We analyzed the spatiotemporal variations in surface air temperature and key climate change indicators over the Tibetan Plateau during a common valid period from 1979 to 2018 to evaluate the performance of different d...We analyzed the spatiotemporal variations in surface air temperature and key climate change indicators over the Tibetan Plateau during a common valid period from 1979 to 2018 to evaluate the performance of different datasets on various timescales.We used observations from 22 in-situ observation sites,the CRA-40/Land(CRA)reanalysis dataset,the China Meteorological Forcing Dataset(CMFD),and the ERA-Interim(ERA)reanalysis dataset.The three datasets are spatially consistent with the in-situ observations,but slightly underestimate the annual mean surface air temperature.The daily mean surface air temperature estimated by the CRA,CMFD,and ERA datasets is closer to the in-situ observations after correction for elevation.The CMFD shows the best performance in simulating the annual mean surface air temperature over the Tibetan Plateau,followed by the CRA and ERA datasets with comparable performances.The CMFD is relatively accurate in simulating the daily mean surface air temperature over the Tibetan Plateau on an annual scale,whereas both the CRA and ERA datasets perform better in summer than in winter.The increasing trends in the annual mean surface air temperature over the Tibetan Plateau from 1979 to 2018 reflected by the CRA dataset and the CMFD are 0.5℃(10 yr)^(-1),similar to the in-situ observations,whereas the warming rate in the ERA dataset is only 0.3℃(10 yr)^(-1).The trends in the length of the growing season derived from the in-situ observations,the CRA,CMFD,and ERA datasets are 5.3,4.8,6.1,and 3.2 day(10 yr)^(-1),respectively.Our analyses suggest that both the CRA dataset and the CMFD perform better than the ERA dataset in modeling the changes in surface air temperature over the Tibetan Plateau.展开更多
In recent years,the detection of image copy-move forgery(CMFD)has become a critical challenge in verifying the authenticity of digital images,particularly as image manipulation techniques evolve rapidly.While deep con...In recent years,the detection of image copy-move forgery(CMFD)has become a critical challenge in verifying the authenticity of digital images,particularly as image manipulation techniques evolve rapidly.While deep convolutional neural networks(DCNNs)have been widely employed for CMFD tasks,they are often hindered by a notable limitation:the progressive reduction in spatial resolution during the encoding process,which leads to the loss of critical image details.These details are essential for the accurate detection and localization of image copy-move forgery.To overcome the limitations of existing methods,this paper proposes a Transformer-based approach for CMFD and localization as an alternative to conventional DCNN-based techniques.The proposed method employs a Transformer structure as an encoder to process images in a sequence-to-sequence manner,substituting the feature correlation calculations of previous methods with self-attention computations.This allows the model to capture long-range dependencies and contextual nuances within the image,preserving finer details that are typically lost in DCNN-based approaches.Moreover,an appropriate decoder is utilized to ensure precise reconstruction of image features,thereby enhancing both the detection accuracy and localization precision.Experimental results demonstrate that the proposed model achieves superior performance on benchmark datasets,such as USCISI,for image copy-move forgery detection.These results show the potential of Transformer architectures in advancing the field of image forgery detection and offer promising directions for future research.展开更多
基金Supported by the Second Tibetan Plateau Scientific Expedition and Research Program(2019QZKK1001)Science Funds from Beijing Meteorological Service(BMBKJ202003008)。
文摘We analyzed the spatiotemporal variations in surface air temperature and key climate change indicators over the Tibetan Plateau during a common valid period from 1979 to 2018 to evaluate the performance of different datasets on various timescales.We used observations from 22 in-situ observation sites,the CRA-40/Land(CRA)reanalysis dataset,the China Meteorological Forcing Dataset(CMFD),and the ERA-Interim(ERA)reanalysis dataset.The three datasets are spatially consistent with the in-situ observations,but slightly underestimate the annual mean surface air temperature.The daily mean surface air temperature estimated by the CRA,CMFD,and ERA datasets is closer to the in-situ observations after correction for elevation.The CMFD shows the best performance in simulating the annual mean surface air temperature over the Tibetan Plateau,followed by the CRA and ERA datasets with comparable performances.The CMFD is relatively accurate in simulating the daily mean surface air temperature over the Tibetan Plateau on an annual scale,whereas both the CRA and ERA datasets perform better in summer than in winter.The increasing trends in the annual mean surface air temperature over the Tibetan Plateau from 1979 to 2018 reflected by the CRA dataset and the CMFD are 0.5℃(10 yr)^(-1),similar to the in-situ observations,whereas the warming rate in the ERA dataset is only 0.3℃(10 yr)^(-1).The trends in the length of the growing season derived from the in-situ observations,the CRA,CMFD,and ERA datasets are 5.3,4.8,6.1,and 3.2 day(10 yr)^(-1),respectively.Our analyses suggest that both the CRA dataset and the CMFD perform better than the ERA dataset in modeling the changes in surface air temperature over the Tibetan Plateau.
基金support from the General Program of the National Natural Science Foundation of China(GrantNo.62072123)Key R&D Initiatives in Guangdong Province(Grant No.2021B0101220006)+2 种基金the Guangdong Provincial Department of Education’s Key Field Projects for Ordinary Colleges and Universities(Grant Nos.2020ZDZX3059,2022ZDZX1012,2023ZDZX1008)Key R&D Projects in Jiangxi Province(Grant No.20212BBE53002)Key R&D Projects in Yichun City(Grant No.20211YFG4270).
文摘In recent years,the detection of image copy-move forgery(CMFD)has become a critical challenge in verifying the authenticity of digital images,particularly as image manipulation techniques evolve rapidly.While deep convolutional neural networks(DCNNs)have been widely employed for CMFD tasks,they are often hindered by a notable limitation:the progressive reduction in spatial resolution during the encoding process,which leads to the loss of critical image details.These details are essential for the accurate detection and localization of image copy-move forgery.To overcome the limitations of existing methods,this paper proposes a Transformer-based approach for CMFD and localization as an alternative to conventional DCNN-based techniques.The proposed method employs a Transformer structure as an encoder to process images in a sequence-to-sequence manner,substituting the feature correlation calculations of previous methods with self-attention computations.This allows the model to capture long-range dependencies and contextual nuances within the image,preserving finer details that are typically lost in DCNN-based approaches.Moreover,an appropriate decoder is utilized to ensure precise reconstruction of image features,thereby enhancing both the detection accuracy and localization precision.Experimental results demonstrate that the proposed model achieves superior performance on benchmark datasets,such as USCISI,for image copy-move forgery detection.These results show the potential of Transformer architectures in advancing the field of image forgery detection and offer promising directions for future research.
