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 125 毫秒
1.
针对传统卷积神经网络时间成本高的不足,对卷积神经网络进行了改进,减少其卷积核的数量,增加池化方式.为解决真实场景中自动驾驶系统和辅助驾驶系统中的道路交通标志识别问题,将改进的卷积神经网络运用到道路交通标志识别当中,以达到在较短时间内识别出交通标志的目的.以图形数据集GTRSB实景交通标志图像数据作为样本,用改进的卷积神经网络对实景交通标志进行识别,其识别总体准确率达到98.38%.实验结果表明,本方法可以在保持较高识别准确率的同时减少其识别的时间.  相似文献   

2.
基于彩色图像的指示标志检测   总被引:3,自引:0,他引:3  
智能交通系统是近年来国内外广泛关注的研究课题,在基于计算机视觉系统的交通标志识别中,关键步骤之一是如何快速有效地检测并提取交通标志。该文提出了基于色彩的指示标志检测与提取的快速有效方法。它包括:对实景图像进行色调不变的彩色增强,彩色聚类,形态滤波,模板匹配,特征点判别等步骤。最后,由搜索到的特征点从增强后的彩色图像中抽取指示标志。实验显示该方法能快速有效地从实景图像中检测指示标志。  相似文献   

3.
介绍了一种基于颜色分割和区域描述的交通标志检测方法。该方法利用在RGB空间的颜色聚类算法分割出感兴趣色彩区域,然后进行形态学处理,最后结合Hu不变矩对道路交通标志进行识别。识别结果表明,基于颜色聚类和Hu不变矩的交通标志识别方法具有很强的抗图像平移、缩放和旋转识别能力,并具有实现简单、识别速度快、准确率高等特点,有较高的实用价值。  相似文献   

4.
在智能交通系统中要求交通标志识别具有良好的鲁棒性、实时性,并且实际交通环境中可能因路标模糊、光照强弱、尺度大小、复杂背景等因素的问题,导致交通标志识别准确率很低。针对上述问题,提出了利用深度学习方法设计卷积神经网络,并通过卷积和池采样的多层处理,结合目标检测方法中的RPN网络结构,以提取图像的候选区域,从而对候选区域进行特征提取,最后利用全连接网络实现对特征图进行回归处理,获取检测目标的位置及识别。实验结果表明,该方法能有效地提高检测精度和计算效率,降低错误率,对于光照、旋转等不良因素下交通标志检测具有较好的稳定性和准确性,有效地提高了交通标志识别效率,具有良好的泛化能力和适应性,且满足一定的实时性的要求。  相似文献   

5.
如何从实景中有效地提取出交通标志是交通标志识别系统的关键,在分析中国道路限速交通标志的颜色和几何形状两种先验特征的基础上,以一种新的颜色滤波方法为基础,获得红色像素在Lab颜色空间中聚类范围的椭圆模型,提取出图像中的红色区域,得到二值化图像,然后采用基于梯度信息的Hough变换圆检测方法获得二值化图像中的圆和椭圆区域,从而实现一种将交通标志先验特征与机器学习算法相融合的智能检测方法。  相似文献   

6.
基于模糊小波神经网络的交通标志识别方法研究   总被引:2,自引:1,他引:1  
对交通标志进行实时、正确的识别,是车辆自动导航中一个重要方面.该文介绍了一种基于模糊小波神经网络的交通标志识别方法.该方法首先利用不变矩来提取图像特征,然后将特征向量输入模糊小波神经网络进行识别.该网络以小波函数作为模糊隶属函数,将模糊技术与神经网络相结合,利用神经网络实现模糊推理,并可对隶属函数的形状进行实时调整,从而使网络具有更强的学习和自适应能力.实验证明,该方法具有较高的识别精度和速度,在车辆自动导航中具有较高的应用价值.  相似文献   

7.
基于不变矩和神经网络的交通标志识别方法研究   总被引:5,自引:0,他引:5  
在交通标志实时识别过程中,由于参考图像与实测图像不是同时获取的,因此摄像机与被摄交通标志之间的位置难以保证完全相同。于是,所获取的参考交通标志图像与实测交通标志图像之间就可能产生几何失真。几何失真将对于图像识别的结果带来很大的影响。因此,需要寻找一种具有旋转和比例不变性的图像识别方法,以满足实际应用中的需要。针对上述问题,提出了一种基于不变矩和神经网络的交通标志识别算法。实验结果表明,所提出的识别算法具有很好的识别能力。  相似文献   

8.
综合颜色和形状的圆形交通标志检测方法   总被引:1,自引:0,他引:1       下载免费PDF全文
快速、可靠的交通标志检测是对其进行准确识别的前提,以颜色分割为基础,提出了一种基于曲线拟合的圆形交通标志检测算法。首先利用交通标志的颜色特征预分割出潜在的交通标志区域,然后针对圆形交通标志轮廓具有圆形这一关键特征,通过边缘检测并采用非线性最小二乘技术准确的确定出图像中的圆形交通标志区域。实验结果表明了算法的有效性。  相似文献   

9.
交通标志检测是进行交通标志识别系统的关键技术,提出一种基于图像的颜色和形状进行交通标志检测的方法.首先对图像进行灰度拉伸和噪声滤出的预处理,然后利用改进的K-means聚类算法对彩色图像进行颜色分割,最后采用基于Hough变换的形状检测技术对交通标志中的特殊形状进行定位,从而实现交通标志的检测.实验结果显示,该方法在各种复杂背景条件下检测出结果的平均正确率达到93.0%,优于同条件的算法且具有较高的实时性.  相似文献   

10.
交通标志检测是交通标志识别的难点。在复杂的交通环境下,采用传统颜色空间的固定阈值分割进行交通标志检测的方法鲁棒性差,难以准确有效地检测出交通标志。提出了一种基于三分量色差法和Ostu法的交通标志检测方法。首先,通过计算图像R、G、B分量的差值来得到红、蓝、黄三种颜色分量,然后利用Ostu法分别对它们进行阈值分割,得到交通标志的检测结果。实验结果表明,该算法的检测准确率和实时性满足实际要求。  相似文献   

11.
Performance factors such as robustness, speed, and tractability are important for the realization of practical computing systems. The aim of soft computing is to achieve these factors in practice by tolerating imprecision and uncertainty instead of depending on exact mathematical computations. The ink drop spread (IDS) method is a modeling technique that has been proposed as a new approach to soft computing. This method is characterized by a modeling process that uses image information without including complex formulas. In this study, the performance of the IDS method is investigated in terms of robustness, speed, and tractability, which are typical criteria that determine the importance of soft computing tools. Robustness is evaluated on the basis of noise tolerance and fault tolerance. Tractability is discussed from the viewpoints of interpretability and transparency. Based on comparative evaluations with artificial neural networks and fuzzy inference systems, this study demonstrates that the IDS method has superior capability to function as a soft computing tool.   相似文献   

12.
LS—SVM在混沌时间序列预测中的应用   总被引:9,自引:0,他引:9  
孙德山  吴今培 《微机发展》2004,14(1):21-22,25
支持向量机是一种基于统计学习理论的新颖的机器学习方法,该方法已广泛用于解决分类和回归问题。文中将最小二乘支持向量机算法应用于混沌时间序列预测中,并同BP网络及RBF网络的预测结果进行了比较分析。仿真实验表明,该方法具有很好的泛化能力和一定的噪声容忍能力。  相似文献   

13.
邢锦江  冯允成 《计算机工程》2006,32(3):31-33,36
为了提供一种简便和有效的真随机数生成方法,通过分析环境噪声数据的统计特性,提出同余法和正负法将声音波形数据进行处理。去除声波数据的周期性、连续性、相关性,提取其随机性。同真随机数采集系统设计相关的电路噪声问题以及采集速度问题也作了讨论。用此思路构建的真随机数采集系统原型已经能够用来产生性能较佳的真随机数,证明了设计思路和处理方法的正确有效性。  相似文献   

14.
Noise artifacts are one of the key obstacles in applying continuous monitoring and computer-assisted analysis of lung sounds. Traditional adaptive noise cancellation (ANC) methodologies work reasonably well when signal and noise are stationary and independent. Clinical lung sound auscultation encounters an acoustic environment in which breath sounds are not stationary and often correlate with noise. Consequently, capability of ANC becomes significantly compromised. This paper introduces a new methodology for extracting authentic lung sounds from noise-corrupted measurements. Unlike traditional noise cancellation methods that rely on either frequency band separation or signal/noise independence to achieve noise reduction, this methodology combines the traditional noise canceling methods with the unique feature of time-sprit stages in breathing sounds. By employing a multi-sensor system, the method first employs a high-pass fdter to elinfinate the off-band noise, and then performs time-shared bfind identification and noise cancellation with recursion from breathing cycle to cycle. Since no frequency separation or signal/noise independence is required, this method potentially has a robust and reliable capability of noise reduction, complementing the traditional methods.  相似文献   

15.

The hyperspectral image (HSI) denoising has been widely utilized to improve HSI qualities. Recently, learning-based HSI denoising methods have shown their effectiveness, but most of them are based on synthetic dataset and lack the generalization capability on real testing HSI. Moreover, there is still no public paired real HSI denoising dataset to learn HSI denoising network and quantitatively evaluate HSI methods. In this paper, we mainly focus on how to produce realistic dataset for learning and evaluating HSI denoising network. On the one hand, we collect a paired real HSI denoising dataset, which consists of short-exposure noisy HSIs and the corresponding long-exposure clean HSIs. On the other hand, we propose an accurate HSI noise model which matches the distribution of real data well and can be employed to synthesize realistic dataset. On the basis of the noise model, we present an approach to calibrate the noise parameters of the given hyperspectral camera. Besides, on the basis of observation of high signal-to-noise ratio of mean image of all spectral bands, we propose a guided HSI denoising network with guided dynamic nonlocal attention, which calculates dynamic nonlocal correlation on the guidance information, i.e., mean image of spectral bands, and adaptively aggregates spatial nonlocal features for all spectral bands. The extensive experimental results show that a network learned with only synthetic data generated by our noise model performs as well as it is learned with paired real data, and our guided HSI denoising network outperforms state-of-the-art methods under both quantitative metrics and visual quality.

  相似文献   

16.
Unlike computer systems, organisms have high adaptability in dealing with environmental changes or noise. The ability to evolve, self-organizing dynamics, and a closed structure–function relationship are the three principle features embedded in biological structures that provide great malleability to environmental change. Computer systems have fast processing speed for performing heavy computational tasks. One of the objectives in this research is to capture these three biological features and implement them onto a digital circuit. The proposed hardware (called neuromolecular hardware), is the integration of inter- and intraneuronal information processing applied to the pattern recognition problem domain. This approach was tested on the Quartus II system, a simulation tool for digital circuits. The experimental result showed good self-organizing capability in selecting significant bits for differentiating patterns and insignificant bits for tolerating noise. The proposed digital circuit also exhibited a closed structure–function relationship. This implied that this hardware embraced an adaptive fitness landscape that facilitated processing spatiotemporal information.  相似文献   

17.
Registration errors between two images produce artefacts when their per-pixel difference is used to detect changes. These artefacts constitute a source of noise hampering image interpretation. In this article, an adaptive filtering approach for misregistration artefact reduction is presented. Both univariate and multichannel images are considered. The proposed filters rely on robust statistics, switching mechanisms and a misregistration-induced change estimation model. An evaluation performed on a synthetic image confirms (1) the high efficiency of the approach in both reducing misregistration artefacts and preserving real changes and (2) the advantages of the method over the Knoll and Delp filter when applied to a difference image. Experiments conducted on real images show, from careful visual analysis, that the adaptive filters are able to remove misregistration noise with good capability whilst preserving real changes. The methods are designed for handling misregistration errors in the order of one to two pixels.  相似文献   

18.
将小波模极大值滤噪法用于处理含噪示波计时电位信号。研究了信号中有用信息频带大小和噪音频带大小对滤噪结果的影响。示波计时电位信号经小波模极大值滤噪法处理后,切口宽度和相对深度变化很小,切口相对深度的相对误差仅为1.3%。实验结果表明:该方法适合于对信号中有用信息频带较窄且位于低频的信号滤噪或有用信息频带与噪音频带(即有用信息的频率上限与噪音的频率下限)有一定差距的含噪信号中高频噪音的滤除。  相似文献   

19.
A Statistically-Switched Adaptive Vector Median Filter   总被引:2,自引:0,他引:2  
This paper presents a new cost-effective, adaptive multichannel filter taking advantage of switching schemes, robust order-statistic theory and approximation of the multivariate dispersion. Introducing the statistical control of the switching between the vector median and the identity operation, the developed filter enhances the detail-preserving capability of the standard vector median filter. The analysis and experimental results reported in this paper indicate that the proposed method is capable of detecting and removing impulsive noise in multichannel images. At the same time, the method is computationally efficient and provides excellent balance between the noise attenuation and signal-detail preservation. Excellent performance of the proposed method is tested using standard test color images as well as real images related to emerging virtual restoration of artworks. Rastislav Lukac: Corresponding author e-mail: lukacr@dsp.utoronto.ca, Web: http//www.dsp.utoronto.ca/∼lukacr.  相似文献   

20.
窄带移动自组网在同一时隙内只能有一个节点发送数据,否则会导致碰撞,降低网络吞吐量性能。移动自组网中引入码分多址技术可以显著提高吞吐量,将网络状态建模为马尔可夫链,提出一种多址干扰和噪声共同影响下的吞吐量性能分析方法,综合考虑了扩频码同步和解调误码率的影响。数值仿真表明提高信噪比、编码纠错能力,以及增加扩频码长度均能提升网络吞吐量性能,低信噪比时编码纠错能力对性能提升比高信噪比时明显。编码纠错能力和扩频码长度均存在最佳的取值范围,超过此范围继续提高则对吞吐量性能的提升作用不再明显,因此需要根据实际环境优化网络参数。提出的新的分析方法具有更加普遍的适用性,更符合网络的真实情况,仿真实验对网络设计具有参考价值,可以用作网络参数选取的理论依据。  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