Quantitative prediction of phase content is of great importance to control and optimize the heat treat-ment process of steels.In this work,a model for predicting the phase content of tempered high carbon steels was pr...Quantitative prediction of phase content is of great importance to control and optimize the heat treat-ment process of steels.In this work,a model for predicting the phase content of tempered high carbon steels was proposed by taking a martensitic 100Cr6 bearing steel as a model case.The microstructural transformations during tempering were studied using thermal analysis,transmission electron microscopy(TEM),and X-ray diffraction(XRD).Kinetics analysis of thermal evolution by employing the isoconver-sional method,and assisted by TEM and XRD characterization,were performed to quantitatively estimate the volume fractions of different phases after tempering.A series of isothermal tempering experiments were designed to verify the model.The predicted results were in good agreement with the experimental results of XRD and electrolytic extraction measurements.展开更多
We propose a novel unsupervised image captioning method.Image captioning involves two fields of deep learning,natural language processing and computer vision.The excessive pursuit ofmodel evaluation results makes the ...We propose a novel unsupervised image captioning method.Image captioning involves two fields of deep learning,natural language processing and computer vision.The excessive pursuit ofmodel evaluation results makes the caption style generated by the model too monotonous,which is difficult to meet people’s demands for vivid and stylized image captions.Therefore,we propose an image captioning model that combines text style transfer and image emotion recognition methods,with which the model can better understand images and generate controllable stylized captions.The proposed method can automatically judge the emotion contained in the image through the image emotion recognition module,better understand the image content,and control the description through the text style transfermethod,thereby generating captions thatmeet people’s expectations.To our knowledge,this is the first work to use both image emotion recognition and text style control.展开更多
基金This work was financially supported by the National Natural Science Foundation of China(Nos.51605355 and 52104381)the National Key R&D Program of China(No.2020YFA0714900)+2 种基金“111 Project”(No.B17034)the Innovative Research TeamDevelopment Program of Ministry of Education of China(No.IRT_17R83)the China Postdoctoral Science Foundation(No.2021M702539)and the State Key Laboratory for Advanced Metals and Materials.
文摘Quantitative prediction of phase content is of great importance to control and optimize the heat treat-ment process of steels.In this work,a model for predicting the phase content of tempered high carbon steels was proposed by taking a martensitic 100Cr6 bearing steel as a model case.The microstructural transformations during tempering were studied using thermal analysis,transmission electron microscopy(TEM),and X-ray diffraction(XRD).Kinetics analysis of thermal evolution by employing the isoconver-sional method,and assisted by TEM and XRD characterization,were performed to quantitatively estimate the volume fractions of different phases after tempering.A series of isothermal tempering experiments were designed to verify the model.The predicted results were in good agreement with the experimental results of XRD and electrolytic extraction measurements.
基金supported by the National Key Research&Development Program (Grant No.2018YFC0831700)National Natural Science Foundation of China (Grant No.61671064,No.61732005).
文摘We propose a novel unsupervised image captioning method.Image captioning involves two fields of deep learning,natural language processing and computer vision.The excessive pursuit ofmodel evaluation results makes the caption style generated by the model too monotonous,which is difficult to meet people’s demands for vivid and stylized image captions.Therefore,we propose an image captioning model that combines text style transfer and image emotion recognition methods,with which the model can better understand images and generate controllable stylized captions.The proposed method can automatically judge the emotion contained in the image through the image emotion recognition module,better understand the image content,and control the description through the text style transfermethod,thereby generating captions thatmeet people’s expectations.To our knowledge,this is the first work to use both image emotion recognition and text style control.
