高性能并行比特变换运动估计硬件架构设计

为了满足便携式实时全高清视频的处理要求,该文基于1维源像素线性阵列,提出一种新的多宏块并行比特变换运动估计结构,克服以往2维阵列消耗资源较多且延时大的不足。该文结构易于并行扩展且更为节约资源,进一步还针对脉动胞元和数据存储器这两个系统瓶颈进行优化设计。FPGA实现结果表明,与同类设计相比,该文设计在面积和速度上均有改善,LUTs资源节约43%,DFFs资源节约25%,BRAMs数目节约75%,性能提升32%。     

Probability Constrained Search Range Determination for Fast Motion Estimation

In this paper, we propose new adaptive search range motion estimation methods where the search ranges are constrained by the probabilities of motion vector differences and a search point sampling technique is applied to the constrained search ranges. Our new methods are based on our previous work, in which the search ranges were analytically determined by the probabilities. Since the proposed adaptive search range motion estimation methods effectively restrict the search ranges instead of search point sampling patterns, they provide a very flexible and hardware‐friendly approach in motion estimation. The proposed methods were evaluated and tested with JM16.2 of the H.264/AVC video coding standard. Experiment results exhibit that with negligible degradation in PSNR, the proposed methods considerably reduce the computational complexity in comparison with the conventional methods. In particular, the combined method provides performance similar to that of the hybrid unsymmetrical‐cross multi‐hexagon‐grid search method and outstanding merits in hardware implementation.     

一种快速块运动估计的分级筛选方法

该文提出了一种新的快速块匹配运动估计方法,分级筛选法。该方法将搜索最佳匹配块的过程 分为若干个筛选级别,在初始的级别中用很少的运算代价通过简单的特征匹配先淘汰一部分候选块;然后在上一级剩余的候选块中,逐级用更加细致的特征继续筛选;直至找到最佳匹配块。实验结果表明,在估计精度非常相近的前提下,该文方法的速度是全搜索方法的12~14倍,而且该方法有很好的稳定性。     

One‐to‐All Broadcasting of Odd Networks for One‐Port and All‐Port Models

Odd networks were introduced in the context of graph theory. However, their potential as fault‐tolerant multiprocessor networks has been shown. Broadcasting is one of the most important communication primitives used in multiprocessor networks. In this letter, we introduce efficient one‐to‐all broadcasting schemes of odd networks for one‐port and all‐port models. We show the broadcasting time of the former is 2d?2 and that of the latter is d–1. The total time steps taken by the proposed algorithms are optimal.     

Serial and Parallel FPGA-based Variable Block Size Motion Estimation Processors

H.264/AVC is the latest video coding standard adopting variable block size motion estimation (VBS-ME), quarter-pixel accuracy, motion vector prediction and multi-reference frames for motion estimation. These new features result in much higher computation requirements than previous coding standards. In this paper we propose a novel most significant bit (MSB) first bit-serial architecture for full-search block matching VBS-ME, and compare it with systolic implementations. Since the nature of MSB-first processing enables early termination of the sum of absolute difference (SAD) calculation, the average hardware performance can be enhanced. Five different designs, one and two dimensional systolic and tree implementations along with bit-serial, are compared in terms of performance, pixel memory bandwidth, occupied area and power consumption.

A Low‐Complexity and High‐Quality Image Compression Method for Digital Cameras

This letter proposes a new near‐lossless image compression method with only one line buffer cost for a digital camera with Bayer format image. For such format data, it can provide a low average compression rate (4.24bits/pixel) with high‐image quality (larger than 46.37dB where the error of every pixel is less than two). The experimental results show that the near‐lossless compression method has better performance than JPEG‐LS (lossless) with δ = 2 for a Bayer format image.     

Fast Weighted Bit Flipping Algorithm for Higher-Speed Decoding of Low-Density Parity-Check Codes

Because of the speed limitation of the conventional bit-selection strategy in the exi- sting weighted bit flipping algorithms, a high- speed Low-Density Parity-Check （LDPC） dec- oder cannot be realised. To solve this problem, we propose a fast weighted bit flipping algo- rithm. Specifically, based on the identically dis- tributed error bits, a parallel bit-selection met- hod is proposed to reduce the selection delay of the flipped bits. The delay analysis demon- strates that, the decoding speed of LDPC codes can be significantly improved by the proposed algorithm. Furthermore, simulation results ver- ify the validity of the proposed algorithm.     

相关信源的随机加权信号子空间估计算法

针对空间相关信号源使得阵列的协方差矩阵的有效秩降低的问题,引入随机加权子阵,打破了信号的相关性,得到一种新相关信源的信号子空间的估计算法。仿真实验表明该算法可以有效地去相关,与空间平滑法相比具有更好的估计性能。     

一种基于小波域的运动估计方法

提出了一种基于多尺度边缘检测(MED)与连续小波变换(CWT)相结合的运动目标估计方法。首先,利用多尺度边缘检测算法,得到抗噪性好、边缘连续清晰的运动目标边缘图像。然后,利用连续小波变换把目标序列图像映射到运动状态参数空间。该方法同时拥有MED和CWT的优良特性。实验表明,该算法是对旋转、遮挡和有噪等复杂运动目标进行运动估计的有效方法。     

Microparticle‐Based Soft Electronic Devices: Toward One‐Particle/One‐Pixel

While microparticle (MP) assemblies have long attracted academic interest, few practical applications of assembled MPs have been achieved because of technological difficulties related to MP synthesis, MP position registration, and the absence of device concepts. The precise positioning of functional MPs in a proper stencil can produce flexible/stretchable electronic devices, even when the MPs themselves are rigid. In recent years, remarkable progress has been made in the programmable position registration of MPs, production of functional MPs, and concepts for MP‐based, pixel‐type electronic devices. This progress report reviews the recent technological advances in MP assembly and discusses the technological challenges preventing the realization of the one‐particle/one‐pixel concept.     

运动估计块匹配算法的分析研究

在对目前运动估计快速块匹配算法研究的基础上，描述了运动估计的基本原理；揭示了提高运动估计效率的关键技术，并对相关的算法进行了分析和比较；提出了运动估计算法今后的研究方向。     

基于离散FRFT的SAR动目标参数估计方法

为了有效减少SAR系统的数据量和降低脉冲重复频率，提出了一种基于离散分数阶傅里叶变换（FRFT）的动目标参数估计方法。首先对距离向脉冲压缩之后的数据进行DPCA对消处理，其次利用Hough变换估计目标跨航向维的运动参数并校正距离单元走动，然后基于离散的FRFT变换矩阵构造稀疏基矩阵，建立压缩感知重构模型，通过对模型的优化求解获得FRFT的最优阶数，进而估计动目标沿航迹维的速度和位置。最后，仿真实验验证了所提方法能够有效实现低脉冲重复频率条件下的地面运动目标参数估计。     

Low‐Complexity Multi‐size Cyclic‐Shifter for QC‐LDPC Codes

The decoding process of a quasi‐cyclic low‐density parity check code requires a unique type of rotator. These rotators, called multi‐size cyclic‐shifters (MSCSs), rotate input data with various sizes, where the size is the amount of data to be rotated. This paper proposes a low‐complexity MSCS structure for the case when the sizes have a nontrivial common divisor. By combining the strong points of two previous structures, the proposed structure achieves the smallest area. The experimental results show that the area reduction was more than 14.7% when the proposed structure was applied to IEEE 802.16e as an example.     

Fine‐Motion Estimation Using Ego/Exo‐Cameras

Robust motion estimation for human–computer interactions played an important role in a novel method of interaction with electronic devices. Existing pose estimation using a monocular camera employs either ego‐motion or exo‐motion, both of which are not sufficiently accurate for estimating fine motion due to the motion ambiguity of rotation and translation. This paper presents a hybrid vision‐based pose estimation method for fine‐motion estimation that is specifically capable of extracting human body motion accurately. The method uses an ego‐camera attached to a point of interest and exo‐cameras located in the immediate surroundings of the point of interest. The exo‐cameras can easily track the exact position of the point of interest by triangulation. Once the position is given, the ego‐camera can accurately obtain the point of interest's orientation. In this way, any ambiguity between rotation and translation is eliminated and the exact motion of a target point (that is, ego‐camera) can then be obtained. The proposed method is expected to provide a practical solution for robustly estimating fine motion in a non‐contact manner, such as in interactive games that are designed for special purposes (for example, remote rehabilitation care systems).     

一种运动估计的快速搜索方法

本文提出了动态图像压缩中估计运动矢量的一种快速搜索算法。分析与实验表明，该算法的计算量大大低于二维全搜索，而匹配效果优于三步法等快速算法，目硬件实现简单。它适用于动态图像的实时编码。     

一种用于视频编码的块运动估计算法——块特征匹配预测搜索算法

本文提出了一种块特征匹配预测搜索ＢＦＭＰＳ算法,可以用于视频压缩的一些国际标准,如Ｈ．２６１,Ｈ．２６３,ＭＰＥＧ１,ＭＰＥＧ２,ＨＤＴＶ中,ＢＦＭＰＳ算法充分利用了序列图像的实际矢量与预测矢量之间距离的空间分布特征,中心偏置分布特性和时间上的相关特性,并采用了中止判决准则,可以明显地减少了运动搜索复杂度,ＢＦＭＰＳ算法在匹配运算中采用 简单有交的块特征匹配准则函数,相应的块匹配计算复杂度,数据读     

A Low‐Complexity 128‐Point Mixed‐Radix FFT Processor for MB‐OFDM UWB Systems

In this paper, we present a fast Fourier transform (FFT) processor with four parallel data paths for multiband orthogonal frequency‐division multiplexing ultra‐wideband systems. The proposed 128‐point FFT processor employs both a modified radix‐2⁴ algorithm and a radix‐2³ algorithm to significantly reduce the numbers of complex constant multipliers and complex booth multipliers. It also employs substructure‐sharing multiplication units instead of constant multipliers to efficiently conduct multiplication operations with only addition and shift operations. The proposed FFT processor is implemented and tested using 0.18 µm CMOS technology with a supply voltage of 1.8 V. The hardware‐ efficient 128‐point FFT processor with four data streams can support a data processing rate of up to 1 Gsample/s while consuming 112 mW. The implementation results show that the proposed 128‐point mixed‐radix FFT architecture significantly reduces the hardware cost and power consumption in comparison to existing 128‐point FFT architectures.     

Low‐Complexity Symbol Timing Offset Estimation Schemes for OFDM Systems

In this paper, we propose three symbol synchronization schemes for Orthogonal Frequency Division Multiplex (OFDM) systems. The cyclic extension preceding OFDM symbols is of decisive importance for these schemes. The first scheme uses the phase‐differential coding of the received OFDM signal. The second and the third schemes use the length of the received OFDM signal. All three schemes make symbol synchronization possible, even though there is a frequency offset in the system. Simulation results show that these schemes can be used to synchronize an OFDM system over AWGN and multi‐path fading channels.     

An SAD‐Based Selective Bi‐prediction Method for Fast Motion Estimation in High Efficiency Video Coding

As the next‐generation video coding standard, High Efficiency Video Coding (HEVC) has adopted advanced coding tools despite the increase in computational complexity. In this paper, we propose a selective bi‐prediction method to reduce the encoding complexity of HEVC. The proposed method evaluates the statistical property of the sum of absolute differences in the motion estimation process and determines whether bi‐prediction is performed. A performance comparison of the complexity reduction is provided to show the effectiveness of the proposed method compared to the HEVC test model version 4.0. On average, 50% of the bi‐prediction time can be reduced by the proposed method, while maintaining a negligible bit increment and a minimal loss of image quality.     

结合Hough变换与运动估计的车道线提取方法

车道线识别是安全辅助驾驶和智能驾驶系统的核心研究内容,对控制危险驾驶和疲劳驾驶均有显著的作用,通常利用Hough变换对直线检测的容错性和鲁棒性,可以对车载摄像头拍摄到的车道线进行有效的检出.巧妙地将分块Hough变换和图像块的运动估计相结合,极大地降低了车道线检测和跟踪的算法复杂度,实现了车道线的实时识别与跟踪.实验表明,采用该方法既可以得到稳定的检测结果,又能提高检测的速度,保留了Hough变换的容错性和鲁棒性.