With the explosive growth of false information on social media platforms, the automatic detection of multimodalfalse information has received increasing attention. Recent research has significantly contributed to mult...With the explosive growth of false information on social media platforms, the automatic detection of multimodalfalse information has received increasing attention. Recent research has significantly contributed to multimodalinformation exchange and fusion, with many methods attempting to integrate unimodal features to generatemultimodal news representations. However, they still need to fully explore the hierarchical and complex semanticcorrelations between different modal contents, severely limiting their performance detecting multimodal falseinformation. This work proposes a two-stage detection framework for multimodal false information detection,called ASMFD, which is based on image aesthetic similarity to segment and explores the consistency andinconsistency features of images and texts. Specifically, we first use the Contrastive Language-Image Pre-training(CLIP) model to learn the relationship between text and images through label awareness and train an imageaesthetic attribute scorer using an aesthetic attribute dataset. Then, we calculate the aesthetic similarity betweenthe image and related images and use this similarity as a threshold to divide the multimodal correlation matrixinto consistency and inconsistencymatrices. Finally, the fusionmodule is designed to identify essential features fordetectingmultimodal false information. In extensive experiments on four datasets, the performance of the ASMFDis superior to state-of-the-art baseline methods.展开更多
With the rapid development of Unmanned Aerial Vehicle(UAV)technology,change detection methods based on UAV images have been extensively studied.However,the imaging of UAV sensors is susceptible to environmental interf...With the rapid development of Unmanned Aerial Vehicle(UAV)technology,change detection methods based on UAV images have been extensively studied.However,the imaging of UAV sensors is susceptible to environmental interference,which leads to great differences of same object between UAV images.Overcoming the discrepancy difference between UAV images is crucial to improving the accuracy of change detection.To address this issue,a novel unsupervised change detection method based on structural consistency and the Generalized Fuzzy Local Information C-means Clustering Model(GFLICM)was proposed in this study.Within this method,the establishment of a graph-based structural consistency measure allowed for the detection of change information by comparing structure similarity between UAV images.The local variation coefficient was introduced and a new fuzzy factor was reconstructed,after which the GFLICM algorithm was used to analyze difference images.Finally,change detection results were analyzed qualitatively and quantitatively.To measure the feasibility and robustness of the proposed method,experiments were conducted using two data sets from the cities of Yangzhou and Nanjing.The experimental results show that the proposed method can improve the overall accuracy of change detection and reduce the false alarm rate when compared with other state-of-the-art change detection methods.展开更多
Objective: The objective of the study is to verify the clinical validity of the following kits with the comparative experimental analysis and evaluate whether their performance can meet the clinical requirements, i.e....Objective: The objective of the study is to verify the clinical validity of the following kits with the comparative experimental analysis and evaluate whether their performance can meet the clinical requirements, i.e. Class III in vitro diagnostic reagent “Herpes Simplex Virus (HSV) Type II Nucleic Acid Detection Kit (PCR-Fluorescence Probe Method)” of Daan Gene Co., Ltd. (Daan kit for short) and “Herpes Simplex Virus (HSV) Type II Nucleic Acid Detection Kit (Fluorescence PCR Method)” of Wuhan Biot Gene Co., Ltd. (Biot kit for short). Method: In the study process, the samples were divided into positive and negative groups according to the control test results, and the clinical application performance of Daan kit and Biot kit was evaluated by comparing their test results. Results: The results show that two kits indicate the same test results, i.e. 26 positive and 107 negative samples in a total of 133 male urethral discharge samples, and 32 positive and 238 negative samples in a total of 270 female cervical secretion samples. Conclusion: It can be concluded from the clinical test that Daan and Biot Herpes Simplex Virus (HSV) Type II Nuc- leic Acid Test Kits are reliable, accurate, safe, convenient for use, stable and high-value in the clinical application.展开更多
Recently,a new research trend in our video salient object detection(VSOD)research community has focused on enhancing the detection results via model self-fine-tuning using sparsely mined high-quality keyframes from th...Recently,a new research trend in our video salient object detection(VSOD)research community has focused on enhancing the detection results via model self-fine-tuning using sparsely mined high-quality keyframes from the given sequence.Although such a learning scheme is generally effective,it has a critical limitation,i.e.,the model learned on sparse frames only possesses weak generalization ability.This situation could become worse on“long”videos since they tend to have intensive scene variations.Moreover,in such videos,the keyframe information from a longer time span is less relevant to the previous,which could also cause learning conflict and deteriorate the model performance.Thus,the learning scheme is usually incapable of handling complex pattern modeling.To solve this problem,we propose a divide-and-conquer framework,which can convert a complex problem domain into multiple simple ones.First,we devise a novel background consistency analysis(BCA)which effectively divides the mined frames into disjoint groups.Then for each group,we assign an individual deep model on it to capture its key attribute during the fine-tuning phase.During the testing phase,we design a model-matching strategy,which could dynamically select the best-matched model from those fine-tuned ones to handle the given testing frame.Comprehensive experiments show that our method can adapt severe background appearance variation coupling with object movement and obtain robust saliency detection compared with the previous scheme and the state-of-the-art methods.展开更多
针对同步定位与地图建立(simultaneous localization and mapping,SLAM)算法在动态环境下存在位姿估计和地图构建误差较大的问题,提出一种光流语义分割方法用于增加动态场景下的可运行性。将ORB-SLAM2系统与YOLOv5模型结合,对传入图像...针对同步定位与地图建立(simultaneous localization and mapping,SLAM)算法在动态环境下存在位姿估计和地图构建误差较大的问题,提出一种光流语义分割方法用于增加动态场景下的可运行性。将ORB-SLAM2系统与YOLOv5模型结合,对传入图像提取特征点的同时将YOLOv5网络模型语义分割后的物体分为高、中、低动态物体。利用运动一致性检测算法,对三种检测物体动态阈值判断,辨别其是否需要剔除特征点,增加ORB-SLAM2算法在动态环境下的运行精度。为加快系统运行速度,用LK光流法加快普通帧与普通帧之间的匹配,其原理为使用LK光流匹配特征点代替ORB特征点匹配,大大的缩小运行时间,同时运行误差变化不大。实验在TUM数据集下测试,平均每一帧提取2000个特征点,在增加LK光流后缩短0.01 s以上,若在900帧数据集下,可缩短9 s.其绝对轨迹误差对比于ORB-SLAM2和DS-SLAM平均提升在95%与30%以上,证明了算法在动态场景下良好的运行精度与鲁棒性。展开更多
近年来,无人机因体积小、灵活性好等优势被广泛应用在车辆跟踪领域。当无人机在高空飞行时,其捕捉的图像中车辆目标存在像素点少、拥挤以及被遮挡的情况。现有的多目标跟踪研究方法在车辆被遮挡过程中发生非线性运动时,使用卡尔曼滤波预...近年来,无人机因体积小、灵活性好等优势被广泛应用在车辆跟踪领域。当无人机在高空飞行时,其捕捉的图像中车辆目标存在像素点少、拥挤以及被遮挡的情况。现有的多目标跟踪研究方法在车辆被遮挡过程中发生非线性运动时,使用卡尔曼滤波预测,会出现车辆位置预测不准确的问题。为了解决这些问题,采用先检测后跟踪(tracking by detection,TBD)范式,对YOLOv8检测算法进行改进,在网络结构中引入了BiFormer稀疏动态注意力模块,用于提取小目标特征信息。同时使用轻量级上采样算子CARAFE替换原最近邻插值上采样,减少上采样过程中小目标特征丢失的问题。提出一种轻量化跟踪模型FA-SORT,针对SORT算法提出三点改进:改进KF、添加速度方向一致性匹配和检测值匹配。在自制地组合了多个车辆数据集上验证改进的YOLOv8算法。实验结果表明,与YOLOv8相比,精确率(precision)提高了0.97%,召回率(recall)提高了0.898%。对所提出的FA-SORT算法使用UAVDT数据集进行验证,结果表明,与现有的多目标跟踪算法相比,HOTA指标首个达到70.05%,IDF1达到87.45%,跟踪速度达到29.93 FPS。验证了FA-SORT跟踪算法在多车辆跟踪任务中的优越性。展开更多
文摘With the explosive growth of false information on social media platforms, the automatic detection of multimodalfalse information has received increasing attention. Recent research has significantly contributed to multimodalinformation exchange and fusion, with many methods attempting to integrate unimodal features to generatemultimodal news representations. However, they still need to fully explore the hierarchical and complex semanticcorrelations between different modal contents, severely limiting their performance detecting multimodal falseinformation. This work proposes a two-stage detection framework for multimodal false information detection,called ASMFD, which is based on image aesthetic similarity to segment and explores the consistency andinconsistency features of images and texts. Specifically, we first use the Contrastive Language-Image Pre-training(CLIP) model to learn the relationship between text and images through label awareness and train an imageaesthetic attribute scorer using an aesthetic attribute dataset. Then, we calculate the aesthetic similarity betweenthe image and related images and use this similarity as a threshold to divide the multimodal correlation matrixinto consistency and inconsistencymatrices. Finally, the fusionmodule is designed to identify essential features fordetectingmultimodal false information. In extensive experiments on four datasets, the performance of the ASMFDis superior to state-of-the-art baseline methods.
基金National Natural Science Foundation of China(No.62101219)Natural Science Foundation of Jiangsu Province(Nos.BK20201026,BK20210921)+1 种基金Science Foundation of Jiangsu Normal University(No.19XSRX006)Open Research Fund of Jiangsu Key Laboratory of Resources and Environmental Information Engineering(No.JS202107)。
文摘With the rapid development of Unmanned Aerial Vehicle(UAV)technology,change detection methods based on UAV images have been extensively studied.However,the imaging of UAV sensors is susceptible to environmental interference,which leads to great differences of same object between UAV images.Overcoming the discrepancy difference between UAV images is crucial to improving the accuracy of change detection.To address this issue,a novel unsupervised change detection method based on structural consistency and the Generalized Fuzzy Local Information C-means Clustering Model(GFLICM)was proposed in this study.Within this method,the establishment of a graph-based structural consistency measure allowed for the detection of change information by comparing structure similarity between UAV images.The local variation coefficient was introduced and a new fuzzy factor was reconstructed,after which the GFLICM algorithm was used to analyze difference images.Finally,change detection results were analyzed qualitatively and quantitatively.To measure the feasibility and robustness of the proposed method,experiments were conducted using two data sets from the cities of Yangzhou and Nanjing.The experimental results show that the proposed method can improve the overall accuracy of change detection and reduce the false alarm rate when compared with other state-of-the-art change detection methods.
文摘Objective: The objective of the study is to verify the clinical validity of the following kits with the comparative experimental analysis and evaluate whether their performance can meet the clinical requirements, i.e. Class III in vitro diagnostic reagent “Herpes Simplex Virus (HSV) Type II Nucleic Acid Detection Kit (PCR-Fluorescence Probe Method)” of Daan Gene Co., Ltd. (Daan kit for short) and “Herpes Simplex Virus (HSV) Type II Nucleic Acid Detection Kit (Fluorescence PCR Method)” of Wuhan Biot Gene Co., Ltd. (Biot kit for short). Method: In the study process, the samples were divided into positive and negative groups according to the control test results, and the clinical application performance of Daan kit and Biot kit was evaluated by comparing their test results. Results: The results show that two kits indicate the same test results, i.e. 26 positive and 107 negative samples in a total of 133 male urethral discharge samples, and 32 positive and 238 negative samples in a total of 270 female cervical secretion samples. Conclusion: It can be concluded from the clinical test that Daan and Biot Herpes Simplex Virus (HSV) Type II Nuc- leic Acid Test Kits are reliable, accurate, safe, convenient for use, stable and high-value in the clinical application.
基金supported in part by the CAMS Innovation Fund for Medical Sciences,China(No.2019-I2M5-016)National Natural Science Foundation of China(No.62172246)+1 种基金the Youth Innovation and Technology Support Plan of Colleges and Universities in Shandong Province,China(No.2021KJ062)National Science Foundation of USA(Nos.IIS-1715985 and IIS1812606).
文摘Recently,a new research trend in our video salient object detection(VSOD)research community has focused on enhancing the detection results via model self-fine-tuning using sparsely mined high-quality keyframes from the given sequence.Although such a learning scheme is generally effective,it has a critical limitation,i.e.,the model learned on sparse frames only possesses weak generalization ability.This situation could become worse on“long”videos since they tend to have intensive scene variations.Moreover,in such videos,the keyframe information from a longer time span is less relevant to the previous,which could also cause learning conflict and deteriorate the model performance.Thus,the learning scheme is usually incapable of handling complex pattern modeling.To solve this problem,we propose a divide-and-conquer framework,which can convert a complex problem domain into multiple simple ones.First,we devise a novel background consistency analysis(BCA)which effectively divides the mined frames into disjoint groups.Then for each group,we assign an individual deep model on it to capture its key attribute during the fine-tuning phase.During the testing phase,we design a model-matching strategy,which could dynamically select the best-matched model from those fine-tuned ones to handle the given testing frame.Comprehensive experiments show that our method can adapt severe background appearance variation coupling with object movement and obtain robust saliency detection compared with the previous scheme and the state-of-the-art methods.
文摘近年来,无人机因体积小、灵活性好等优势被广泛应用在车辆跟踪领域。当无人机在高空飞行时,其捕捉的图像中车辆目标存在像素点少、拥挤以及被遮挡的情况。现有的多目标跟踪研究方法在车辆被遮挡过程中发生非线性运动时,使用卡尔曼滤波预测,会出现车辆位置预测不准确的问题。为了解决这些问题,采用先检测后跟踪(tracking by detection,TBD)范式,对YOLOv8检测算法进行改进,在网络结构中引入了BiFormer稀疏动态注意力模块,用于提取小目标特征信息。同时使用轻量级上采样算子CARAFE替换原最近邻插值上采样,减少上采样过程中小目标特征丢失的问题。提出一种轻量化跟踪模型FA-SORT,针对SORT算法提出三点改进:改进KF、添加速度方向一致性匹配和检测值匹配。在自制地组合了多个车辆数据集上验证改进的YOLOv8算法。实验结果表明,与YOLOv8相比,精确率(precision)提高了0.97%,召回率(recall)提高了0.898%。对所提出的FA-SORT算法使用UAVDT数据集进行验证,结果表明,与现有的多目标跟踪算法相比,HOTA指标首个达到70.05%,IDF1达到87.45%,跟踪速度达到29.93 FPS。验证了FA-SORT跟踪算法在多车辆跟踪任务中的优越性。
