首页 | 本学科首页   官方微博 | 高级检索  
相似文献
 共查询到20条相似文献,搜索用时 31 毫秒
1.
Extremely high correlation between the probability that a single cell will fire and the amplitude of microelectrode electroencephalogram was established in on-line, real-time computer experiments. Such congruence between spike and wave shows orderly biological variation under sensory control. Congruence, as a measure of brain activity, provides an immediate estimate of regional neural homogeneity of function of cell populations in brain.  相似文献   

2.
长输管道的泄漏检测和定位对管道安全平稳运行意义极其重大,在以软件计算为主的检测方法里,模糊神经网络模型综合了模糊算法和神经网络模型的优点,能较好地适应长输管道的非线性特征。本文采用中亚地区某管道某相邻两站场的历史数据训练基于高木-关野(Takagi-Sugeno)模糊神经网络的预测模型,使用STONER管道仿真软件产生实时数据,用一种较简单的软方法较好的实现管道泄漏定位,该种方法对中亚某长输管道这类没有专门硬件泄漏检测设备和系统的管线有一定的实用意义。  相似文献   

3.
Erratum     
《Science (New York, N.Y.)》1986,232(4751):698
The caption for the photograph accompanying the review of Neurobiology of Arachnids (28 Mar., p. 1610) should have read, "Peocilotheria regalis, an Indian bird-spider. Beneath it is a spike train recorded from a sensory neuron responding to mechanical stimuli in the spider Cupiennius salei."  相似文献   

4.
[目的]基于深度学习的钻蛀振动识别模型结合边缘计算可实现林业蛀干害虫钻蛀振动长期监测和实时预警,但要求大幅压缩智能识别模型的参数量和运算量.本研究采用深度学习模型压缩算法,在不损失精度的前提下,对已有的钻蛀振动识别模型进行压缩,减小模型的体积并提升模型在嵌入式平台的识别速度.[方法]首先采集双条杉天牛钻蛀振动和背景噪声...  相似文献   

5.
春小麦数量性状的相关和通径分析   总被引:3,自引:1,他引:3  
为了明确小麦主要数量性状之间的相关关系以及各性状对单穗粒重的贡献大小,本试验选用了20个小麦品种(系),针对15个主要数量性状进行了研究.结果表明:单穗粒重、千粒重、每穗粒数、第1节长的遗传力很高;每穗粒数与单穗粒重呈极显著正相关,株高、千粒重、第1节、第2节与单穗粒重呈显著正相关;株高、I2、每穗小穗数、千粒重对单穗粒重的遗传直接通径为正贡献,每穗粒数、第1节、IL、第2节通过I2,穗长通过每穗小穗数对单穗粒重的间接贡献较大。  相似文献   

6.
目的利用红外自动感应相机对野生动物进行图像监测是对野生动物保护管理的有效手段,为了解决野外复杂背景环境导致的野生动物监测图像自动识别准确率低的问题,提出一种基于感兴趣区域(ROI)与卷积神经网络(CNN)的野生动物物种自动识别方法。方法以红外自动感应相机在内蒙古赛罕乌拉国家自然保护区内拍摄的马鹿、斑羚、猞猁、狍和野猪这5种国家级陆生保护动物的图像为实验样本,采用基于回归算法的目标检测方法,对监测图像中野生动物区域进行检测并分割,生成ROI图像,减少复杂背景信息对物种识别的干扰;利用裁剪、仿射变换等方式对样本数据进行扩充;构建基于全局-局部的VGG16双通道网络模型对样本图像进行训练,最后接入分类器输出物种识别结果。同时,构建了基于VGG19的双通道网络模型对样本图像进行训练,并与本研究训练结果进行比较;另外,将样本图像分别输入本研究算法与VGG16、R-CNN、Fast R-CNN算法进行训练,对比不同算法下的识别效果。结果利用本研究模型对样本图像进行训练时,测试集的平均识别精度均值MAP达到0.912,相对于VGG19结构下的训练模型和VGG16、R-CNN、Fast R-CNN,得到了更高的MAP值。结论相比于其他算法,本研究提出的物种识别模型更适合于复杂背景下的野生动物监测图像的物种识别,可以得到更高的MAP值与更优的识别效果。   相似文献   

7.
Short light flashes can appear brighter than longer flashes. This brightness enhancement has often been attributed to neural transients occurring shortly after stimulus onset. This attribution assumes an equivalence between the totality of the response to a stimulus of a given duration and the instantaneous response at a given time after stimulus onset. Recordings from Limulus photoreceptors indicate that this attribution is an example of illusory correlation.  相似文献   

8.
《数据结构》课程中的案例教学初探   总被引:10,自引:0,他引:10  
李克清 《长江大学学报》2004,1(4):135-136,i006
针对<数据结构>教学过程中存在着的"重知识传授,轻编程实践"矛盾,在教学过程中引进案例教学模式,提出了课堂引导案例与深化讨论案例的设计.教师精心准备典型案例,积极引导学生阅读、思考、分析、讨论和实践,收到了良好的教学效果.  相似文献   

9.
Insect-eating bats find their aerial food by sonar, through emitting ultrasonic chirps and locating sources of echoes. Certain moths have ears sensitive to these chirps and can detect bats well beyond the range of the bats' sonar. On hearing a distant bat, many moths turn and fly directly away from the source of ultrasound. Only one sense cell in each ear of a moth provides the primary nervous information for this response. This article describes my initial attempts to find out how a moth's central nervous system processes the train of chirps reaching its two ears. The ear of a restrained moth is exposed to a sequence of artifically generated ultrasonic pulses that approximates the cries made by a bat. This stimulus can be varied with respect to ultrasonic frequency (pitch), pulse intensity, pulse duration, the interval between pulses, and pulse-train duration. The more sensitive acoustic sense cell responds to all frequencies between about 15,000 and 80,000 cycles per second, but the signal that it transmits to the moth's central nervous system contains no measure of frequency within this range. However, this nerve signal reports variations in the other parameters of the stimulus. The acoustic fiber connects, in the central nervous system, with various nerve cells that transform the signal farther. The signal from a pulse-marker neuron contains no measures of pulse intensity or pulse duration, reporting only changes in interpulse interval and pulse-train duration. A train-marker neuron reports only the duration of the pulse train. The stimulus parameters may be likened to keys, each of which is necessary to gain admittance through a given door but becomes superfluous once this door has been passed. This analogy suggests one of the ways in which a signal is transformed in its passage through the nervous system, and how its specificity is assured in eliciting a given response. In addition to undergoing this kind of transformation, neural signals generated in the two directionally sensitive ears must be combined if a flying moth is to steer a course away from a distant bat. Neurons have been discovered in the central ganglia which summate signals from the right and left ears. Other neurons are inhibited in their activity by stimulation of one ear. The moth may combine signals from these neurons with motor-nerve information on the attitude of its own wings, which act as oscillating baffles modifying its directional acoustic sensitivity 20 to 40 times a second as it flaps an erratic path through the darkness.  相似文献   

10.
黄岩  毕坤 《安徽农业科学》2011,(8):4700-4702,4786
通过建立地面太阳总辐射的Elman网络估算模型,利用北京市1971~1995年的气象数据训练网络,并对1996~2000年的数据进行估算。对比分析结果表明,基于Elman网络的地面太阳总辐射估算方法具有较高的精度。  相似文献   

11.
张聪 《安徽农业科学》2006,34(21):5746-5747
分析了G.723.1协议编解码算法的特点,提出了利用DirectX技术来进行储粮害虫声信号数据编解码处理的方法。介绍了如何使用DirectSound技术来实现储粮害虫声信号的获取和回放。  相似文献   

12.
【目的】针对传统植保无人机在定量喷施作业时由于飞行速度的变化造成施药不均匀以及传统控制算法无法满足无人机变量喷雾系统所需的实时性和稳定性等问题,设计一种基于神经网络PID的自适应无人机变量喷雾系统。【方法】采用风压变送器测出无人机的飞行速度,根据速度采用脉宽调制(PWM)方法进行自适应变量喷雾,同步用流量传感器测出实际喷雾流量,融合BP神经网络PID控制算法调节喷雾流量。由MATLAB构建BP神经网络PID控制算法,并与PID、模糊PID和神经元PID对比及分析;田间试验过程中,对比分析无人机定量喷雾与随飞行速度改变的变量喷雾效果,采用水敏纸获取雾滴沉积量分布,分别从整体区域、飞行方向和喷杆方向评价沉积量分布的均匀性。【结果】算法仿真对比试验结果表明,与PID、模糊PID和神经元PID相比,BP神经网络PID阶跃响应上升时间分别少28.57%、84.73%和31.03%,正弦跟踪平均误差分别小63.01%、87.03%和0.58%,方波跟踪平均误差分别小74.00%、79.53%和6.80%,鲁棒性强,无静差,超调量为1.20%;喷雾对比试验结果表明,本系统能够根据飞行速度自适应调节喷雾流量,实际流量与目标流量的平均偏差为8.43%,水敏纸扫描结果表明总体区域雾滴沉积量的变异系数对比定量喷雾平均降低26.25%,喷杆方向平均降低18.79%。【结论】该研究结果可为农业航空变量喷雾技术的应用提供理论基础。  相似文献   

13.
针对传统渠道输水PID控制方法响应速度慢、超调量大、参数不能在线自调整的不足,根据RBF神经网络和渠道输水特点,提出了将传统渠道下游常水位输水PID控制和RBF人工神经网络结合的输水控制方法,使输水控制具有自学习、自适应、容错性和鲁棒性。推导了RBF网络整定PID输水控制调节器的算法。仿真结果表明,基于RBF网络的PID输水控制方法,能够通过不断学习自动调整控制参数,使输水控制过程超调量小、响应速度快,具有不需要特意选择或计算控制参数的特点。因此,基于RBF神经网络的参数非线性PID控制更适合进行渠道输水这样高度非线性系统的实时控制过程。  相似文献   

14.
高清视频车辆检测及跟踪系统的设计与实现   总被引:1,自引:0,他引:1  
针对目前视频车辆检测与跟踪系统存在检测精度低、跟踪稳定性差等问题,设计基于数字信号处理器(DSP)和现场可编程门阵列(FPGA)结构的高清视频车辆检测和跟踪系统。该系统采用高清摄像机为图像采集单元,利用现场可编程门阵列对采集到的IP视频图像进行实时图像解码和图像预处理,高性能DSP做为系统主控单元实现前景提取、车辆检测、识别和跟踪等功能,DSP与FPGA之间的数据交换通过两路高速串行接口连接,以满足运算处理时大批量中间数据的交互;采用基于背景图像差分检测方法进行运动目标的实时检测,通过计算目标物体的紧密度对运动目标进行分类,利用区域特征跟踪法来快速跟踪图像序列中的车辆目标。试验结果表明:与普通视频相比,高清视频条件下对视频图像进行处理,在定位及寻找物体边缘方面具有优势,提高检测精度10%以上,能够实现对运动车辆的实时、准确、快速跟踪。  相似文献   

15.
土壤水分含量是影响作物生长的重要因素,精确预测技术对水资源的合理利用与管理具有重要的指导意义。土壤水分运动是一个复杂的时间序列系统,其变化与区域气候条件和生态环境密切相关,具有明显的随机性波动。利用人工神经网络的方法对河南驻马店地区的土壤含水量进行预报,利用表层土壤含水量资料计算了一些与深层土壤含水量相关的预报因子,用以建立驻马店地区深层土壤含水量的神经网络预报模型,并应用独立样本进行了初步的模拟预报检验。结果表明,预报模型取得了令人满意的效果,应用神经网络的方法预报深层土壤含水量是可行的。  相似文献   

16.
The neural encoding of serial order was studied in the motor cortex of monkeys performing a context-recall memory scanning task. Up to five visual stimuli were presented successively on a circle (list presentation phase), and then one of them (test stimulus) changed color; the monkeys had to make a single motor response toward the stimulus that immediately followed the test stimulus in the list. Correct performance in this task depends on memorization of the serial order of the stimuli during their presentation. It was found that changes in neural activity during the list presentation phase reflected the serial order of the stimuli; the effect on cell activity of the serial order of stimuli during their presentation was at least as strong as the effect of motor direction on cell activity during the execution of the motor response. This establishes the serial order of stimuli in a motor task as an important determinant of motor cortical activity during stimulus presentation and in the absence of changes in peripheral motor events, in contrast to the commonly held view of the motor cortex as just an "upper motor neuron."  相似文献   

17.
The ability of rats to use their whiskers for fine tactile discrimination rivals that of humans using their fingertips. Rats perform discriminations rapidly and accurately while palpating the environment with their whiskers. This suggests that whisker deflections produce a robust and reliable neural code. Whisker primary afferents respond with highly reproducible temporal spike patterns to transient stimuli. Here we show that, with the use of a linear kernel, any of these reproducible response trains recorded from an individual neuron can reliably predict complex whisker deflections. These predictions are significantly improved by integrating responses from neurons with opposite angular preferences.  相似文献   

18.
An electrical response which reproduces the waveform and frequency of the sound stimulus can be recorded from the central neural pathway for audition. Controversy has existed for some years over whether this frequency-following response (FFR) is neural or an artifact such as remote pickup of the cochlear microphonic or cross talk in the recording system. Two experiments resolve this issue by demonstrating that the frequency-following response depends upon functionally intact neural pathways. The frequency-following response, as well as auditory evoked potentials, is abolished by section of the eighth nerve; it is reversibly abolished by cooling of the cochlear nucleus.  相似文献   

19.
在分析发动机结构参数和运转参数对发动机性能影响的基础上,提出了一种基于组合式神经网络的柴油机性能状态评估预测模型.该模型首先运用动态聚类法将大样本分成若干小组,然后分别用于子网络训练.性能评估时,运用模糊识别法选择相关的子网络进行评估分析.实例验证表明,这种模型能有效解决大样本下神经网络训练速度慢和难以收敛的问题,提高柴油机性能评估预测精度.  相似文献   

20.
Recent development of neural prosthetics for assisting paralyzed patients has focused on decoding intended hand trajectories from motor cortical neurons and using this signal to control external devices. In this study, higher level signals related to the goals of movements were decoded from three monkeys and used to position cursors on a computer screen without the animals emitting any behavior. Their performance in this task improved over a period of weeks. Expected value signals related to fluid preference, the expected magnitude, or probability of reward were decoded simultaneously with the intended goal. For neural prosthetic applications, the goal signals can be used to operate computers, robots, and vehicles, whereas the expected value signals can be used to continuously monitor a paralyzed patient's preferences and motivation.  相似文献   

设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号