In the process of geothermal exploitation and utilization, the reinjection amount of used geothermal water in super-deep and porous reservoir is small and significantly decreases over time. This has been a worldwide p...In the process of geothermal exploitation and utilization, the reinjection amount of used geothermal water in super-deep and porous reservoir is small and significantly decreases over time. This has been a worldwide problem, which greatly restricts the exploitation and utilization of geothermal resources. Based on a large amount of experiments and researches, the reinjection research on the tail water of Xianyang No.2 well, which is carried out by combining the application of hydrogeochemical simulation, clogging mechanism research and the reinjection experiment, has achieved breakthrough results. The clogging mechanism and indoor simulation experiment results show: Factors affecting the tail water reinjection of Xianyang No.2 well mainly include chemical clogging, suspended solids clogging, gas clogging, microbial clogging and composite clogging, yet the effect of particle migration on clogging has not been found; in the process of reinjection, chemical clogging was mainly caused by carbonates(mainly calcite), silicates(mainly chalcedony), and a small amount of iron minerals, and the clogging aggravated when the temperature rose; suspended solids clogging also aggravated when the temperature rose, which showed that particles formed by chemical reaction had a certain proportion in suspended solids.展开更多
The effect of deep cryogenic treatment on the formation of reversed austenite （RA） in super martensitic stainless steel was investigated. RA was found to form in steels without （A） and with （B） deep cryogenic tr...The effect of deep cryogenic treatment on the formation of reversed austenite （RA） in super martensitic stainless steel was investigated. RA was found to form in steels without （A） and with （B） deep cryogenic treatment. The volume fraction of RA initially increased and then decreased with increasing tempering temperature over 550-- 750 ℃ for the two steels, which were quenched at 1050 ℃. In addition, for both with and without deep cryogenic treatment, the RA content reached a maximum value at 650 ℃ although the RA content in steel B was greater than that in steel A over the entire range of tempering temperatures. Furthermore, the hardness （HRC） of steel B was greater than that of steel A at tempering temperatures of 550--750 ℃. From these results, the basic mechanism for the formation of RA in steels A and B was determined to be Ni diffusion. However, there were more Ni enriched points, a lower degree of enrichment, and a shorter diffusion path in steel B. It needed to be noted that the shapes of the RA consisted of blocks and stripes in both steels. These shapes resulted because the RA redissolved and trans- formed to martensite along the martensitic lath boundaries when the tempering temperature was 650--750 ℃, and a portion of RA in the martensitie lath divided the originally wider martensitic laths into a number of thinner ones. In- terestingly, the RA redissolved more rapidly in steel B and consequently resulted in a stronger refining effect.展开更多
Single image super-resolution has attracted increasing attention and has a wide range of applications in satellite imaging, medical imaging, computer vision, security surveillance imaging, remote sensing, objection de...Single image super-resolution has attracted increasing attention and has a wide range of applications in satellite imaging, medical imaging, computer vision, security surveillance imaging, remote sensing, objection detection, and recognition. Recently, deep learning techniques have emerged and blossomed, producing " the state-of-the-art” in many domains. Due to their capability in feature extraction and mapping, it is very helpful to predict high-frequency details lost in low-resolution images. In this paper, we give an overview of recent advances in deep learning-based models and methods that have been applied to single image super-resolution tasks. We also summarize, compare and discuss various models from the past and present for comprehensive understanding and finally provide open problems and possible directions for future research.展开更多
Purpose: To detect small diagnostic signals such as lung nodules in chest radiographs, radiologists magnify a region-of-interest using linear interpolation methods. However, such methods tend to generate over-smoothed...Purpose: To detect small diagnostic signals such as lung nodules in chest radiographs, radiologists magnify a region-of-interest using linear interpolation methods. However, such methods tend to generate over-smoothed images with artifacts that can make interpretation difficult. The purpose of this study was to investigate the effectiveness of super-resolution methods for improving the image quality of magnified chest radiographs. Materials and Methods: A total of 247 chest X-rays were sampled from the JSRT database, then divided into 93 training cases with non-nodules and 154 test cases with lung nodules. We first trained two types of super-resolution methods, sparse-coding super-resolution (ScSR) and super-resolution convolutional neural network (SRCNN). With the trained super-resolution methods, the high-resolution image was then reconstructed using the super-resolution methods from a low-resolution image that was down-sampled from the original test image. We compared the image quality of the super-resolution methods and the linear interpolations (nearest neighbor and bilinear interpolations). For quantitative evaluation, we measured two image quality metrics: peak signal-to-noise ratio (PSNR) and structural similarity (SSIM). For comparative evaluation of the super-resolution methods, we measured the computation time per image. Results: The PSNRs and SSIMs for the ScSR and the SRCNN schemes were significantly higher than those of the linear interpolation methods (p p p Conclusion: Super-resolution methods provide significantly better image quality than linear interpolation methods for magnified chest radiograph images. Of the two tested schemes, the SRCNN scheme processed the images fastest;thus, SRCNN could be clinically superior for processing radiographs in terms of both image quality and processing speed.展开更多
近年来,深度强化学习在复杂决策和控制任务中得到了广泛应用,并在游戏AI领域展现了卓越性能。基于双重深度Q网络的方法,提出一种通过智能体与Super Mario Bros环境的持续交互、逐步学习并优化游戏策略。首先,利用gym-super-mario-bros...近年来,深度强化学习在复杂决策和控制任务中得到了广泛应用,并在游戏AI领域展现了卓越性能。基于双重深度Q网络的方法,提出一种通过智能体与Super Mario Bros环境的持续交互、逐步学习并优化游戏策略。首先,利用gym-super-mario-bros框架构建训练环境,并通过帧跳、灰度转换和图像缩放等技术提升训练效率。其次,智能体采用DDQN架构结合卷积神经网络进行特征提取,并通过经验回放和目标网络减少Q值波动。最后,通过衰减的epsilon-greedy策略平衡探索与利用。实验结果表明,该方法能有效提升智能体表现。展开更多
基金funded by National Science Foundation Project in 2015 (No.41472221)
文摘In the process of geothermal exploitation and utilization, the reinjection amount of used geothermal water in super-deep and porous reservoir is small and significantly decreases over time. This has been a worldwide problem, which greatly restricts the exploitation and utilization of geothermal resources. Based on a large amount of experiments and researches, the reinjection research on the tail water of Xianyang No.2 well, which is carried out by combining the application of hydrogeochemical simulation, clogging mechanism research and the reinjection experiment, has achieved breakthrough results. The clogging mechanism and indoor simulation experiment results show: Factors affecting the tail water reinjection of Xianyang No.2 well mainly include chemical clogging, suspended solids clogging, gas clogging, microbial clogging and composite clogging, yet the effect of particle migration on clogging has not been found; in the process of reinjection, chemical clogging was mainly caused by carbonates(mainly calcite), silicates(mainly chalcedony), and a small amount of iron minerals, and the clogging aggravated when the temperature rose; suspended solids clogging also aggravated when the temperature rose, which showed that particles formed by chemical reaction had a certain proportion in suspended solids.
文摘The effect of deep cryogenic treatment on the formation of reversed austenite （RA） in super martensitic stainless steel was investigated. RA was found to form in steels without （A） and with （B） deep cryogenic treatment. The volume fraction of RA initially increased and then decreased with increasing tempering temperature over 550-- 750 ℃ for the two steels, which were quenched at 1050 ℃. In addition, for both with and without deep cryogenic treatment, the RA content reached a maximum value at 650 ℃ although the RA content in steel B was greater than that in steel A over the entire range of tempering temperatures. Furthermore, the hardness （HRC） of steel B was greater than that of steel A at tempering temperatures of 550--750 ℃. From these results, the basic mechanism for the formation of RA in steels A and B was determined to be Ni diffusion. However, there were more Ni enriched points, a lower degree of enrichment, and a shorter diffusion path in steel B. It needed to be noted that the shapes of the RA consisted of blocks and stripes in both steels. These shapes resulted because the RA redissolved and trans- formed to martensite along the martensitic lath boundaries when the tempering temperature was 650--750 ℃, and a portion of RA in the martensitie lath divided the originally wider martensitic laths into a number of thinner ones. In- terestingly, the RA redissolved more rapidly in steel B and consequently resulted in a stronger refining effect.
基金the support from the Shanxi Hundred People Plan of China
文摘Single image super-resolution has attracted increasing attention and has a wide range of applications in satellite imaging, medical imaging, computer vision, security surveillance imaging, remote sensing, objection detection, and recognition. Recently, deep learning techniques have emerged and blossomed, producing " the state-of-the-art” in many domains. Due to their capability in feature extraction and mapping, it is very helpful to predict high-frequency details lost in low-resolution images. In this paper, we give an overview of recent advances in deep learning-based models and methods that have been applied to single image super-resolution tasks. We also summarize, compare and discuss various models from the past and present for comprehensive understanding and finally provide open problems and possible directions for future research.
文摘Purpose: To detect small diagnostic signals such as lung nodules in chest radiographs, radiologists magnify a region-of-interest using linear interpolation methods. However, such methods tend to generate over-smoothed images with artifacts that can make interpretation difficult. The purpose of this study was to investigate the effectiveness of super-resolution methods for improving the image quality of magnified chest radiographs. Materials and Methods: A total of 247 chest X-rays were sampled from the JSRT database, then divided into 93 training cases with non-nodules and 154 test cases with lung nodules. We first trained two types of super-resolution methods, sparse-coding super-resolution (ScSR) and super-resolution convolutional neural network (SRCNN). With the trained super-resolution methods, the high-resolution image was then reconstructed using the super-resolution methods from a low-resolution image that was down-sampled from the original test image. We compared the image quality of the super-resolution methods and the linear interpolations (nearest neighbor and bilinear interpolations). For quantitative evaluation, we measured two image quality metrics: peak signal-to-noise ratio (PSNR) and structural similarity (SSIM). For comparative evaluation of the super-resolution methods, we measured the computation time per image. Results: The PSNRs and SSIMs for the ScSR and the SRCNN schemes were significantly higher than those of the linear interpolation methods (p p p Conclusion: Super-resolution methods provide significantly better image quality than linear interpolation methods for magnified chest radiograph images. Of the two tested schemes, the SRCNN scheme processed the images fastest;thus, SRCNN could be clinically superior for processing radiographs in terms of both image quality and processing speed.
文摘近年来,深度强化学习在复杂决策和控制任务中得到了广泛应用,并在游戏AI领域展现了卓越性能。基于双重深度Q网络的方法,提出一种通过智能体与Super Mario Bros环境的持续交互、逐步学习并优化游戏策略。首先,利用gym-super-mario-bros框架构建训练环境,并通过帧跳、灰度转换和图像缩放等技术提升训练效率。其次,智能体采用DDQN架构结合卷积神经网络进行特征提取,并通过经验回放和目标网络减少Q值波动。最后,通过衰减的epsilon-greedy策略平衡探索与利用。实验结果表明,该方法能有效提升智能体表现。
