A reconfigurable computing architecture based on Field Programmable Gate Array (FPGA) technology is implemented for the Electrical Capacitance Tomography (ECT) system. The ECT system is used to image the multi-phase f...
详细信息
A reconfigurable computing architecture based on Field Programmable Gate Array (FPGA) technology is implemented for the Electrical Capacitance Tomography (ECT) system. The ECT system is used to image the multi-phase flow when gas/liquid or solid/liquid phases occurs. In the ECT systems, an exhaustive computational image reconstruction algorithm has to vastly processed large amount of data. The software algorithms and hardware parameters are adjusted based on a Hardware-software codesign process using commercially available tools. The hardware system consists of capacitive sensors, wireless nodes and FPGA module. Rr4wesults show that implementing the ECT image reconstruction algorithm on the FPGA platform achives fast performance and small design density.
It is a trend now that computing power through parallelism is provided by multi-core systems or heterogeneous architectures for High Performance Computing (HPC) and scientific computing. Although many algorithms have ...
详细信息
ISBN:
(纸本)9781509052530
It is a trend now that computing power through parallelism is provided by multi-core systems or heterogeneous architectures for High Performance Computing (HPC) and scientific computing. Although many algorithms have been proposed and implemented using sequential computing, alternative parallel solutions provide more suitable and high performance solutions to the same problems. In this paper, three parallelization strategies are proposed and implemented for a dynamic programming based cloud smoothing application, using both shared memory and non-shared memory approaches. The experiments are performed on NVIDIA GeForce GT750m and Tesla K20m, two GPU accelerators of Kepler architecture. Detailed performance analysis is presented on partition granularity at block and thread levels, memory access efficiency and computational complexity. The evaluations described show high approximation of results with high efficiency in the parallel implementations, and these strategies can be adopted in similar data analysis and processing applications.
With rapid development in mobile devices with high quality image and video processing capabilities, it is desirable or necessary to implement steganography technology within such devices in some applications such as s...
详细信息
ISBN:
(纸本)9781509003051
With rapid development in mobile devices with high quality image and video processing capabilities, it is desirable or necessary to implement steganography technology within such devices in some applications such as source authentication. In this paper, we propose a method of implementing information hiding within a video with less processing time and memory requirement. The message to be secretly communicated is hidden in the video by modifying blocks of Discrete Wavelet Transform coefficients of feature regions. We choose corners as specific features. The localizable capability of both corner detection algorithm and the Discrete Wavelet Transform makes it possible for the entire embedding method to be localized. This makes the entire process of steganography computationally faster and memory efficient. Our simulations show that the proposed method has a processing time that is up to 2.6 times more efficient, in average, than other traditional methods without incurring memory overhead.
AprilTags and other passive fiducial markers require specialized algorithms to detect markers among other features in a natural scene. The vision processing steps generally dominate the computation time of a tag detec...
详细信息
ISBN:
(纸本)9781509037636
AprilTags and other passive fiducial markers require specialized algorithms to detect markers among other features in a natural scene. The vision processing steps generally dominate the computation time of a tag detection pipeline, so even small improvements in marker detection can translate to a faster tag detection system. We incorporated lessons learned from implementing and supporting the AprilTag system into this improved system. This work describes AprilTag 2, a completely redesigned tag detector that improves robustness and efficiency compared to the original AprilTag system. The tag coding scheme is unchanged, retaining the same robustness to false positives inherent to the coding system. The new detector improves performance with higher detection rates, fewer false positives, and lower computational time. Improved performance on small images allows the use of decimated input images, resulting in dramatic gains in detection speed.
Texture synthesis is a fast growing technique in imageprocessing, and has been widely used in real time processes. During this process a texture is taken as sample input and various methods and techniques are applied...
详细信息
ISBN:
(纸本)9781467384384
Texture synthesis is a fast growing technique in imageprocessing, and has been widely used in real time processes. During this process a texture is taken as sample input and various methods and techniques are applied to synthesis that texture to produce a large synthesized texture with user defined size and similar texture characteristics. Methods and techniques used are tiling and synthesis based on patches, pixels and exemplar. Quality of the synthesized texture and synthesis time is one of the major concerns during the process. A survey is taken on various techniques used to improve the quality and synthesis time of the texture.
Sophisticated computational imaging algorithms require both high performance and good energy-efficiency when executed on mobile devices. Recent trend has been to exploit the abundant data-level parallelism found in ge...
详细信息
Sophisticated computational imaging algorithms require both high performance and good energy-efficiency when executed on mobile devices. Recent trend has been to exploit the abundant data-level parallelism found in general purpose programmable GPUs. However, for low-power mobile use cases, generic GPUs consume excessive amounts of power. This paper proposes a programmable computational imaging processor with 16-bit half-precision SIMD floating point vector processing capabilities combined with power efficiency of an exposed datapath. In comparison to traditional VLIW architectures with similar computational resources, the exposed datapath reduces the register file traffic and complexity. These and the specific optimizations enabled by the explicit programming model enable extremely good power-performance. When synthesized on a 28nm ASIC technology, the accelerator consumes 71mW of power while running a state-of-the-art denoising algorithm, and occupies only 0.2mm 2 of chip area. For the algorithm, energy usage per frame is 7mJ, which is 10x less than the best found GPU-based implementation.
Compressed sensing is a technique that is suitable for compressing and recovering signals having sparse representations in certain bases. Compressed sensing has been widely used to optimize the measurement process of ...
详细信息
ISBN:
(纸本)9791092279115
Compressed sensing is a technique that is suitable for compressing and recovering signals having sparse representations in certain bases. Compressed sensing has been widely used to optimize the measurement process of power and bandwidth constrained systems like wireless body sensor network. The central issues with compressed sensing are mainly the construction of measurement matrices and the development of efficient recovery algorithms. In this paper, we proposed a simple and fast recovery algorithm which performed a thresholding in the discrete cosine transform domain. We validated it by recovering electrocardiogram and electromyogram signals taken from the Phyiobank database. The simulation and experimental results have shown that the proposed recovery algorithm was 25 and 12 times faster than orthogonal matching pursuit and stagewise orthogonal matching pursuit, respectively. In addition, depending on the compression ratio, the signal-to-noise ratio of recovered signals were improved up to 2 dB.
Early forest fire alarm systems are critical in making prompt response in the event of unexpected *** cameras,improvements in memory,and enhanced computation power have all enabled the design and real-time application...
详细信息
ISBN:
(纸本)9781467383196
Early forest fire alarm systems are critical in making prompt response in the event of unexpected *** cameras,improvements in memory,and enhanced computation power have all enabled the design and real-time application of fire detecting algorithms using light and smallsize embedded surveillance *** is vital in situations where the performance of traditional forest fire monitoring and detection techniques are *** paper presents a forest fire monitoring and detection method with visual sensors onboard unmanned aerial vehicle(UAV).Both color and motion features of fire are adopted for the design of the studied forest fire detection *** is for the purpose of improving fire detection performance,while reducing false alarm *** experiments are conducted to demonstrate the effectiveness of the studied forest fire detection methodologies.
Automatic object recognition for texture-less objects using computer vision is a difficult task in comparison of textured one since class discriminative information is rarely available. Herein, an algorithm to count s...
详细信息
Automatic object recognition for texture-less objects using computer vision is a difficult task in comparison of textured one since class discriminative information is rarely available. Herein, an algorithm to count such objects using shape and color attributes for recognition with scale, rotation and illumination invariance is proposed. Initially, the algorithm extracts shape and color features of the prototype image to find its instance in the real-time pre-processed scene image captured by the vision interface. The pre-processing is achieved by morphological boundary extraction and segmentation techniques. Color and shape features are extracted based on mean hue value and Hu-moments respectively from the obtained segments. SVM, kNN, neural network and tree-bagging are then applied for classification. Tree-bagging is found to eclipse over the other classifiers in terms of accuracy. Finally, the classified objects are counted and localized in the image by drawing bounding boxes around them. A desktop application of the proposed algorithm is also developed. To assess the performance of the proposed algorithm, experimentation has been carried out for various objects having different shapes and colors. The algorithm proved out to be robust and effective for recognition and counting of the texture-less objects.
Compare with the infra-ray light gaze tracking systems, the visible light gaze tracking (VLGT) design provides new applications to consumer electronics. However, the VLGT suffers from the technical difficulties of acc...
详细信息
ISBN:
(纸本)9781467383653
Compare with the infra-ray light gaze tracking systems, the visible light gaze tracking (VLGT) design provides new applications to consumer electronics. However, the VLGT suffers from the technical difficulties of accommodating various illumination conditions and unstable image features. These system design issues lead to the problem of low accuracy in estimating iris center location and high computational complexity. Leveraging from our previous work, we further improve the algorithm of ellipse matching for the iris region and the mapping function, whereas the average angular errors are less than 0.45° for both horizontal and vertical directions.
暂无评论