Atmospheric CO2 retrieval with a look-up-table based method by combining near and thermal infrared spectra

In order to precisely retrieve the atmospheric CO2 , a retrieval method based on both near infrared （NIR） and thermal infrared （TIR） is established firstly. Then a look-up-table （LUT） based fast line-by-line radiative transfer model （RTM） was integrated into the retrieval procedure to accelerate radiative transfer calculations. The LUT stores gas absorption cross-sections as a function of temperature, pressure and wavenumber. It could greatly reduce calculating time in radiative transfer compared to direct line-by-line method. Then retrieval was simulated using NIR, TIR and both bands. The retrieved CO2 profiles suggest joint approach could reconstruct CO2 profile better than those using NIR or TIR alone. Joint retrieval using both bands simultaneously could provide better constrain to CO2 vertical distribution in the whole troposphere.     

A novel algorithm for selective current limit used in AMOLED panel

A selective current limit algorithm based on color space conversion is proposed in this paper. There are mainly two advantages for this algorithm. Firstly, only the luminance component Y is decreased, and the color component Cb or Cr is not changed; thus, the color distortion is avoided in the process of current limit. Secondly, the current of pixels with high luminance is limited while that of those pixels with low luminance is not, which means current limit is executed selectively. For the same value of mean square error, the simulation results show that the ratio of pixels with high luminance of the image processed by the proposed algorithm is zero, and that processed by the net power control and the algorithm based on color space conversion is 0.081 and 0.0854, respectively. This algorithm is implemented on a field‐programmable gate array to drive the image display on the AMOLED panel.     

人脸检测的继承式集成学习方法

基于"遗传+变异"模式,提出继承式集成学习方法框架,它可以训练出四种不同形式的层叠分类器。除了基于"无遗传"模式的基本层叠分类器与基于"全部遗传"模式的嵌入式层叠分类器两种传统方法之外,还有基于"部分遗传+部分变异"模式的特征继承层叠分类器与弱分类器继承层叠分类器。虽然后两种层叠分类器都有一定的继承代价,但是其拟合性更好,可以更好地均衡收敛速度和扩展性能,其综合性能优于传统方法。基于RAB、GAB算法与LUT弱分类器的正面直立人脸检测实验结果表明了新的继承式集成学习方法的有效性。     

A Universal BIST Approach for Virtex-Ultrascale Architecture

Interconnected cells, Configurable Logic Blocks (CLBs), and input/output (I/O) pads are all present in every Field Programmable Gate Array (FPGA) structure. The interconnects are formed by the physical paths for connecting the blocks . The combinational and sequential circuits are used in the logic blocks to execute logical functions. The FPGA includes two different tests called interconnect testing and logical testing. Instead of using an additional circuitry, the Built-in-Self-Test (BIST) logic is coded into an FPGA, which is then reconfigured to perform its specific operation after the testing is completed. As a result, additional test circuits for the FPGA board are no longer required. The FPGA BIST has no area overhead or performance reduction issues like conventional BIST. A resource-efficient testing scheme is essential to assure the appropriate operation of FPGA look-up tables for effectively testing the functional operation. In this work, the Configurable Logic Blocks (CLBs) of virtex-ultrascale FPGAs are tested using a BIST with a simple architecture. To evaluate the CLBs’ capabilities including distributed modes of operation of Random Access Memory (RAM), several types of configurations are created. These setups have the ability to identify 100% stuck-at failures in every CLB. This method is suitable for all phases of FPGA testing and has no overhead or performance cost.     

基于Real Adaboost的肤色分割方法

提出了一种基于Real AdaBoost算法构造的肤色置信度分类器及动态阈值相结合的肤色分割方法。根据肤色在YCrCb色度空间的聚类性,通过大量肤色和非肤色样本用Real AdaBoost训练一族查找表(LUT)型圆形弱分类器,组成一个能输出连续置信度的强分类器,利用肤色强分类器计算图像中像素的肤色相似度,最后用大津法确定阈值对肤色相似图进行二值分割。实验表明,该方法能较好地描述肤色分布,误检率低,鲁棒性好。     

基于性能的LUT结构的FPGA的再设计算法

本文提出一个基于性能的LUT结构的FPGA的再设计算法,该算法采用特征函数以及对原布尔网络进行相应的约束实现电路的再设计。因为不改变网络的拓扑结构,从而避免了在再设计过程中重新考虑电路的时延和布局布线结果。     

Multiple-tapped-delay-line hardware-linearisation technique based on wire load regulation

This article describes designing, implementation and tuning processes of multiple-tapped-delay-line (MTDL). Obtained MTDL can be implemented in various field-programmable-logic-devices (FPGA) devices and applied for time-to-digital-converters (TDC) construction. The task of tuning process is the tapped-delay-line (TDL) linearisation, and consists of two stages. The first stage depends on selecting an appropriate configurable-logic-block (CLB) for particular delay-segment realization and selecting proper connection between these blocks. The second tuning stage, that is essential from this article viewpoint, depends on inter CLBs connecting wires delay regulation realized directly by load regulation. The Load regulation depends on connecting an appropriate number of unused three-state-buffers or CLB inputs to the wire which delay is adjusted. Depending on the number of inputs connected to the wire its capacitance changes that influences its time-constant and finally changes its time-delay.The MTDL mathematical model, obtained characteristics and results of time-interval (TI) measurements are also presented. The derived TDL model provides information about how the particular wire delay should be changed and in which order the changes should be executed. This makes the designing process predictable and easy to carry out. Presented characteristics confirm the proper operation of presented linearisation technique. The proper operation of the whole measuring module is confirmed by obtained TIs histograms presentation.     

一种基于视觉的反Gamma矫正方法

平板显示器件在灰度实现方式上与CRT存在较大差异.本文研究了SMPDP的发光特性和Gamma特性,提出了一种新的针对于SMPDP的反Gamma矫正方法,此方法综合考虑了SMPDP发光特性、环境光以及人眼视觉系统的影响,并在此基础上进行了优化.为了进一步提高SMPDP的视频图像质量,人眼系统对于亮度的阈值也被考虑到此方法中.此方法在SMPDP视频驱动电路中利用LUT作评估,实验结果表明本文提出的方法相对于传统方法能更好地提高视频图像质量.由于易于实现、精确度高,此方法也可用于其它的平板显示器件.     

基于分布式算法的FIR滤波器的设计与实现

本文介绍了能高效实现固定常数乘法的分布式算法原理,给出了在FPGA中用查找表实现FIR滤波器的算法设计,并以一个16阶低通滤波器为例说明了设计过程.该设计通过Altera公司的EPF10K30器件进行验证,结果正确,工作稳定可靠,满足了设计要求.