基于形变估计与运动补偿的医学CT图像层间超分辨率算法

针对医学断层图像层间分辨率较低的问题,提出了基于形变估计与运动补偿的医学CT图像层间超分辨率算法用于生成切片间图像,从而提高层间分辨率。首先利用U-Net对相邻两幅图像作多尺度特征提取与融合;其次,为了处理层间图像的复杂形变,使用基于自适应协作流的变形扭曲模块来实现相邻切片间的双向形变估计,设计层级信息递进融合模块对金字塔特征层进行特征聚合,对生成图进行运动补偿;最后经过后处理网络以减少异常像素点。该算法在两种CT数据集上进行验证,平均PSNR值分别达到了35.59 dB和30.76 dB,输出图能较好地恢复图像细节。与现有的一些方法对比,相关实验证明了该算法的有效性。

融合特征金字塔与可变形分离卷积的CT图像层间插值方法

针对计算机断层扫描(CT)序列图像中层间分辨率远低于层内分辨率的问题,提出了一种融合特征金字塔与可变形分离卷积的CT图像层间插值网络。该网络由两个模块构成:图像生成模块和图像增强模块。图像生成模块利用MultiResUNet实现对输入图像的特征提取,然后利用两组不同的可变形分离卷积分别对输入图像进行卷积运算生成候选层间图像;图像增强模块通过特征金字塔和图像合成网络实现对输入图像的多尺度特征融合,生成额外专注于上下文细节的图像,以进一步对候选层间图像的纹理细节进行增强。实验结果表明,所提层间插值网络生成的层间图像在定性和定量分析上均取得了较优效果,在图像边缘轮廓及纹理细节处理上表现更加出色,能有效提高CT序列图像的层间分辨率。

Potential Submarine Geologic Hazards at the Entrance of the Pearl River Estuary in the Northern South China Sea

The potential submarine geologic hazards were distinguished and categorized at the entrance of the Pearl River Estuary in the northern South China Sea, based upon the analysis of side scan sonar and sub-bottom profiler surveying data of about 2500 km long, in an area about 2000 km^2 around the Wanshan Archipelago. The data obtained in the survey has the highest spatial resolution by far, which could reveal more detailed distributions and characteristics of the geologic hazards than before. In the study region, three paleo-channels that were buried about 10–30 m below the seabed were found; more than 10 shallow gas areas were discovered. The sand waves found in the region were generally small and located near the islands, and twenty pockmarks found on the seabed were mostly concentrated to north of Zhuzhou island. There are also many man-made obstacles in the region, such as wreckages, pipeline, etc. In this paper we provide a detailed distribution map of the submarine geologic hazards in this region for the first time, and discuss their formation and harmfulness, which will provide a scientific basis for marine engineering construction, marine geologic disaster prevention and mitigation.

Multifractal Characteristics of Intermittent Turbulence in the Urban Canopy Layer

The multifractality of energy and thermal dissipation of fully developed intermittent turbulence is investigated in the urban canopy layer under unstable conditions by the singularity spectrum for the fractal dimensions of sets of singularities characterizing multifractals. In order to obtain high-order moment properties of smallscale turbulent dissipation in the inertial range, an ultrasonic anemometer with a high sampling frequency of 100 Hz was used. The authors found that the turbulent signal could be singular everywhere. Moreover, the singular exponents of energy and thermal dissipation rates are most frequently encountered at around 0.2, which is significantly smaller than the singular exponents for a wind tunnel at a moderate Reynolds number. The evidence indicates a higher intermittency of turbulence in the urban canopy layer at a high Reynolds number, which is demonstrated by the data with high temporal resolution. Furthermore, the temperature field is more intermittent than the velocity field. In addition, a large amount of samples could be used for verification of the results.

Fusion of ERT Images Based on Dempster-Shafer's Evidence Theory

In this paper, an electrical resistance tomography(ERT) imaging method is used as a classifier, and then the Dempster-Shafer's evidence theory with fuzzy clustering is integrated to improve the ERT image quality. The fuzzy clustering is applied to determining the key mass function, and dealing with the uncertain, incomplete and inconsistent measured imaging data in ERT. The proposed method was applied to images with the same investigated object under eight typical current drive patterns. Experiments were performed on a group of simulations using COMSOL Multiphysics tool and measurements with a piece of porcine lung and a pair of porcine kidneys as test materials. Compared with any single drive pattern, the proposed method can provide images with a spatial resolution of about 10% higher, while the time resolution was almost the same.

Recognizing single phospholipid vesicle collisions on carbon fiber nanoelectrode

We recognize the stochastic collisions of dopamine contained phospholipid vesicle on carbon fiber nanoelectrode, extending the observation of discrete collision events on nanoelectrode to biologically relevant analytes. To decrease noise interference to the technique, the dimensions of nanoelectrode was systematically investigated and optimized. Scanning electron microscopy(SEM) further supported the comparable sizes of nanoelectrode and vesicles(~100 nm in diameter). Vesicles collision and rupture on the surface of nanoelectrode led to the dopamine release from vesicles, which could be electrochemically oxidized to dopamine-o-quinone and detected via voltammetry. The comparable size of the nanoelectrode with vesicles and fast voltammetry allowed differentiation of single collision events from the current magnitudes and peak widths in the electrochemical collision experiments, which shows the efficacy of the method to characterize vesicle samples. This work provides a foundation upon which quantitative sensor technology might be built for the detection of dopamine contained vesicles with high spatial and temporal resolution.

A real-size MRPC developed for CBM-TOF

An MRPC prototype for the future CBM-TOF inner area has been developed.The counter assembled with low resistive glass consists of 10×0.22 mm gas gaps and 2×8 readout pads.The size of each pad is 20 mm×20 mm with an interval of 2 mm between pads.The signals of the counter are collected by 4 eight-pin connectors.Preliminary tests performed with cosmic-rays showed an efficiency over 98%and a time resolution around 60 ps.The measured crosstalk,dark current and dark rate were at a very low levels.Results under uniform irradiation performed at JINR with high energy deuteron showed a time resolution of40–50 ps and over 98%efficiency at a low flux rate,meanwhile at a flux up to 70 kHz/cm2,93%efficiency and 77 ps time resolution still remained.These excellent performances call fully meet the demand of CBM-TOF inner wall.

Retrieval of middle and upper atmospheric wind based on non-full circular fringe recorded by Fabry-Perot Inteferometer

Fabry-Perot Interferometer(FPI) has been used widely for wind measurements of the middle and upper atmosphere.To date, most of FPIs have been based on full-closed circular fringe, which needs 15–25 min to obtain a group of wind velocity(zonal and meridional). However, it is hard to improve the temporal resolution because full-closed circular fringe in several directions cannot be easily imaged onto the same Charge-Coupled Device(CCD) with enough airglow intensity. In this paper, a data processing method is proposed for non-full circular fringe of FPI, which can support CCD with enough area of observations in several directions simultaneously. The method is focused on the center determination of non-full fringe. It includes radial cross-section, peak coordinate determination, and center calculation. Based on the calculated center, the fringe is annular summed. Then its radius is determined subsequently using Gaussian fitting. Finally, the wind is retrieved from the fringe radius. For validation, fringes from two ground-based FPIs were used, which are deployed in Kelan(38.71°N, 111.58°E) and Xinglong(40.40°N, 117.59°E) in China. The results retrieved from non-full fringes of FPIs were compared with that from full-closed circular fringe. The averaged wind deviation between them demonstrates reasonable difference with 5.38 ms^-(1) for 892.0 nm airglow emission, 5.81 ms^-(1) for 630.0 nm emission, and 3.03 ms^-(1) for 557.7 nm emission. Besides, wind results of Xinglong FPI are compared roughly with measurements of meteor radar which is deployed in Ming Tombs of Beijing(40.3°N,116.2°E). Good agreement demonstrates that this method is robust enough for FPI wind retrieval of mesosphere and thermosphere.