基于YCbCr颜色空间和Fisher判别分析的棉花图像分割研究

棉花的分割是采棉机器人研究的关键技术。本文分别在HSV、HIS和YCbCr颜色空间下,首先根据棉花的颜色信息与背景颜色信息的差距,对样本图像中的各个对象(棉絮、棉枝、土壤等)分类; 其次根据分类结果分别提取各类在各颜色空间下的样本像素值; 再根据类间离散度最大和类内离散度最小的准则计算出Fisher判别向量和各类的质心; 最后按照像素值离各质心最近的准则进行图像分割。结果表明, 在YCbCr颜色空间下产生的分割噪声最小,选取此颜色空间,采用贴标签的方法自适应去噪。实验仿真表明,本方法可有效避免阳光直射和阴影的干扰,对各种情况都能准确分割,分割准确率达90.44%。     

利用频域有限元法分析内嵌管式辐射地板频域热特性

分析内嵌管式辐射地板的频域热特性,便于进一步了解内嵌管式辐射地板的动态热特征,为辐射地板的系统和控制设计提供重要的参考依据。建立了内嵌管式辐射地板的频域有限元模型,同时采用ANSYS软件建立了内嵌管式辐射地板的时域传热模型,通过2个模型的对比说明了频域有限元模型的准确性。采用频域有限元模型计算并分析了内嵌管式辐射地板的频域热特性。结果表明,在低频区域,地板的频域热响应基本不随频率变化,其传热过程接近于稳态,而在高频区域,地板的频域热响应随频率的变化十分剧烈,地板传热呈现明显的动态特征。当房间内存在较多成分的高频热扰时,关于辐射地板的传热计算应采用动态计算方法。     

抗PVY马铃薯品种遗传多样性的RAPD分析

采用RAPD技术分析了国内外37种抗PVY马铃薯资源。提取马铃薯(Solanumtuberosum)新鲜叶片的总DNA,用17种RAPD引物进行随机多态性扩增,利用遗传多样性信息,进行亲源关系的分子鉴定。试验表明:平均每10个碱基引物扩增出6￣21条谱带,共获得164条特异性谱带,平均每个引物扩增获得9.8条多态性谱带,多态性比率平均为76.3%。RAPD分析表明37种抗PVY马铃薯资源之间的遗传距离介于0.06￣0.68之间,平均值为0.35,平均遗传距离介于0.27￣0.50之间,聚类分析将37种抗PVY材料划分为三个类,聚类结果同材料的血缘关系密切,并且表现出一定的地域特性。根据扩增的特异性谱带可进行抗PVY马铃薯资源的分子鉴定,为抗PVY育种亲本选配提供了依据。     

基于RAPD标记的芥蓝种质资源遗传多样性分析

调查了44份芥蓝种质的植物学性状,并利用RAPD分子标记分析了其遗传多样性。结果从150条随机引物中筛选出20个引物,20个引物共扩增出177条谱带,其中多态谱带105条,平均每个引物扩增出8.9条谱带和5.3条多态性谱,多态性比率为59.32%。基于RAPD标记,利用NTSYS-pc2.11构建了聚类树状图谱,遗传距离为0.70时,44份芥蓝资源可聚成六大类群。芥蓝种质存在着一定的遗传多样性,但原产华南而且主要产区也在华南,遗传多样性要小于芸薹属其他蔬菜。     

OTSU最佳阈值法在棉花幼苗识别中的应用研究

 棉花幼苗的识别与分离是实现自动化移栽的关键步骤,本文用视觉图像处理技术对棉花幼苗的识别进行了研究。对于用摄像头获取的棉花幼苗图像,在分析棉花幼苗及背景颜色的基础上,采用了OTSU法将表征在RGB颜色空间内的彩色图像转化为灰度图像,并使用非线性滤波法对灰度图像进行去噪处理,通过最大类间方差法选取了最佳分割阈值,将灰度图像转化为二值化图像。为了更好地提取并识别棉花幼苗特征,又细化处理了二值化图像,设计了去骨刺处理,消除了细化后图像骨刺对特征识别的影响,并且使用了加权模板匹配法,设计了防误判匹配法则。应用这些方法,实现了棉花幼苗的识别,说明用图像处理方法识别棉花幼苗是完全可行的。     

基于颜色阈值的田间籽棉图像分割技术

为正确分割田间籽棉图像,将棉花与背景视为二个类别,在典型的未成熟籽棉图像和不同质量等级的成熟/过熟籽棉图像中,用肉眼选取20 000个白棉、黄染棉和污染棉等棉花像素以及20 000个棉株、土壤等背景像素,在RGB、HSI、La*b*和Hunter颜色空间下获取二类像素之间的颜色阈值,基于阈值进行图像分割,选取噪声较少的HSI和La*b*颜色空间,进一步基于形态学滤波器去噪,实验结果表明,907幅籽棉图像分割的准确率为87.21%和86.33%。HSI颜色空间更适合分割成熟籽棉图像,La*b*颜色空间则适合未成熟籽棉;颜色阈值覆盖范围广,基于速度的阈值分割法能够适应田间籽棉环境。     

基于OpenCV的Android手机植物叶片几何参数测量系统

通过Android 手机对植物叶片图像进行处理以测量其长、宽、周长、面积等几何参数,避免其他测量方法的缺点,提高数据采集效率。首先,获取含有已知尺寸矩形框的叶片图像。然后,对图像进行校正以消除倾斜失真。在校正阶段,采用了更为鲁棒的特征点检测法代替可能出现错误结果的Hough 变换法来处理矩形框,得到其4 个角点。根据求得的角点与标准图像的角点对图像进行映射得到校正图像。最后,处理校正图像计算叶片几何参数。参数计算提出以叶片轮廓为基础的方法,所有参数均通过对轮廓的处理求出,提高了计算效率。此外,基于轮廓求面积不需要对叶片进行形态学处理消除空隙及不连续,还可避免统计像素方法中将噪声像素误认为叶片造成的计算不精确。在求叶片长、宽时,给出了2 种方法供用户根据叶片实际形状选择。另外,在编程方面,采用了OpenCV-Android-SDK,大幅提高了图像处理速度。结果表明,精确度测试中,对打印出的规则图形几何参数测量的结果误差均在2%以内;而实际叶片的测量结果中长、宽的误差在2%以内,周长误差小于4%,面积误差低于3%。耗时测试中,相同图像处理算法采用OpenCV后,处理速度明显提升;对多部不同配置手机的测试中,处理1 幅分辨率为2448×3264 的叶片图像的耗时均在3 s 以内。另外,对于同一叶片图像,不同手机测量的结果完全一致。实验表明,该系统不仅操作简单、速度快、通用性强,而且对叶片几何参数的测量精确度也较高。     

贵州桂花种质资源遗传多样性的ISSR分析

通过ISSR分子标记,对贵州不同地区的54份桂花种质进行了遗传多样性研究.选取11条多态性引物用于正式扩增,获得清晰、重复性好的DNA谱带93条,多态位性谱带80条,占总扩增带的86％,平均每个引物扩增8.45条;应用810和830的引物组合,构建了桂花种质的ISSR指纹图谱,可以有效区分所有供试材料;利用NTSYS和POPGENE 32进行数据分析,计算出相似系数与遗传距离,构建了54个样本的聚类树状图;以遗传相似系数0.69为阈值把54份供试桂花材料分为4类,聚类结果与地域及海拔具有一定相关性.因此,ISSR技术能够有效地用于桂花种质资源的鉴别及遗传关系的解析.     

晋旱125×（昔野×501）大豆群体同工酶研究及聚类分析

选用晋旱125×(昔野×501)大豆杂交组合的后代群体选育的较稳定的四个有代表性的特殊类型为实验材料,对双亲及子代不同类型品种的POD酶和酯酶进行电泳分析,发现POD酶和酯酶谱存在明显差异,2种同工酶的酶带数分别为21和24条。同时采用DPS数据分析软件对这两类酶带谱进行聚类分析。最终从聚类树状图上根据遗传距离将亲本与子代清晰区别开来,结果显示:在POD酶和酯酶的聚类图上6个品种均被聚为两类,子代在遗传距离上偏向于母本晋旱125。     

中国农产品期货价格的分形和多重分形特征

为了探索中国农产品期货价格中存在的非线性特征,根据大连豆粕等农产品期货价格的时间序列数据,应用R/S分析方法,分析了上述价格系统中存在的分形特征,得到了不同时间标度下农产品期货价格的Hurst指数,从而发现了系统对信息的长期记忆性;跟踪了不同时延下Hurst指数的演化轨迹,并根据系统不同的动力学行为将其划分为不同阶段;引入多重分形分析研究了该分形系统中分形结构的不规则性。研究结果显示：在豆粕期货价格的短期收益率接近随机游走,系统中含有较强的噪声干扰,价格行为难以预测;豆粕期货价格的中长期收益率呈现出正向趋势性,具有对历史信息的长期记忆;系统中的噪声干扰较小,具有更强的稳定性。长期收益率具有高稳定性特征,反应出市场供需的基本面特征;豆粕期货价格分形系统均具有非平凡的多重分形特征。     

Test Method of Indoor Noise for Micro-surfacing Mix

Micro-surfacing mix has been wildly used in the pavement maintenance for its water proofing, anti-skidding and wear resistance. However, micro-surfacing pavement has higher noise than the common asphalt pavement and its noise can not be measured effectively and accurately. Therefore, the indoor noise test method of micro-surfacing mix based on wet track abrasion test was explored. The micro-surfacing mix specimens were used to imitate the micro-surfacing pavement. The wet track abrasion tester was used to provide the vehicle loads. The sound level meter was used to measure the noise between tire and pavement. Furthermore, the indoor noise of nine kinds of micro-surfacing mixes with different gradation was examined. The results show that the equivalent sound level in the second time period has good relationship with the surface texture depth by sad patch method. Then the test method and test procedure for the indoor noise of micro-surfacing mix based on wet track abrasion test were given. Finally, two micro-surfacing mixtures with different gradations were made to validate the test method. The results show that the evaluation method can effectively distinguish the difference of micro-surfacing mixes with different gradation and can supplement the mix design method of micro-surfacing mixes.     

Filtering Method for the Decomposition of Heart Rate Variability Power Spectral

Heart rate variability is a convenient tool to study the autonomic nervous system of human body. It is a focus of studying cardiovascular disease in clinic. The main spectral components are distinguished in the heart rate variability spectrum: ultra low frequency(ULF),very low frequency(VLF), low frequency(LF), high frequency(HF) components .While different part has different physiological meaning. This paper discusses the design of an effective digital filter for heart rate variability power spectral analysis,which provides a scientific measure of heart rate variability power spectral analysis in further studies and clinical applications.     

Application of digital phase-locked loop and filtering technique in PWM rectifier

VHS-ADC high speed digital signal processing platform based on MATLAB and FPGA could be used to build three-phase voltage source rectifier with space voltage vector PWM(SVPWM).The initial phase of coordinate transformation from three phase static coordinate to two phase static coordinate which is called Park transformation and from two phase rotating reference frame to two phase static coordinate is uncertain. There existed noise in acquiring grid voltage and current. Digital phase-locked loop is added to follow the frequency and phase of grid voltage, FIR filter is added to process signal and reduce the interference. External voltage control loop, internal current control loop and the model of coordinate transformation are designed on the VHS-ADC platform. The small power experiment is conducted to verify the stability and feasibility of the control system. The experimental results show that the digital phase-locked loop and FIR digital filter can be applied to three-phase voltage source rectifier.     

The study on EMI principle and suppression method of vehicle ignition system

Vehicle ignition system is the main electromagnetic interference source. Electromagnetic radiation produced by ignition system will produce the interference to electromagnetic environment around. It is necessary to study the electromagnetic interference (EMI) principle, noise paths and suppression method. We analyze the electromagnetic interference principle of ignition system, and harness trouble-shooting related to electromagnetic interference is also tested. Results show that the spark discharge between spark plug electrodes is the main interference source in ignition system. Then, the paths of conducted interference and the reason of radiated emission are discussed in detail. Finally, two kinds of power supply cord EMI filters are designed and implemented on the basis of analysis results. Experimental results indicate that the filter can reduce the noise of power-wire in broadband range. Simultaneously, the radiation emission performance from power wire is also reduced observably. The proposed mixing filter has been demonstrated in this paper to perform satisfactorily with excellent effectiveness in suppression of conducted EMI from 10 MHz to 110 MHz, and the object car can also pass the EMC standard tests.     

Adaptive Enhancementt of Heart Sound Based on Simulink

In the course of recording heart sound, it is inevitable that many kinds of noise will be merged in the main signal. These kinds of noise disbennifit the heart sounds, even import awful influence to analysis result. So, before further processings of phonocardiographic records, noise must be suppressed. Because heart-sound is a highly nonstationary signal, this paper presents the application of adaptive filter method to the elimination of noise,and at the same time, employs the datum of heart-sound plot as the reference noise. With simulink, a modeling and simulation tool for dynamical systems, the model of LMS adaptive heart-sound enhancement is established.Tests prove this method makes effective denoising and heart-sound enhancement.     

THE RESEARCH OF MONITORING AND DIAGNOSING BALL BEARING FATIQUE ON LINE

This paper describes the research work of analyzing vibration signal to monitor and diagnose on line the ball bearing fatique. Carrying out a large mumber of life test on ball bearing test machine, it is discovered that in the so called "second effect" vibration signal as the bearing fatique occurs, apart from the two widely used spetrum charactersof "fraverse frequency" and "high frequency shock pulse" , there is another important but rarely known opcctrum characteristic of high frequency peaks existing in the middle-to-high frequency band.The generating mechanism, effect factors and test method of those high frequency peaks are researched tentativ tentatively. Two binds of device for monitoring abnd diagnos sing ball learing fatique on line are developed based ion this new type spectrum characteristic. One consists of a bandpass filter, the other is based on the spetrum characters which corns from the DDS ( Dynamic Data System ) modeling     

Filtering Construct of Adaptive Filter Grou Pwith Multi-scale Decomposition

The principle and method of the adaptive filter and the filtering with wavelet transform were analyzed, and the model and method of adaptive filtering with wavelet transforms for the transient signal was established. The separated noise of signal by the multi-scale decomposition of wavelet transforms, was the input signal of adaptive filter, and accordingly the optimal filtering method of signal-noise decomposition was realized. By the adaptive filter grou Pbased on the wavelet transform, the optimal filtering to the multi-noise of signal is achieved at the same time, and the method presented in this paper has the excellent filtering capability. Examples of application demonstrate that this method presented is excellent to realize the optimal estimate to the valuable signal and noise of the transient signal in the same frequency segment.     

EEG Signal Denoising Based on Wavelet Transform

The article focuses on the method of noise cancellation for EEG signal. The method of notch filter is discussed. According to the frequency of noise and the principle of notch filter, the design result of the notch filter and the denoised signal are presented. Then, the analysis of EEG signal are proposed based on wavelet transform (WT) and noise cancellation using WT. Wavelet transform is a multi-resolution time-frequency analysis method. It can decompose mixed signal into signals at different frequency bands. The EEG signal is analyzed and denoised using WT, then the results are presented respectively. Comparing the experiment results shows that WT can detect and process noise in the EEG signal effectively.     

具特异性状人工合成小麦的条锈病抗性改良

为了解3份具特异优良性状但高感条锈病的人工合成小麦SHW-Z1、SHW-Z2和SHW-Z4感病性的遗传特点,进行更好的育种利用。用高抗条锈病的普通小麦材料5157与上述人工合成小麦分别进行正反杂交,对6个杂交组合的亲本、F¹世代的条锈病抗性与F²代的条锈病抗感分离情况进行了分析以探究其感病性的遗传特点,结果表明：（1）本研究的普通小麦和人工合成小麦杂交后代的条锈病抗性由多对基因控制,遗传上表现出加-显效应;(2)SHW-Z1条锈病的抗性改良效果优于SHW-Z2和SHW-Z4;（3）本研究材料的条锈病抗性基因可能还受到遗传背景的影响。本研究对这3份人工合成小麦的条锈病抗性改良和育种利用提供了理论依据,同时可为相关研究提供参考。     

Dynamic Characteristic of Micromachined Arrayed Microphone

Several kinds of microphone developed via silicon Micro Electro Mechanical System (MEMS) processes in recent years are investigated in this paper. The analysis and research of the merits and demerits of them show that: these conventional microphones mainly function as acoustic electric converters; the filtering against noise is implemented by a special filter circuit or a digital filter. Therefore, a design method for silicon micro array filter microphone is presented, where the filtering is achieved while converting signals. The effects of damping, silicon reed number and sensor element pair on dynamic characteristic are discussed; it follows that: Increasing damping can reduce microphone's sensitivity, while increasing silicon reed number can improve the sensitivity curve, furthermore, the implementation of the filter function is simultaneous with acquiring of the noise signal.