Search

Text tracking is to track multiple texts in a video,and construct a trajectory for each text. Existing methodstackle this task by utilizing the tracking-by-detection frame-work, i.e., detecting the text instances in each frame andassociating the corresponding text instances in consecutiveframes. We argue that the tracking accuracy of this paradigmis severely limited in more complex scenarios, e.g., owing tomotion blur, etc., the missed detection of text instances causesthe break of the text trajectory. In addition, different textinstances with similar appearance are easily confused, leadingto the incorrect association of the text instances. To this end,a novel spatio-temporal complementary text tracking model isproposed in this paper. We leverage a Siamese ComplementaryModule to fully exploit the continuity characteristic of the textinstances in the temporal dimension, which effectively alleviatesthe missed detection of the text instances, and hence ensuresthe completeness of each text trajectory. We further integratethe semantic cues and the visual cues of the text instance intoa unified representation via a text similarity learning network,which supplies a high discriminative power in the presence oftext instances with similar appearance, and thus avoids the mis-association between them. Our method achieves state-of-the-art performance on several public benchmarks. The source codeis available at https://github.com/lsabrinax/VideoTextSCM., Comment: update Fig.7, in the third row of part (c), the second and third frame is wrong and we update the right pictures

Osmotic energy reserves in estuaries, coupled with the ubiquitous solar energy, could be harnessed through emerging nanofluidic membranes to reduce the energy crisis. Herein, we mixed WS

The checkpoint/restart mechanism is critical in a preemptive system because clusters with this mechanism will be improved in terms of fault tolerance, load balance, and resource utilization. As graphics processing units (GPUs) have more recently become commonplace with the advent of general-purpose computation, and open computing language (OpenCL) programs are portable across various CPUs and GPUs, it is increasingly important to set up checkpoint/restart mechanism in OpenCL programs. However, due to the complexity of the internal computational state of the GPU, there is currently no effective and reasonable checkpoint/restart scheme for OpenCL applications. This paper proposes a feasible system, checkpoint/restart state (CRState), to achieve checkpoint/restart in GPU kernels. The computation states including heap, data segments, local memory, stack and code segments in the underlying hardware are identified and concretized in order to establish an association between the underlying level state and the application level representation. Then, a pre-compiler is developed to insert primitives into OpenCL programs at compile time so that major components of the computation state will be extracted at runtime. Since the computation state is duplicated at application level, such OpenCL programs can be preempted and ported across heterogeneous devices. A comprehensive example and ten authoritative benchmark programs are selected to demonstrate the feasibility and effectiveness of the proposed system.

Delineating the crucial waves in electrocardiogram records is a paramount work for the automatic diagnosis system of heart diseases. In this paper, a novel method is described to determine the boundaries and the peaks of P waves, QRS complexes and T waves by utilizing twelve-lead electrocardiogram signals. It avoids the difficulty of setting the thresholds when determining the boundaries of crucial waves and also the trouble of selection of wavelet basis as the wavelet-based method does. The signals are first preprocessed by a bandpass filter. After that, the locations of QRS complexes are identified. And based on the QRS locations, adaptive search windows are set to detect the locations of P waves and T waves. Then, a method called local distance transform decides the wave boundary in each lead. Finally, the final boundary determination rule is applied to obtain reliable boundaries. We justify the performance of our algorithm on LUDB database. When the tolerance window interval is 40ms, the peak accuracies of P wave, QRS complex and T wave are all beyond 98% and their boundary accuracies are all above 96%. Compared with the derivative threshold method and the wavelet-based method where the tolerance window interval is 150ms, the algorithm shows a sensitivity and a positive predictive value of peaks and boundaries greater than or equal to 98.43% and 96.44% for the P wave, 99.89% and 99.86% for the QRS complex and 99.21% and 99.85% for the T wave. For the critera of average error and standard deviation, our method has the performance similar to those methods. In addition, our algorithm can also handle such several situations where the boundary determination of crucial waves is tough as high T wave, high noise and baseline wandering well.

Lung cancer is one of the most diagnosable forms of cancer worldwide. The early diagnoses of pulmonary nodules in computed tomography (CT) chest scans are crucial for potential patients. Recent researches have showed that the methods based on deep learning have made a significant progress for the medical diagnoses. However, the achievements on identification of pulmonary nodules are not yet satisfactory enough to be adopted in clinical practice. It is largely caused by either the existence of many false positives or the heavy time of processing. With the development of fully convolutional networks (FCNs), in this study, we proposed a new method of identifying the pulmonary nodules. The method segments the suspected nodules from their environments and then removes the false positives. Especially, it optimizes the network architecture for the identification of nodules rapidly and accurately. In order to remove the false positives, the suspected nodules are reduced using the 2D models. Furthermore, according to the significant differences between nodules and non-nodules in 3D shapes, the false positives are eliminated by integrating into the 3D models and classified via 3D CNNs. The experiments on 1000 patients indicate that our proposed method achieved 97.78% sensitivity rate for segmentation and 90.1% accuracy rate for detection. The maximum response time was less than 30 s and the average time was about 15 s. Graphical Abstract This paper has proposed a new method of identifying the pulmonary nodules. The method segments the suspected nodules from CT images and removes the false positives. As shown in the above, the proposed approach consists of three stages. In stage I, raw data are filtered and normalized. The clean normalized data are then segmented in stage II to extract the suspected nodular lesions through 2D FCNs. Stage III is to remove some false positives generated at stage II via 3D CNNs and outputs the final results. The experiments on 1000 patients indicate that our proposed method has achieved 97.78% sensitivity rate for segmentation and 90.1% accuracy rate for detection. The maximum response time was less than 30 s and the average time was about 15 s.

For two-dimensional (2D) continuity characteristics of pulmonary nodules CT images, a sequence segmentation model based on U-shaped structure network and Convolutional Long Short-Term Memory (ConvLSTM) network is proposed to fully obtain the context space characteristics of image slices. In order to solve the problem of limited number of annotated samples in pulmonary nodules segmentation task, a segmentation method based on multi-task learning framework is proposed, which uses the annotated data of different types of tasks to mine the potential common characteristics among tasks; aiming at the problem of unbalanced category distribution in pulmonary nodules segmentation task, the design method of unified loss function under the multi-task learning framework is studied, and an optimization strategy integrating image prior knowledge and dynamic adjustment of multi-task weight is proposed to ensure that each task can complete training and learning efficiently. The experiments based on the LIDC-IDRI dataset demonstrate that the multi-task learning method proposed in this paper for the segmentation of pulmonary nodules under weak supervision is optimized from the three aspects of model design, network structure and constraints, and the MIoU and DSC are improved to 79.23% and 82.26% respectively.

In segmentation task of pulmonary nodules CT image, the U-shaped convolution neural network (UNet) has made great progress in segmentation accuracy and stability. However, for the three-dimensional (3D) connection characteristics of medical sequence slices, 3D segmentation networks often can’t give full play to their potential and advantages constrained by such factors as the number of marked samples and the computing capacity of equipment. For this reason, this paper focuses on the fusion methods of 2D and 3D UNet in medical image segmentation task: the first is to give full play to the advantages of 2D and 3D networks respectively, and propose a segmentation network model based on two-level hierarchical structure; the second is to make full use of the context information between the image slices and propose a 2.5D segmentation network model. The experiments based on LIDC-IDRI dataset show that the above two methods achieve a good balance in terms of computational performance, segmentation accuracy and model stability.

Checkpoint/restart is an important mechanism to achieve fault tolerance, load balancing and resources sharing in a preemptive system. As Graphics Processing Unit (GPU) becomes quite popular in high performance computing as well as OpenCL programs are portable across various CPUs and GPUs, checkpoint/restart of OpenCL programs on GPUs is in demand. However, due to the intricacy of computation states inside GPUs, there is no effective checkpoint/restart scheme for heterogeneous devices now. This paper proposes a feasible system, CRState, to achieve checkpoint/restart in GPU kernels. With the assistant of a pre-compiler, the primitives are inserted into programs. In run-time, the computation state existing in the underlying hardware is concretized and reconstructed at application level and is ported to heterogeneous devices. Comprehensive experiments have been conducted to demonstrate CRState's feasibility and effectiveness. The experimental results also indicate that CRState has the potential to reschedule resources and balance workload across heterogeneous devices.

Lung cancer is one of the most diagnosable form of cancer worldwide. Recent researches have showed that the diagnoses of pulmonary nodules in Computed Tomography (CT) chest scans based on deep learning have made a significant progress for the medical diagnoses. However, the existence of many false positives or the high costs of processing time make it impossible to apply to clinical practice. Toward this purpose, this paper proposed a new image processing method to improve the performance by exploiting the power of acceleration technologies via OpenCL. We use parallel programming and pipeline models to parallelize the CT image preprocessing, and classify them by 3D CNNs according to the significant differences between nodules and non-nodules in 3D shapes. Extensive experimental results have shown that image processing can be accelerated significantly on GPU. In addition, the experiments on 500 patients indicate that our proposed method improved the performance by \(12.5\%\) and achieved \(97.78\%\) sensitivity rate for segmentation.

Police Patrol is a very useful method for managing the social security. Many researches focus on how to obtain the best strategy for police patrol. We got this research topic from The Sixth National Post-Graduate Mathematical Contest in Modeling. In the proposed approach, Greedy Policy and Randomized Strategy are combined with Genetic Algorithm to solve these problems, simulation result shows the assignment and patrol strategy for police cars have a good effect with great concealment.

Characters of PbGeTe single layer is likely affected by various factors. The adhesion of PbGeTe single layer and PbGeTe/ZnS multilayer deposited on Si substrate by PVD method is investigated by means of x-ray diffraction. The correlation of layer growing rate and the preferred orientation of Si wafer is studied by the grind angle to measure the thickness method. The particle structure of films on various surface situations is studied by the image analysis. It has been noticed, that the adhesion of PbGeTe single layer is stronger in strength than that of the PbTe single layer, which shows little correlation with the preferred orientation of the substrate. The adhesive strength of the films can be improved by inserting thin layer of Ge or oxide layer. We have found that the layer growing rate varies with the preferred orientation of the substrate, we have also noticed that the particle structure of the films can be affected by the roughness of the substrate and the polishing method. Finally, the refractive index of Pb1-xGexTe(x equals 0.08) single layer was calculated.© (1998) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.