Peak load and wind energy emission pressure rise more as wind energy penetration keeps growing,which affects the stabilization of the PS(power system).This paper suggests integrated optimal dispatching of thermal powe...Peak load and wind energy emission pressure rise more as wind energy penetration keeps growing,which affects the stabilization of the PS(power system).This paper suggests integrated optimal dispatching of thermal power generators and BESS(battery energy storage system)taking wind energy emission grading punishment and deep peak clipping into consideration.Firstly,in order to minimize wind abandonment,a hierarchical wind abandonment penalty strategy based on fuzzy control is designed and introduced,and the optimal grid-connected power of wind energy is determined as a result of minimizing the peak cutting cost of the system.Secondly,considering BESS and thermal power,the management approach of BESS-assisted virtual peak clipping of thermal power generators is aimed at reducing the degree of deep peak clipping of thermal power generators and optimizing the output of thermal power generators and the charging and discharging power of BESS.Finally,Give an example of how this strategy has been effective in reducing abandonment rates by 0.66% and 7.46% individually for different wind penetration programs,and the daily average can reduce the peak clipping power output of thermal power generators by 42.97 and 72.31 MWh and enhances the effect and economy of system peak clipping.展开更多
BACKGROUND Hem-o-lok clips are typically used to control the cystic duct and vessels during laparoscopic cholecystectomy(LC)and common bile duct exploration for stones in the bile duct and gallbladder.Here,we report a...BACKGROUND Hem-o-lok clips are typically used to control the cystic duct and vessels during laparoscopic cholecystectomy(LC)and common bile duct exploration for stones in the bile duct and gallbladder.Here,we report a unique example of Hem-o-lok clip movement towards the duodenal bulb after LC,appearing as a submucosal tumor(SMT).Additionally,we provide initial evidence of gradual and evolving endoscopic manifestations of Hem-o-lok clip migration to the duodenal bulb wall and review the available literature.CASE SUMMARY A 72-year-old man underwent LC for gallstones,and Hem-o-lok clips were used to ligate both the cystic duct and cystic artery.Esophagogastroduodenoscopy(EGD)2 years later revealed an SMT-like lesion in the duodenal bulb.Due to the symptomatology,the clinical examination did not reveal any major abnormalities,and the patient was followed up as an outpatient.A repeat EGD performed 5 months later revealed an SMT-like lesion in the duodenal bulb with raised edges and a central depression.A third EGD was conducted,during which a Hem-o-lok clip was discovered connected to the front side of the duodenum.The clip was extracted easily using biopsy forceps,and no complications occurred.Two months after the fourth EGD,the scar was surrounded by normal mucosa.CONCLUSION Clinicians should be aware of potential post-LC complications.Hem-o-lok clips should be removed if symptomatic.展开更多
视频-文本检索作为一项被广泛应用于现实生活中的多模态检索技术受到越来越多的研究者的关注.近来,大部分视频文本工作通过利用大规模预训练模型中所学到的视觉与语言之间的匹配关系来提升文本视频间跨模态检索效果.然而,这些方法忽略...视频-文本检索作为一项被广泛应用于现实生活中的多模态检索技术受到越来越多的研究者的关注.近来,大部分视频文本工作通过利用大规模预训练模型中所学到的视觉与语言之间的匹配关系来提升文本视频间跨模态检索效果.然而,这些方法忽略了视频、文本数据都是由一个个事件组合而成.倘若能捕捉视频事件与文本事件之间的细粒度相似性关系,将能帮助模型计算出更准确的文本与视频之间的语义相似性关系,进而提升文本视频间跨模态检索效果.因此,提出了一种基于CLIP生成多事件表示的视频文本检索方法(CLIP based multi-event representation generation for video-text retrieval,CLIPMERG).首先,通过利用大规模图文预训练模型CLIP的视频编码器(ViT)以及文本编码器(Tansformer)分别将视频、文本数据转换成视频帧token序列以及文本的单词token序列;然后,通过视频事件生成器(文本事件生成器)将视频帧token序列(单词token序列)转换成k个视频事件表示(k个文本事件表示);最后,通过挖掘视频事件表示与文本事件表示之间的细粒度关系以定义视频、文本间的语义相似性关系.在3个常用的公开视频文本检索数据集MSR-VTT,DiDeMo,LSMDC上的实验结果表明所提的CLIPMERG优于现有的视频文本检索方法.展开更多
Although most pipes are restrained by retaining clips in aircraft,the influence of the clip parameters on the vibration of the fluid-conveying pipe has not been revealed.By considering the clip width,a new dynamic mod...Although most pipes are restrained by retaining clips in aircraft,the influence of the clip parameters on the vibration of the fluid-conveying pipe has not been revealed.By considering the clip width,a new dynamic model of a fluid-conveying pipe restrained by an intermediate clip is established in this paper.To demonstrate the necessity of the proposed model,a half pipe model is established by modeling the clip as one end.By comparing the two models,it is found that the half pipe model overestimates the critical velocity and may estimate the dynamical behavior of the pipe incorrectly.In addition,with the increase in the clip stiffness,the conversion processes of the first two modes of the pipe are shown.Furthermore,by ignoring the width of the clip,the effect of the flow velocity on the accuracy of a concentrated restraint clip model is presented.When the flow velocity is close to the critical velocity,the accuracy of the concentrated restraint clip model significantly reduces,especially when the width of the clip is large.In general,the contribution of this paper is to establish a dynamic model of the fluid-conveying pipe which can describe the influence of the clip parameters,and to demonstrate the necessity of this model.展开更多
A computer vision approach through Open AI’s CLIP, a model capable of predicting text-image pairs, is used to create an AI agent for Dixit, a game which requires creative linking between images and text. This paper c...A computer vision approach through Open AI’s CLIP, a model capable of predicting text-image pairs, is used to create an AI agent for Dixit, a game which requires creative linking between images and text. This paper calculates baseline accuracies for both the ability to match the correct image to a hint and the ability to match up with human preferences. A dataset created by previous work on Dixit is used for testing. CLIP is utilized through the comparison of a hint to multiple images, and previous hints, achieving a final accuracy of 0.5011 which surpasses previous results.展开更多
当前,随着国内摄像头数量的迅猛增长,每天所产生的庞大视频数据不仅对人力和物力资源构成了巨大的负担,而且导致了昂贵的成本开支。针对这一问题,本研究聚焦于解决细粒度目标检测领域存在的具体问题。本研究基于深度学习技术,结合Yolov...当前,随着国内摄像头数量的迅猛增长,每天所产生的庞大视频数据不仅对人力和物力资源构成了巨大的负担,而且导致了昂贵的成本开支。针对这一问题,本研究聚焦于解决细粒度目标检测领域存在的具体问题。本研究基于深度学习技术,结合Yolov4目标检测和CLIP特征分析,提出了一种综合的图像分析方法,以降低视频数据处理的成本。目前,现有的细粒度目标检测方法在处理大规模视频数据时面临着一系列挑战。这些挑战包括但不限于人工标注成本太高,而且无法保证标注的全面性,人工标注不如用户反馈及时有效;泛化能力只太弱,定制化成本太高,大多数AI任务都需要case by case实现。为了解决这些问题,本研究首先利用Yolov4模型对输入图像进行人物检测,以高效地实现目标的准确分割。随后,针对每个分割的人物,本实验采用CLIP模型进行深度特征分析,其泛化能力强且训练语料完全不需要人工标注的特点使捕捉图像和语言之间的语义精准关联。通过本研究的实验结果,本研究验证了该方法在人物检测方面的卓越表现,并展示了在基于CLIP的特征分析中显著的语义一致性。这一创新方法有望显著降低视频数据处理的成本和工作量,为细粒度目标检测领域的进一步研究提供了新的方向。展开更多
BACKGROUND Posterior reversible encephalopathy syndrome(PRES)is characterized mainly by occipital and parietal lobe involvement,which can be reversible within a few days.Herein,we report a rare case of PRES that devel...BACKGROUND Posterior reversible encephalopathy syndrome(PRES)is characterized mainly by occipital and parietal lobe involvement,which can be reversible within a few days.Herein,we report a rare case of PRES that developed after craniotomy for an unruptured intracranial aneurysm(UIA).CASE SUMMARY A 59-year-old man underwent clipping surgery for the treatment of UIA arising from the left middle cerebral artery.Clipping surgery was performed uneventfully,and he regained consciousness quickly immediately after the surgery.At the 4th hour after surgery,he developed a disorder of consciousness and aphasia.Magnetic resonance imaging revealed cortical and subcortical T2/FLAIR hyperintensities in the parietal,occipital,and frontal lobes ipsilaterally,without restricted diffusion,consistent with unilateral PRES.With conservative treatment,his symptoms and radiological findings almost completely disappeared within weeks.In our case,the important causative factor of PRES was suspected to be a sudden increase in cerebral perfusion pressure associated with temporary M1 occlusion.CONCLUSION Our unique case highlights that,to our knowledge,this is the second report of PRES developing after craniotomy for the treatment of UIA.Surgeons must keep PRES in mind as one of the causes of perioperative neurological abnormality following clipping of an UIA.展开更多
计算机视觉(Computer Vision,CV)与自然语言处理(Natural Language Processing,NLP)技术已逐渐趋于成熟,结合视觉和语言的多模态领域技术将成为学界和业界的研究热点。文章使用CLIP预训练模型,结合图像与语言两种模态信息,进一步将图像...计算机视觉(Computer Vision,CV)与自然语言处理(Natural Language Processing,NLP)技术已逐渐趋于成熟,结合视觉和语言的多模态领域技术将成为学界和业界的研究热点。文章使用CLIP预训练模型,结合图像与语言两种模态信息,进一步将图像拓展至视频,利用FFmpeg处理视频,并对视频与文本信息进行嵌入(embedding)和余弦相似度匹配,从而实现利用纯文本检索视频中符合该文本语义的片段。展开更多
基金supported by Jilin Province Higher Education Teaching Reform Research Project in 2021(JLJY202186163419).
文摘Peak load and wind energy emission pressure rise more as wind energy penetration keeps growing,which affects the stabilization of the PS(power system).This paper suggests integrated optimal dispatching of thermal power generators and BESS(battery energy storage system)taking wind energy emission grading punishment and deep peak clipping into consideration.Firstly,in order to minimize wind abandonment,a hierarchical wind abandonment penalty strategy based on fuzzy control is designed and introduced,and the optimal grid-connected power of wind energy is determined as a result of minimizing the peak cutting cost of the system.Secondly,considering BESS and thermal power,the management approach of BESS-assisted virtual peak clipping of thermal power generators is aimed at reducing the degree of deep peak clipping of thermal power generators and optimizing the output of thermal power generators and the charging and discharging power of BESS.Finally,Give an example of how this strategy has been effective in reducing abandonment rates by 0.66% and 7.46% individually for different wind penetration programs,and the daily average can reduce the peak clipping power output of thermal power generators by 42.97 and 72.31 MWh and enhances the effect and economy of system peak clipping.
文摘BACKGROUND Hem-o-lok clips are typically used to control the cystic duct and vessels during laparoscopic cholecystectomy(LC)and common bile duct exploration for stones in the bile duct and gallbladder.Here,we report a unique example of Hem-o-lok clip movement towards the duodenal bulb after LC,appearing as a submucosal tumor(SMT).Additionally,we provide initial evidence of gradual and evolving endoscopic manifestations of Hem-o-lok clip migration to the duodenal bulb wall and review the available literature.CASE SUMMARY A 72-year-old man underwent LC for gallstones,and Hem-o-lok clips were used to ligate both the cystic duct and cystic artery.Esophagogastroduodenoscopy(EGD)2 years later revealed an SMT-like lesion in the duodenal bulb.Due to the symptomatology,the clinical examination did not reveal any major abnormalities,and the patient was followed up as an outpatient.A repeat EGD performed 5 months later revealed an SMT-like lesion in the duodenal bulb with raised edges and a central depression.A third EGD was conducted,during which a Hem-o-lok clip was discovered connected to the front side of the duodenum.The clip was extracted easily using biopsy forceps,and no complications occurred.Two months after the fourth EGD,the scar was surrounded by normal mucosa.CONCLUSION Clinicians should be aware of potential post-LC complications.Hem-o-lok clips should be removed if symptomatic.
文摘视频-文本检索作为一项被广泛应用于现实生活中的多模态检索技术受到越来越多的研究者的关注.近来,大部分视频文本工作通过利用大规模预训练模型中所学到的视觉与语言之间的匹配关系来提升文本视频间跨模态检索效果.然而,这些方法忽略了视频、文本数据都是由一个个事件组合而成.倘若能捕捉视频事件与文本事件之间的细粒度相似性关系,将能帮助模型计算出更准确的文本与视频之间的语义相似性关系,进而提升文本视频间跨模态检索效果.因此,提出了一种基于CLIP生成多事件表示的视频文本检索方法(CLIP based multi-event representation generation for video-text retrieval,CLIPMERG).首先,通过利用大规模图文预训练模型CLIP的视频编码器(ViT)以及文本编码器(Tansformer)分别将视频、文本数据转换成视频帧token序列以及文本的单词token序列;然后,通过视频事件生成器(文本事件生成器)将视频帧token序列(单词token序列)转换成k个视频事件表示(k个文本事件表示);最后,通过挖掘视频事件表示与文本事件表示之间的细粒度关系以定义视频、文本间的语义相似性关系.在3个常用的公开视频文本检索数据集MSR-VTT,DiDeMo,LSMDC上的实验结果表明所提的CLIPMERG优于现有的视频文本检索方法.
基金supported by the National Science Funds for Distinguished Young Scholars of China(No.12025204)the Shanghai Municipal Education Commission of China(No.2019-01-07-00-09-E00018)。
文摘Although most pipes are restrained by retaining clips in aircraft,the influence of the clip parameters on the vibration of the fluid-conveying pipe has not been revealed.By considering the clip width,a new dynamic model of a fluid-conveying pipe restrained by an intermediate clip is established in this paper.To demonstrate the necessity of the proposed model,a half pipe model is established by modeling the clip as one end.By comparing the two models,it is found that the half pipe model overestimates the critical velocity and may estimate the dynamical behavior of the pipe incorrectly.In addition,with the increase in the clip stiffness,the conversion processes of the first two modes of the pipe are shown.Furthermore,by ignoring the width of the clip,the effect of the flow velocity on the accuracy of a concentrated restraint clip model is presented.When the flow velocity is close to the critical velocity,the accuracy of the concentrated restraint clip model significantly reduces,especially when the width of the clip is large.In general,the contribution of this paper is to establish a dynamic model of the fluid-conveying pipe which can describe the influence of the clip parameters,and to demonstrate the necessity of this model.
文摘A computer vision approach through Open AI’s CLIP, a model capable of predicting text-image pairs, is used to create an AI agent for Dixit, a game which requires creative linking between images and text. This paper calculates baseline accuracies for both the ability to match the correct image to a hint and the ability to match up with human preferences. A dataset created by previous work on Dixit is used for testing. CLIP is utilized through the comparison of a hint to multiple images, and previous hints, achieving a final accuracy of 0.5011 which surpasses previous results.
文摘当前,随着国内摄像头数量的迅猛增长,每天所产生的庞大视频数据不仅对人力和物力资源构成了巨大的负担,而且导致了昂贵的成本开支。针对这一问题,本研究聚焦于解决细粒度目标检测领域存在的具体问题。本研究基于深度学习技术,结合Yolov4目标检测和CLIP特征分析,提出了一种综合的图像分析方法,以降低视频数据处理的成本。目前,现有的细粒度目标检测方法在处理大规模视频数据时面临着一系列挑战。这些挑战包括但不限于人工标注成本太高,而且无法保证标注的全面性,人工标注不如用户反馈及时有效;泛化能力只太弱,定制化成本太高,大多数AI任务都需要case by case实现。为了解决这些问题,本研究首先利用Yolov4模型对输入图像进行人物检测,以高效地实现目标的准确分割。随后,针对每个分割的人物,本实验采用CLIP模型进行深度特征分析,其泛化能力强且训练语料完全不需要人工标注的特点使捕捉图像和语言之间的语义精准关联。通过本研究的实验结果,本研究验证了该方法在人物检测方面的卓越表现,并展示了在基于CLIP的特征分析中显著的语义一致性。这一创新方法有望显著降低视频数据处理的成本和工作量,为细粒度目标检测领域的进一步研究提供了新的方向。
基金Supported by Clinical Research Grant from Pusan National University Hospital in 2022.
文摘BACKGROUND Posterior reversible encephalopathy syndrome(PRES)is characterized mainly by occipital and parietal lobe involvement,which can be reversible within a few days.Herein,we report a rare case of PRES that developed after craniotomy for an unruptured intracranial aneurysm(UIA).CASE SUMMARY A 59-year-old man underwent clipping surgery for the treatment of UIA arising from the left middle cerebral artery.Clipping surgery was performed uneventfully,and he regained consciousness quickly immediately after the surgery.At the 4th hour after surgery,he developed a disorder of consciousness and aphasia.Magnetic resonance imaging revealed cortical and subcortical T2/FLAIR hyperintensities in the parietal,occipital,and frontal lobes ipsilaterally,without restricted diffusion,consistent with unilateral PRES.With conservative treatment,his symptoms and radiological findings almost completely disappeared within weeks.In our case,the important causative factor of PRES was suspected to be a sudden increase in cerebral perfusion pressure associated with temporary M1 occlusion.CONCLUSION Our unique case highlights that,to our knowledge,this is the second report of PRES developing after craniotomy for the treatment of UIA.Surgeons must keep PRES in mind as one of the causes of perioperative neurological abnormality following clipping of an UIA.
文摘计算机视觉(Computer Vision,CV)与自然语言处理(Natural Language Processing,NLP)技术已逐渐趋于成熟,结合视觉和语言的多模态领域技术将成为学界和业界的研究热点。文章使用CLIP预训练模型,结合图像与语言两种模态信息,进一步将图像拓展至视频,利用FFmpeg处理视频,并对视频与文本信息进行嵌入(embedding)和余弦相似度匹配,从而实现利用纯文本检索视频中符合该文本语义的片段。
