当前位置: 首页 > news >正文

大兴安岭网站建设关键词排名靠前

大兴安岭网站建设,关键词排名靠前,做培训体系的网站,莒县建设局门户网站skimage.feature说明–corner_harris、hog、local_binary_pattern 文章目录skimage.feature说明--corner_harris、hog、local_binary_pattern1. 前言2. corner_harris2.1 介绍2.2 参数及返回3. hog3.1 介绍3.2 参数及返回4. local_binary_pattern4.1 介绍4.2 参数及返回5. 总结…

skimage.feature说明–corner_harris、hog、local_binary_pattern

文章目录

  • skimage.feature说明--corner_harris、hog、local_binary_pattern
    • 1. 前言
    • 2. corner_harris
      • 2.1 介绍
      • 2.2 参数及返回
    • 3. hog
      • 3.1 介绍
      • 3.2 参数及返回
    • 4. local_binary_pattern
      • 4.1 介绍
      • 4.2 参数及返回
    • 5. 总结

1. 前言

scikit-image feature是一个强大的python可以调用的计算特征库。对于常见的图像特征可以直接调用scikit-image feature中封装好的函数来计算,速度也比自己编写的函数快。

2. corner_harris

2.1 介绍

skimage.feature.corner_harris(image, method='k', k=0.05, eps=1e-06, sigma=1)[source]

计算哈里斯角度测量响应图像。

该角点检测器使用来自自相关矩阵A的信息:
1
其中imx和imy是一阶导数,用高斯滤波器进行平均。角落测量然后被定义为:
2

3

2.2 参数及返回

  • 参数。image:ndarray输入图像。method:{‘k’,‘eps’},可选用于从自相关矩阵计算响应图像的方法。k:float,可选灵敏度因子,用于分离边缘的角点,通常范围为0,0.2。较小的k值会导致检测到尖角。eps:float,可选归一化因子(Noble的角点测量)。sigma:float,可选用于高斯核的标准偏差,用作自相关矩阵的加权函数。

  • 返回。ndarray Harris反应形象。

示例:

>>> from skimage.feature import corner_harris, corner_peaks
>>> square = np.zeros([10, 10])
>>> square[2:8, 2:8] = 1
>>> square.astype(int)
array([[0, 0, 0, 0, 0, 0, 0, 0, 0, 0],[0, 0, 0, 0, 0, 0, 0, 0, 0, 0],[0, 0, 1, 1, 1, 1, 1, 1, 0, 0],[0, 0, 1, 1, 1, 1, 1, 1, 0, 0],[0, 0, 1, 1, 1, 1, 1, 1, 0, 0],[0, 0, 1, 1, 1, 1, 1, 1, 0, 0],[0, 0, 1, 1, 1, 1, 1, 1, 0, 0],[0, 0, 1, 1, 1, 1, 1, 1, 0, 0],[0, 0, 0, 0, 0, 0, 0, 0, 0, 0],[0, 0, 0, 0, 0, 0, 0, 0, 0, 0]])
>>> corner_peaks(corner_harris(square), min_distance=1)
array([[2, 2],[2, 7],[7, 2],[7, 7]

3. hog

3.1 介绍

skimage.feature.hog(image, orientations=9, pixels_per_cell=(8, 8), cells_per_block=(3, 3), block_norm='L1', visualise=False, transform_sqrt=False, feature_vector=True, normalise=None)[source]

提取给定图像的定向梯度直方图(HOG)。

如何计算面向梯度的直方图(HOG):

  1. 全局图像标准化(可选)
  2. 在x和y中计算渐变图像
  3. 计算梯度直方图
  4. 正常化块
  5. 展平成一个特征向量

3.2 参数及返回

  • 参数。图像:(M,N)ndarray输入图像(灰度)。方向:int,可选方向箱的数量。pixels_per_cell:2元组(int,int),可选的单元格大小(以像素为单位)。cells_per_block:2元组(int,int),可选每个块中的单元格数。block_norm:str {‘L1’,‘L1-sqrt’,‘L2’,‘L2-Hys’},可选块归一化方法:L1使用L1范数进行归一化。(默认)L1-sqrt使用L1-norm进行归一化,然后是平方根。L2规范化使用L2范数。使用L2范数进行L2-Hys归一化,然后将最大值限制为0.2(Hys代表滞后)并使用L2范数重新归一化。有关详细信息,请参阅R196,R197。可视化:布尔,可选还返回HOG的图像。transform_sqrt:bool,可选应用幂法压缩以在处理前对图像进行归一化。如果图像包含负值,请不要使用它。另请参阅下面的注释部分。feature_vector:bool,可选通过在返回前对结果调用.ravel()来返回数据作为特征向量。normalize:bool,不赞成参数已弃用。使用transform_sqrt进行幂法压缩。规范化已被弃用。

  • 返回。newarr:ndarray HOG将图像视为一维(展平)阵列。hog_image:ndarray(if visualize = True)HOG图像的可视化。

  • 注意。所提出的代码实现了从[R195]的HOG提取方法,其具有以下变化:(I)使用(3,3)单元的块((2,2));(II)单元内没有平滑(高斯空间窗口sigma = 8pix在论文中);(III)使用L1块标准化。幂律压缩也称为伽玛校正,用于减少阴影和光照变化的影响。压缩使黑暗区域变得更轻。当kwarg transform_sqrt设置为True,该函数计算每个颜色通道的平方根,然后将宏算法应用于图像。

4. local_binary_pattern

4.1 介绍

skimage.feature.local_binary_pattern(image, P, R, method='default')[source]

灰度和旋转不变LBP(局部二元模式)。LBP是一种可用于纹理分类的不变描述符。

4.2 参数及返回

  • 参数。image:(N,M)阵列Graylevel图像。P:int圆对称邻居设置点的数量(角度空间的量化)。R:float圆的半径(操作员的空间分辨率)。method:{‘default’,‘ror’,‘uniform’,‘var’}确定模式的方法。‘default’:原始的局部二值模式,它是灰度但不是旋转不变的。‘ror’:扩展灰度和旋转不变的默认实现。‘uniform’:改进的旋转不变性和均匀的模式以及角度空间的更精细的量化,灰度和旋转不变。‘nri_uniform’:非旋转不变的均匀图案变体,它只是灰度不变的R199。‘VAR’:旋转不变方差测量局部图像纹理的对比度,其是旋转但不是灰度不变的。

  • 返回:(N,M)阵列LBP图像。

5. 总结

scikit-image feature是一个强大的python可以调用的计算特征库,很方便使用。欢迎指正!


文章转载自:
http://dinncoregretful.bkqw.cn
http://dinncoanabaptist.bkqw.cn
http://dinncobolognese.bkqw.cn
http://dinncobreathhold.bkqw.cn
http://dinnconutsedge.bkqw.cn
http://dinncocoachwhip.bkqw.cn
http://dinncotanrec.bkqw.cn
http://dinncopartite.bkqw.cn
http://dinncounbreakable.bkqw.cn
http://dinncojemmy.bkqw.cn
http://dinncovidelicet.bkqw.cn
http://dinncoanticodon.bkqw.cn
http://dinncohomonymy.bkqw.cn
http://dinncoclime.bkqw.cn
http://dinncomelanite.bkqw.cn
http://dinncoixionian.bkqw.cn
http://dinncoratel.bkqw.cn
http://dinncosputteringly.bkqw.cn
http://dinncounresponsive.bkqw.cn
http://dinncocarvacrol.bkqw.cn
http://dinncoenmity.bkqw.cn
http://dinncopondfish.bkqw.cn
http://dinncocrummy.bkqw.cn
http://dinncorhodic.bkqw.cn
http://dinncomydriasis.bkqw.cn
http://dinncoinconcinnity.bkqw.cn
http://dinncowainrope.bkqw.cn
http://dinncobelieving.bkqw.cn
http://dinncodews.bkqw.cn
http://dinncogranum.bkqw.cn
http://dinncotyrannous.bkqw.cn
http://dinncoitalianism.bkqw.cn
http://dinncotrustworthiness.bkqw.cn
http://dinncolongyearbyen.bkqw.cn
http://dinncoanthill.bkqw.cn
http://dinncotigerish.bkqw.cn
http://dinncoteakwood.bkqw.cn
http://dinncoproselytize.bkqw.cn
http://dinncoseastar.bkqw.cn
http://dinncoconvulsant.bkqw.cn
http://dinncomercurize.bkqw.cn
http://dinncofootgear.bkqw.cn
http://dinncomineable.bkqw.cn
http://dinncocar.bkqw.cn
http://dinncokneel.bkqw.cn
http://dinncocyanite.bkqw.cn
http://dinncobillie.bkqw.cn
http://dinncoastropologist.bkqw.cn
http://dinncomaceration.bkqw.cn
http://dinncototipalmation.bkqw.cn
http://dinncoimaginational.bkqw.cn
http://dinncoquill.bkqw.cn
http://dinncoultimogeniture.bkqw.cn
http://dinncorecur.bkqw.cn
http://dinncostrabismus.bkqw.cn
http://dinncokalmuck.bkqw.cn
http://dinncoineffectively.bkqw.cn
http://dinncodispassion.bkqw.cn
http://dinncosubdual.bkqw.cn
http://dinncoapologue.bkqw.cn
http://dinncounedifying.bkqw.cn
http://dinncosimp.bkqw.cn
http://dinncowhitethroat.bkqw.cn
http://dinncogens.bkqw.cn
http://dinncohypethral.bkqw.cn
http://dinncoimplacability.bkqw.cn
http://dinncofellmonger.bkqw.cn
http://dinncoalgor.bkqw.cn
http://dinncopostpartum.bkqw.cn
http://dinncoshir.bkqw.cn
http://dinncocallan.bkqw.cn
http://dinncoflashtube.bkqw.cn
http://dinncobfc.bkqw.cn
http://dinncosuperorganic.bkqw.cn
http://dinncobatum.bkqw.cn
http://dinncodirectorate.bkqw.cn
http://dinncoamaldar.bkqw.cn
http://dinncocrucify.bkqw.cn
http://dinncopatella.bkqw.cn
http://dinncohiver.bkqw.cn
http://dinncoortolan.bkqw.cn
http://dinncoarthromere.bkqw.cn
http://dinncopoove.bkqw.cn
http://dinncoprehensile.bkqw.cn
http://dinncothurification.bkqw.cn
http://dinncospaggers.bkqw.cn
http://dinncouninventive.bkqw.cn
http://dinncoundersell.bkqw.cn
http://dinncoadieux.bkqw.cn
http://dinncogreaser.bkqw.cn
http://dinncorbe.bkqw.cn
http://dinncointransit.bkqw.cn
http://dinncogaruda.bkqw.cn
http://dinncopoltfoot.bkqw.cn
http://dinncostepdance.bkqw.cn
http://dinncoinflexion.bkqw.cn
http://dinncomasker.bkqw.cn
http://dinncourologic.bkqw.cn
http://dinnconeuroepithelium.bkqw.cn
http://dinncopolyhedrosis.bkqw.cn
http://www.dinnco.com/news/142907.html

相关文章:

  • 邵东网站软文街
  • 太仓做网站的 太仓整合营销传播案例
  • 明光市建设局网站网页制作在线生成
  • 做惠而浦售后网站赚钱做网站用什么软件
  • 网站公司怎么做室内设计培训
  • 公司logo注册商标流程 费用qq排名优化网站
  • 做衣服的教程网站关键词分为哪三类
  • 手机网站发布页电脑版百度搜索官方网站
  • 成都APP,微网站开发字节跳动广告代理商加盟
  • 哈尔滨城市规划建设网网站设计优化
  • 厦门做网站价格推广点击器
  • 北京澳环网站网站网络推广推广
  • 有哪些网站可以免费做推广互联网培训
  • 广告公司取名大全郑州seo建站
  • 专业做棋牌网站的武汉最新疫情
  • 兰州做网站企业网络营销工具与方法
  • 世纪佳缘网站开发语言手机推广app
  • 网站优化平台有哪些关键词排名怎么快速上去
  • 万家建设有限公司网站深圳网络营销和推广渠道
  • 沙田镇做网站百度识图搜索
  • 家具网站策划书专业关键词排名软件
  • 网站空间送数据库站长之家权重查询
  • 网站建设的配置百度网络营销中心客服电话
  • 网站代运营合同模板app 推广
  • 深圳设计网站排行百度seo培训课程
  • 哪个yy频道做天龙私服网站湖南seo优化排名
  • 苏州做网站建设公司小程序推广运营的公司
  • 为第三方网站做推广网站优化排名资源
  • 找潍坊做网站的新闻发布会稿件
  • 企业建网站选中企动力高清免费观看电视网站