Purpose-The real-time generation of feature descriptors for object recognition is a challenging problem.In this research,the purpose of this paper is to provide a hardware friendly framework to generate sparse feature...Purpose-The real-time generation of feature descriptors for object recognition is a challenging problem.In this research,the purpose of this paper is to provide a hardware friendly framework to generate sparse features that can be useful for key feature point selection,feature extraction,and descriptor construction.The inspiration is drawn from feature formation processes of the human brain,taking into account the sparse,modular,and hierarchical processing of visual information.Design/methodology/approach-A sparse set of neurons referred as active neurons determines the feature points necessary for high-level vision applications such as object recognition.A psycho-physical mechanism of human low-level vision relates edge detection to noticeable local spatial stimuli,representing this set of active neurons.A cognitive memory cell array-based implementation of low-level vision is proposed.Applications of memory cell in edge detection are used for realizing human vision inspired feature selection and leading to feature vector construction for high-level vision applications.Findings-True parallel architecture and faster response of cognitive circuits avoid time costly and redundant feature extraction steps.Validation of proposed feature vector toward high-level computer vision applications is demonstrated using standard object recognition databases.The comparison against existing state-of-the-art object recognition features and methods shows an accuracy of 97,95,69 percent for Columbia Object Image Library-100,ALOI,and PASCAL VOC 2007 databases indicating an increase from benchmark methods by 5,3 and 10 percent,respectively.Originality/value-A hardware friendly low-level sparse edge feature processing system isproposed for recognizing objects.The edge features are developed based on threshold logic of neurons,and the sparse selection of the features applies a modular and hierarchical processing inspired from the human neural system.展开更多
Autonomous lane keeping is an important technology in intelligent transportation,which is used to avoid unnecessary traffic accidents caused by lane departure.To adapt different lighting environment,and make up ordina...Autonomous lane keeping is an important technology in intelligent transportation,which is used to avoid unnecessary traffic accidents caused by lane departure.To adapt different lighting environment,and make up ordinary Hough transform’s shortcomings of tardiness and poor immunity,we propose an improved algorithm by using adaptive gray threshold and setting Region of interest(ROI),to do the quick Hough transform for tracking lane line,implementing autonomous lane keeping. The dynamic adaptive threshold method can be suitable with different lighting conditions and quickly,accurately remove most of the information not relative to lane line.Meanwhile setting ROI can let the program only care about the specific region which can provide useful information and further reduce the processing data.And then on the basic of identification,we put forward some efficient innovation strategy about the control logic of straight state,curve state and the transition state.The experiment proves that this solution greatly raises efficiency.展开更多
Proton-rich nuclei are synthesized via photodisintegration and reverse reactions.To examine this mechanism and reproduce the observed p-nucleus abundances,it is crucial to know the reaction rates and thereby the react...Proton-rich nuclei are synthesized via photodisintegration and reverse reactions.To examine this mechanism and reproduce the observed p-nucleus abundances,it is crucial to know the reaction rates and thereby the reaction cross sections of many isotopes.Given that the number of experiments on the reactions in astrophysical energy regions is very rare,the reaction cross sections are determined by theoretical methods whose accuracy should be tested.In this study,given that ^(121)Sb is a stable seed isotope located in the region of medium-mass p-nuclei,we investigated the cross sections and reaction rates of the ^(121)Sb(α,γ)^(125)I reaction using the TALYS computer code with 432 different combinations of input parameters(OMP,LDM,and SFM).The optimal model combinations were determined using the threshold logic unit method.The theoretical reaction cross-sectional results were compared with the experimental results reported in the literature.The reaction rates were determined using the two input parameter sets most compatible with the measurements,and they were compared with the reaction rate databases:STARLIB and REACLIB.展开更多
This paper presents a novel approach to design robust Source Coupled Logic (SCL) for implementing ultra low power circuits. In this paper, we propose two different source coupled logic structures and analyze the perfo...This paper presents a novel approach to design robust Source Coupled Logic (SCL) for implementing ultra low power circuits. In this paper, we propose two different source coupled logic structures and analyze the performance of these structures with STSCL (Sub-threshold SCL). The first design under consideration is DTPMOS as load device which analyses the performance of Dynamic Threshold SCL (DTSCL) Logic with previous source coupled logic for ultra low power operation. DTSCL circuits exhibit a better power-delay Performance compared with the STSCL Logic. It can be seen that the proposed circuit provides 56% reduction in power delay product. The second design under consideration uses basic current mirror active load device to provide required voltage swing. Current mirror source coupled logic (CMSCL) can be used for high speed operation. The advantage of this design is that it provides 54% reduction in power delay product over conventional STSCL. The main drawback of this design is that it provides a higher power dissipation compared to other source coupled logic structures. The proposed circuit provides lower sensitivity to temperature and power supply variation, with a superior control over power dissipation. Measurements of test structures simulated in 0.18 μm CMOS technology shows that the proposed DTSCL logic concept can be utilized successfully for bias currents as low as 1 pA. Measurements show that existing standard cell libraries offer a good solution for ultra low power SCL circuits. Cadence Virtuoso schematic editor and Spectre Simulation tools have been used.展开更多
云中过冷水识别对于人工影响天气及预防飞机积冰具有重要意义,但过冷水的识别一直是气象探测中的难点,毫米波雷达是连续探测云结构和物理特征的有效工具。利用布设在藏东南水汽通道入口处墨脱地区的Ka波段毫米波云雷达基数据,结合微波...云中过冷水识别对于人工影响天气及预防飞机积冰具有重要意义,但过冷水的识别一直是气象探测中的难点,毫米波雷达是连续探测云结构和物理特征的有效工具。利用布设在藏东南水汽通道入口处墨脱地区的Ka波段毫米波云雷达基数据,结合微波辐射计温度资料,采用基于模糊逻辑法、阈值法进行过冷水识别,识别出的粒子相态包含冰、雪、过冷水及混合态。并利用同址的微波辐射计的液态水路径(liquid water paths,LWP)对墨脱云雷达观测的两个层积云过程的过冷水识别效果进行了分析和初步验证。结果表明:模糊逻辑法和阈值法识别的过冷水基本合理,但模糊逻辑法可以识别更多的过冷水,从定量分析来看,模糊逻辑法相对于阈值法识别的LWP更接近于微波辐射计。藏东南墨脱地区层积云中过冷水的微物理参数与其他地区较为一致,有效半径主要位于7~15μm,液态水含量(liquid water content,LWC)主要分布在0.01~0.3 g/m^(3),但墨脱地区过冷水的分布比其他地区更为丰富,往往云顶、云底及云中同时存在过冷水。展开更多
文摘Purpose-The real-time generation of feature descriptors for object recognition is a challenging problem.In this research,the purpose of this paper is to provide a hardware friendly framework to generate sparse features that can be useful for key feature point selection,feature extraction,and descriptor construction.The inspiration is drawn from feature formation processes of the human brain,taking into account the sparse,modular,and hierarchical processing of visual information.Design/methodology/approach-A sparse set of neurons referred as active neurons determines the feature points necessary for high-level vision applications such as object recognition.A psycho-physical mechanism of human low-level vision relates edge detection to noticeable local spatial stimuli,representing this set of active neurons.A cognitive memory cell array-based implementation of low-level vision is proposed.Applications of memory cell in edge detection are used for realizing human vision inspired feature selection and leading to feature vector construction for high-level vision applications.Findings-True parallel architecture and faster response of cognitive circuits avoid time costly and redundant feature extraction steps.Validation of proposed feature vector toward high-level computer vision applications is demonstrated using standard object recognition databases.The comparison against existing state-of-the-art object recognition features and methods shows an accuracy of 97,95,69 percent for Columbia Object Image Library-100,ALOI,and PASCAL VOC 2007 databases indicating an increase from benchmark methods by 5,3 and 10 percent,respectively.Originality/value-A hardware friendly low-level sparse edge feature processing system isproposed for recognizing objects.The edge features are developed based on threshold logic of neurons,and the sparse selection of the features applies a modular and hierarchical processing inspired from the human neural system.
文摘Autonomous lane keeping is an important technology in intelligent transportation,which is used to avoid unnecessary traffic accidents caused by lane departure.To adapt different lighting environment,and make up ordinary Hough transform’s shortcomings of tardiness and poor immunity,we propose an improved algorithm by using adaptive gray threshold and setting Region of interest(ROI),to do the quick Hough transform for tracking lane line,implementing autonomous lane keeping. The dynamic adaptive threshold method can be suitable with different lighting conditions and quickly,accurately remove most of the information not relative to lane line.Meanwhile setting ROI can let the program only care about the specific region which can provide useful information and further reduce the processing data.And then on the basic of identification,we put forward some efficient innovation strategy about the control logic of straight state,curve state and the transition state.The experiment proves that this solution greatly raises efficiency.
文摘Proton-rich nuclei are synthesized via photodisintegration and reverse reactions.To examine this mechanism and reproduce the observed p-nucleus abundances,it is crucial to know the reaction rates and thereby the reaction cross sections of many isotopes.Given that the number of experiments on the reactions in astrophysical energy regions is very rare,the reaction cross sections are determined by theoretical methods whose accuracy should be tested.In this study,given that ^(121)Sb is a stable seed isotope located in the region of medium-mass p-nuclei,we investigated the cross sections and reaction rates of the ^(121)Sb(α,γ)^(125)I reaction using the TALYS computer code with 432 different combinations of input parameters(OMP,LDM,and SFM).The optimal model combinations were determined using the threshold logic unit method.The theoretical reaction cross-sectional results were compared with the experimental results reported in the literature.The reaction rates were determined using the two input parameter sets most compatible with the measurements,and they were compared with the reaction rate databases:STARLIB and REACLIB.
文摘This paper presents a novel approach to design robust Source Coupled Logic (SCL) for implementing ultra low power circuits. In this paper, we propose two different source coupled logic structures and analyze the performance of these structures with STSCL (Sub-threshold SCL). The first design under consideration is DTPMOS as load device which analyses the performance of Dynamic Threshold SCL (DTSCL) Logic with previous source coupled logic for ultra low power operation. DTSCL circuits exhibit a better power-delay Performance compared with the STSCL Logic. It can be seen that the proposed circuit provides 56% reduction in power delay product. The second design under consideration uses basic current mirror active load device to provide required voltage swing. Current mirror source coupled logic (CMSCL) can be used for high speed operation. The advantage of this design is that it provides 54% reduction in power delay product over conventional STSCL. The main drawback of this design is that it provides a higher power dissipation compared to other source coupled logic structures. The proposed circuit provides lower sensitivity to temperature and power supply variation, with a superior control over power dissipation. Measurements of test structures simulated in 0.18 μm CMOS technology shows that the proposed DTSCL logic concept can be utilized successfully for bias currents as low as 1 pA. Measurements show that existing standard cell libraries offer a good solution for ultra low power SCL circuits. Cadence Virtuoso schematic editor and Spectre Simulation tools have been used.
文摘云中过冷水识别对于人工影响天气及预防飞机积冰具有重要意义,但过冷水的识别一直是气象探测中的难点,毫米波雷达是连续探测云结构和物理特征的有效工具。利用布设在藏东南水汽通道入口处墨脱地区的Ka波段毫米波云雷达基数据,结合微波辐射计温度资料,采用基于模糊逻辑法、阈值法进行过冷水识别,识别出的粒子相态包含冰、雪、过冷水及混合态。并利用同址的微波辐射计的液态水路径(liquid water paths,LWP)对墨脱云雷达观测的两个层积云过程的过冷水识别效果进行了分析和初步验证。结果表明:模糊逻辑法和阈值法识别的过冷水基本合理,但模糊逻辑法可以识别更多的过冷水,从定量分析来看,模糊逻辑法相对于阈值法识别的LWP更接近于微波辐射计。藏东南墨脱地区层积云中过冷水的微物理参数与其他地区较为一致,有效半径主要位于7~15μm,液态水含量(liquid water content,LWC)主要分布在0.01~0.3 g/m^(3),但墨脱地区过冷水的分布比其他地区更为丰富,往往云顶、云底及云中同时存在过冷水。
