image thinning algorithms are widely used in imageprocessing to simplify elaboration preserving geometrical features. Standard approaches are based on iterative methods and on distance transforms. Both techniques are...
详细信息
ISBN:
(纸本)9781479952205
image thinning algorithms are widely used in imageprocessing to simplify elaboration preserving geometrical features. Standard approaches are based on iterative methods and on distance transforms. Both techniques are well known to be computationally intensive. In this work we propose a parallel, fast and flexible hardware architecture for image thinning to achieve real-time performance. The test case is the 720 x 576 PAL standard video at 25 frame per second (fps). Synthesis was performed for a Stratix ii FPGA EP2S30 and for a standard cell 65 nm CMOS technology. The former showed a usage of 4 % slices and 1 % registers, the latter gave an occupation of 5 kgates for the processing core. The execution time for one frame was 0.03 s on the FPGA and 0.009 s on the 65 nm, resulting in a maximum throughput of 33 fps and 111 fps, respectively.
This paper presents a distributed solution for the development of deformable model-based medical image segmentation methods. The design and implementation stages of the segmentation methods usually require a lot of ti...
详细信息
This paper presents a parallel remeshing algorithm for distributed-memory architectures. It is an iterative parallel algorithm that divides the areas to be remeshed into multiple pieces which can be distributed to as ...
详细信息
ISBN:
(纸本)9788494284472
This paper presents a parallel remeshing algorithm for distributed-memory architectures. It is an iterative parallel algorithm that divides the areas to be remeshed into multiple pieces which can be distributed to as many processing elements as possible, in order for these pieces to be remeshed concurrently by a third-party sequential remesher. Then, remeshed pieces are reintegrated into the distributed mesh, and this process is iterated until all relevant areas of the mesh have been remeshed. Any sequential remesher can be used, provided it allows some of the mesh elements not to be modified, so as to preserve interfaces between pieces. Our method, which has been implemented in the PaMPA library, is validated by a set of experiments involving both isotropic and anisotropic meshes.
Dim objects tracking, like asteroids or satellites, require high performance algorithm. Track Before Detect (TBD) algorithms could be applied for such scenario, but the computation power is very demanding. Giant amoun...
详细信息
ISBN:
(纸本)9781479950812
Dim objects tracking, like asteroids or satellites, require high performance algorithm. Track Before Detect (TBD) algorithms could be applied for such scenario, but the computation power is very demanding. Giant amount of trajectories that could be processed is the main problem. The application of GPGPU allows the computation using optimized Downsampled Spatio Temporal TBD (ST-TBD) algorithm for small part of image sequences. Message Passing Interface (MPICH2) is applied for the image transmission and state space exchange between computers and GPGPUs. Analysis of the system with multiple processing computers, for bandwidth limited network, is provided. The system could be limited by the bandwidth of network interface, not by GPGPU, quite often. The derived formula for this system allows the design with best optimization of GPGPUs.
This paper reviews and compares the performance of several methods to detect target tracks in image sequences. The targets are assumed to be sub-pixel or not resolved by the imaging system, and moving over a static ba...
详细信息
ISBN:
(纸本)9781628410266
This paper reviews and compares the performance of several methods to detect target tracks in image sequences. The targets are assumed to be sub-pixel or not resolved by the imaging system, and moving over a static background. To process the resulting large amount of data requires simple, fast and robust processingmethods to quickly find and display tracks of moving targets in a single image. An object moving through a pixel in a scene will momentarily perturb the pixel intensity signal, introducing a change of both skewness and kurtosis in the intensity histogram relative to an undisturbed pixel. Numerical experiments show that for Gaussian and Poisson distributed system noise higher order moments (>2) perform better than second order detectors.
Non-negative matrix factorization (NMF) is a powerful dimension reduction method and has been widely used in many pattern recognition and computer vision problems. However, conventional NMF methods are neither robust ...
详细信息
ISBN:
(纸本)9781479919611
Non-negative matrix factorization (NMF) is a powerful dimension reduction method and has been widely used in many pattern recognition and computer vision problems. However, conventional NMF methods are neither robust enough as their loss functions are sensitive to outliers, nor discriminative because they completely ignore labels in a dataset. In this paper, we proposed a correntropy supervised NMF (CSNMF) to simultaneously overcome aforementioned deficiencies. In particular, CSNMF maximizes the correntropy between the data matrix and its reconstruction in low-dimensional space to inhibit outliers during learning the subspace, and narrows the minimizes the distances between coefficients of any two samples with the same class labels to enhance the subsequent classification performance. To solve CSNMF, we developed a multiplicative update rules and theoretically proved its convergence. Experimental results on popular face image datasets verify the effectiveness of CSNMF comparing with NMF, its supervised variants, and its robustified variants.
In our previous work, we have provided tools for an efficient characterization of biomedical images using Legendre and Zernike moments, showing their relevance as biomarkers for classifying image tiles coming from bon...
详细信息
In our previous work, we have provided tools for an efficient characterization of biomedical images using Legendre and Zernike moments, showing their relevance as biomarkers for classifying image tiles coming from bone tissue regeneration studies (Ujaldon, 2009) [24]. As part of our research quest for efficiency, we developed methods for accelerating those computations on GPUs (Martin-Requena and Ujaldon, 2011) [10,9]. This new stage of our work focuses on the efficient data partitioning to optimize the execution on many-cores and clusters of GPUs to attain gains up to three orders of magnitude when compared to the execution on multi-core CPUs of similar age and cost using 1 Mpixel images. We deploy a successive and successful chain of optimizations which exploit symmetries in trigonometric functions and access patterns to image pixels which are effectively combined with massive data parallelism on GPUs to enable (1) real-time processing for our set of input biomedical images, and (2) the use of high-resolution images in clinical practice. (C) 2013 Elsevier Inc. All rights reserved.
Positron Emission Mammography (PEM) imaging systems with the ability in detection of millimeter-sized tumors were developed in recent years. And some of them have been well used in clinical applications. In considerat...
详细信息
Positron Emission Mammography (PEM) imaging systems with the ability in detection of millimeter-sized tumors were developed in recent years. And some of them have been well used in clinical applications. In consideration of biopsy application, a double-plane PET is more practical than a ring detector for the convenience of breast immobilization. In this study, the possibility of the imaging-guided intervention under a double-plane PET was studied. The distance between the planes is changeable with different breast size. And the system matrix is computed in real-time for every plane spacing. The biopsy motivation under the double-plane PET requires a high sensitivity, great resolution performance and a quickly reconstruction speed. The sensitivity result shows the central value is 13.28% with the 6cm spacing and 200-750keV energy window. The resolution modeling method shows the resolution recovery achieved in the cross-plane. The 3D-OSEM algorithm is applied. The reconstruction is supported with GPU (GeForce GTX 660). For the GPU acceleration, both projection and back-projection procedures use the tube model. Each tube is distributed with a thread and calculated in parallel in the GPU. A double plane PET is simulated above Geant4 Application for Emission Tomography (GATE) software based on MC methods. With all the span tubes calculated, the GPU takes 19 seconds to reconstruct the image while the CPU takes 1037 seconds. The image quality was tested with different acquisition times. The series of biopsy images are showed. The point and rod source both could be imaged in the simulated biopsy process.
暂无评论