For satellite communication, large amount of data storage and transmission are involved as the satellites send data all the time, all day. Storing all these data and analyzing them for various purposes is possible usi...
详细信息
ISBN:
(纸本)9781467393393
For satellite communication, large amount of data storage and transmission are involved as the satellites send data all the time, all day. Storing all these data and analyzing them for various purposes is possible using small low cost memory devices only with the help of image compression. image compression is the process of removing the redundant information from the image and it can be stored to reduce the storage size, transmission bandwidth and time. image compression aims at removing duplication from the source image and is essential for applications such as transmission and storage in an efficient form. The objective of the work is to develop an efficient low power image compression algorithm which compress it with higher compression ratio in such a way that the output compressed image becomes compatible for satellite communication. The proposed system should own a light weight algorithm which has the characteristics of minimum power consumption, less compression time and should meet a higher compression ratio. To do image compression, quad tree fractal image compression and an adaptive fractal wavelet image compression algorithm are selected and their performance in terms of mean square error, ratio of compression and peak signal to noise ratio are evaluated.
Novel security detection technology are needed to meet the growing demand of subway operation. In this paper, an improved faulting detection algorithm for subway tunnel segment is proposed. A combined denoising techni...
详细信息
ISBN:
(纸本)9781467393393
Novel security detection technology are needed to meet the growing demand of subway operation. In this paper, an improved faulting detection algorithm for subway tunnel segment is proposed. A combined denoising technique is used to convert the depth image of faulting acquired by Kinect into binary image of the height difference which can be processed by digital image. The obvious advantage of this method is that it can avoid the interference of environmental light and improve the detection speed. In addition, we focus on how to classify and identify different types of faulting line based on the idea of deep learning algorithm. And a fast CNN (convolution neural network) has been built to classify and identify faulting line. Finally, it shows that the proposed algorithm is effective through the experimental data.
The proceedings contain 53 papers. The topics discussed include: high performance multi-standard architecture for DCT computation in H.264/AVC high profile and HEVC codecs;architecture and programming model support fo...
ISBN:
(纸本)9791092279016
The proceedings contain 53 papers. The topics discussed include: high performance multi-standard architecture for DCT computation in H.264/AVC high profile and HEVC codecs;architecture and programming model support for efficient heterogeneous computing on tightly-coupled shared-memory clusters;a neural model for hardware plasticity in artificial vision systems;system-level PMC-driven energy estimation models in RVC-CAL video codec specifications;a resource-aware nearest neighbor search algorithm for k-dimensional trees;accuracy and performance analysis of Harris corner computation on tightly-coupled processor arrays;a linear state model for PDR+WLAN positioning;SiPM based smart pixel for photon counting integrated streak camera;a coarse-grained reconfigurable wavelet denoiser exploiting the multi-dataflow composer tool;foreground object features extraction with GLCM texture descriptor in FPG;and noise-agnostic adaptive image filtering without training references on an evolvable hardware platform.
The possibility of more intuitive human-machine interfaces has sparked the development of new visual technologies. The way humans interact with elements of their environment should not be limited to the screens of pho...
详细信息
ISBN:
(纸本)9781509037889
The possibility of more intuitive human-machine interfaces has sparked the development of new visual technologies. The way humans interact with elements of their environment should not be limited to the screens of phones or computers. Other alternatives where a sensation of spatial freedom are under development. Projection systems, using continuous light on surrounding surfaces, represent a major area of exploration. The great level of development in artificial vision hardware and software tools enables the acquisition of data from the user and his/her environment, while in the background, a software can analyze in real time the variations of the scene without intervention of the user. This kind of data processing makes possible the integration between what the user is doing and seeing. The device proposed in this paper uses an arrangement of infrared sensors that capture the hand gestures from the user, and then points towards a projection surface in the user workspace. A gesture recognition software platform recreates the 3D environment of the user and analyzes the motion of the key points of the user hands. After obtaining these data, a process of comparison with previously established patterns determines if the user is performing some kind of preset command with his hands. If so, the system immediately converts this signal into a command that will be executed with the assistance of an Arduino platform. The embedded platform carries out a previously established protocol, in which, by making use of a mechatronic system of 2 degrees of freedom that supports a micro projector, it allows the user to adjust the position of the projected image over a surface, allowing the user to use a 360° virtual space. The main goal of this system is to generate an interface where the gestures of the hands of the user not only allow him/her to interact with software-level elements, but also with mechatronic components that may be physically present such as robots or home automation
Automatic segmentation of the Epicardium and Endocardium plays an important role in the diagnosis of myocardial ischemia. Due to the speckle noise of ultrasound images and the complexity of cardiac tissue, the segment...
详细信息
Automatic segmentation of the Epicardium and Endocardium plays an important role in the diagnosis of myocardial ischemia. Due to the speckle noise of ultrasound images and the complexity of cardiac tissue, the segmentation is still manual or semi-automatic. A fully automatic segmentation method based on Convolutional Neural Network(CNN) is proposed in this paper: localization segmentation method, which can obtain the position of the Epicardium and Endocardium based on the output of the network, then the segmentation is completed according to reprocessing based on the position. We concentrate the solutions of the following problems to supervised learning models: 1) the need of training images which have obvious features;2) the design of network;3) correlation between group images. The performance of our approach is evaluated using four different datasets containing 800 ultrasound images of Sprague-Dawley rats. The quality grades of the manual and automatic segmentation methods have been calculated, and the proportion of poor segmentation quality is decreased from 3.48% to 2.39%, which shows that our proposed method has improved segmentation accuracy and stability. The efficiency of our fully automatic segmentation method is higher than traditional segmentation strategies. The proposed method can optimize the clinical procedure and decrease the variability of manual segmentations.
Optimizing connected component labeling is currently a very active research field. The current most effective algorithms although close in their design are based on different memory/computation trade-offs. This paper ...
详细信息
ISBN:
(纸本)9791092279061
Optimizing connected component labeling is currently a very active research field. The current most effective algorithms although close in their design are based on different memory/computation trade-offs. This paper presents a review of these algorithms and a detailed benchmark on several Intel and ARM embedded processors that allows to focus on their advantages and drawbacks and to highlight how processor architecture impact them.
We present in this paper Video++, a new framework targeting image and video applications running on multi-core processors. While offering a high expressive power, we show that it generates code running up to 32 times ...
详细信息
Chirp-sequence-based Frequency Modulation Continuous Wave (FMCW) radar is effective at detecting range and velocity of a target. However, the target detection algorithm is based on two-dimensional Fast Fourier Transfo...
详细信息
ISBN:
(纸本)9791092279061
Chirp-sequence-based Frequency Modulation Continuous Wave (FMCW) radar is effective at detecting range and velocity of a target. However, the target detection algorithm is based on two-dimensional Fast Fourier Transform, which uses a great deal of data over several PRIs (Pulse Repetition Intervals). In particular, if the multiple-receive channel is employed to estimate the angle position of a target;even more computational complexity is required. In this paper, we report on how a newly developed signalprocessing module is implemented in the FPGA, and on its performance measured under test conditions. Moreover, we have presented results from analysis of the use of hardware resources and processing times.
This paper presents the Parallel Heterogeneous Architecture Technology (PHAT), a scalable design methodology for prototyping and evaluating heterogeneous arrays of software-programmable VLIW processors and both manual...
详细信息
ISBN:
(纸本)9791092279061
This paper presents the Parallel Heterogeneous Architecture Technology (PHAT), a scalable design methodology for prototyping and evaluating heterogeneous arrays of software-programmable VLIW processors and both manually designed and automatically-compiled custom hardware accelerators, using a shared memory architecture for communication. We discuss the trade-offs and break-even point for switching from bus-based to network-on-chip interconnects, the interface and protocols for connecting distributed on-chip caches and multi-bank out-of-order offchip- memories, as well as the impact of floorplanning on the quality of results for implementation on Xilinx Virtex 6 LX 760 devices. The capabilities are evaluated at the system-level on the multi-FPGA Convey HC-1ex hybrid-core computer, accessing its high-performance memory system, and integrating r-VEX processor cores with IP blocks for SHA and FFT computations.
暂无评论