity of heaven and earth’.In his work Inquiry on the Great Learning,Wang Yangming systematically elaborated on this notion,emphasizing the philosophical expression of the relationship between humans and nature within an ethical framework and outlining the new implications of traditional Confucian ecological thought.This paper aims to analyze Wang Yangming’s ecological view of‘benevolence as the unity of heaven and earth’by examining the ecological ideas in his Inquiry on the Great Learning.On this basis,it seeks to refine the valuable achievements of traditional Chinese ecological civilization thought and strengthen the theoretical foundation of contemporary ecological ideas with Chinese characteristics.
基金Supported by the National Social Science Fund General Project titled“The Millard’s Review and the Study of Sino-American Legal Civilization Exchange in the First Half of the 20th Century”.
文摘Newspapers and magazines were primary media for Sino-Western cultural exchange in the early 19th century.The Chinese Repository,a monthly periodical founded by American missionary Elijah Bridgman in 1832,reported on Chinese social life as well as legal cases,system,and penalty.These reports served as a crucial window for the West to observe 19th-century Chinese criminal practices and significantly shaped Western perceptions of Chinese penal systems.Analysis of the periodical’s reports and reviews reveals a predominantly critical and negative Western view on China,arising from the collective Western impressions of China at the time and the identities and experiences of The Chinese Repository’s main contributors.The Western impressions of China formed during this period of time have had lasting negative impacts on Western legal demands towards China.Studies on the negative Western impressions of the Chinese view of legal penalty with a case study of The Chinese Repository offer valuable insights into the current Sino-Western legal cultural exchanges and dissemination.
基金supported by the National Natural Science Foundation of China(Nos.62276204 and 62203343)the Fundamental Research Funds for the Central Universities(No.YJSJ24011)+1 种基金the Natural Science Basic Research Program of Shanxi,China(Nos.2022JM-340 and 2023-JC-QN-0710)the China Postdoctoral Science Foundation(Nos.2020T130494 and 2018M633470).
文摘Drone-based small object detection is of great significance in practical applications such as military actions, disaster rescue, transportation, etc. However, the severe scale differences in objects captured by drones and lack of detail information for small-scale objects make drone-based small object detection a formidable challenge. To address these issues, we first develop a mathematical model to explore how changing receptive fields impacts the polynomial fitting results. Subsequently, based on the obtained conclusions, we propose a simple but effective Hybrid Receptive Field Network (HRFNet), whose modules include Hybrid Feature Augmentation (HFA), Hybrid Feature Pyramid (HFP) and Dual Scale Head (DSH). Specifically, HFA employs parallel dilated convolution kernels of different sizes to extend shallow features with different receptive fields, committed to improving the multi-scale adaptability of the network;HFP enhances the perception of small objects by capturing contextual information across layers, while DSH reconstructs the original prediction head utilizing a set of high-resolution features and ultrahigh-resolution features. In addition, in order to train HRFNet, the corresponding dual-scale loss function is designed. Finally, comprehensive evaluation results on public benchmarks such as VisDrone-DET and TinyPerson demonstrate the robustness of the proposed method. Most impressively, the proposed HRFNet achieves a mAP of 51.0 on VisDrone-DET with 29.3 M parameters, which outperforms the extant state-of-the-art detectors. HRFNet also performs excellently in complex scenarios captured by drones, achieving the best performance on the CS-Drone dataset we built.
