1)  Hungary algorithm
Hungary算法
2)  Hu moment
Hu矩
1.
By applying Zernike moment in facial expression recognition, Hu moment invariant and Zernike moment invariant respectively in human facial expression images are calculated.
首次将Zernike矩运用于同一个人的表情识别问题,分别计算了同一个人面部图像的Hu矩和Zernike矩,以模式识别中常用的类间距作为依据,提取了面部图像的各种矩的最好特征和次好特征,并对Hu矩和Zernike矩的识别能力和抗噪性进行了比较,试验结果证明Zernike矩更适合识别细微变化的面部表情。
2.
This paper is mainly dedicated to the comparison of three kinds of moment features; Hu moment, Zernike moment and Wavelet moment and sum up the work and views of previous literatures.
综述了前人使用Hu矩、 Zernike矩和小波矩进行目标识别的研究工作,并比较了它们在具体应用中的特点, 为图像识别算法中目标矩特征的选择提供了一定的依据。
3.
In this paper, an image description method of identifying intravenous needles images by Hu moment invariants is studied.
为此,结合输液针的特点,对多种描述方法做了分析,提出了用Hu矩不变量来描述输液针图像,并用逆光拍摄方法来获取各个角度输液针针尖图像,从而获取了完整的轮廓信息,其二值化后的图像避免了矩值灰度级计算的不稳定性。
3)  SCID-hu
SCID-hu
1.
By intraperitoneal injection of fetal liver cells (FLC,the human immune system (SCID-hu) in severe combined immunodeficiency mouse was reconstituted.
用人胎肝细胞重建重症联合免疫缺陷鼠(SCID)人免疫系统(SCID-hu)的初步结果。
4)  Hu moments
Hu矩
1.
It is proved by theoretical analysis and experiment that Hu moments and Zernike moments combined with neural network can be used for the character recognition, and the recognition effect of Zernike moments is better than Hu moments.
理论分析与实验证明 ,Hu矩和Zernike矩与神经网络相结合 ,可运用于字符图像的识别 ,且Zernike矩的识别效果优于Hu矩。
2.
Hu moments and Zernike moments are used to identify images, and the results of identification using these two methods are compared.
将Hu矩和Zernike矩应用于图象识别 ,比较了两者的识别效果。
5)  Wee1Hu gene
Wee1 Hu
6)  Hu's moment
Hu距
参考词条
补充资料:BP算法
分子式:
CAS号:

性质:又称逆推学习算法,简称BP算法,是1986年鲁梅哈特(D. E. Rumelhart)和麦克莱朗德(J. L. McClelland)提出来的。用样本数据训练人工神经网络(一种模仿人脑的信息处理系统),它自动地将实际输出值和期望值进行比较,得到误差信号,再根据误差信号从后(输出层)向前(输入层)逐层反传,调节各神经层神经元之间的连接权重,直至误差减至满足要求为止。反向传播算法的主要特征是中间层能对输出层反传过来的误差进行学习。这种算法不能保证训练期间实现全局误差最小,但可以实现局部误差最小。BP算法在图像处理、语音处理、优化等领域得到应用。

说明:补充资料仅用于学习参考,请勿用于其它任何用途。