WOLFRAM

Wolfram 语言与系统参考资料中心

LocalAdaptiveBinarize[image,r]

通过将高于范围 r 邻域的均值用1替换，其他数值用0替换，从 image 创建二值图像.

LocalAdaptiveBinarize[image,r,{α,β,γ}]

将高于 的数值用1替换，其他数值用0替换，其中 和 是局部均值和标准差.

更多信息和选项
Details and Options Details and Options
范例  
基本范例  
范围  
Data  
Parameters  
选项  
PerformanceGoal  
Padding  
应用  
属性和关系  
互动范例  
参见
相关指南
历史
按以下格式引用：

LocalAdaptiveBinarize

LocalAdaptiveBinarize[image,r]

通过将高于范围 r 邻域的均值用1替换，其他数值用0替换，从 image 创建二值图像.

LocalAdaptiveBinarize[image,r,{α,β,γ}]

将高于 的数值用1替换，其他数值用0替换，其中 和 是局部均值和标准差.

更多信息和选项

范例

打开所有单元 关闭所有单元

基本范例  (1)

在出现不均匀照度时进行局部自适应阈值化处理彩色图像：

Wolfram Language code: LocalAdaptiveBinarize[[image], 4]

范围  (10)

Data  (3)

局部阈值化处理一个灰度图：

Wolfram Language code: LocalAdaptiveBinarize[[image], 20]

三维体：

Wolfram Language code: LocalAdaptiveBinarize[[image], 10]

彩色三维体：

Wolfram Language code: LocalAdaptiveBinarize[[image], 10]

Parameters  (7)

指定半径参数为 1 以使用 3×3 的邻域：

Wolfram Language code: LocalAdaptiveBinarize[[image], 1]

指定半径参数为 10 以使用 21×21 的邻域：

Wolfram Language code: LocalAdaptiveBinarize[[image], 10]

段像素比其周围的平均局部像素更亮：

Wolfram Language code: LocalAdaptiveBinarize[[image], 3, {1, 0, 0}]

段像素比局部平均像素的 0.9 部分更亮：

Wolfram Language code: LocalAdaptiveBinarize[[image], 3, {0.9, 0, 0}]

段像素比局部平均像素加半个标准差更亮：

Wolfram Language code: LocalAdaptiveBinarize[[image], 3, {1, 0.5, 0}]

段像素比局部平均像素减 0.05 更亮：

Wolfram Language code: LocalAdaptiveBinarize[[image], 3, {1, 0, -0.05}]

指定局部平均值、局部标准差和偏移的阈值参数：

Wolfram Language code: LocalAdaptiveBinarize[[image], 3, {1, 0.4, -.1}]

选项  (8)

PerformanceGoal  (4)

使用 PerformanceGoal->"Quality" 强化结果的质量：

Wolfram Language code: RepeatedTiming[LocalAdaptiveBinarize[[image], 10, PerformanceGoal -> "Quality"]]

使用 PerformanceGoal->"Speed" 加强计算速度：

Wolfram Language code: RepeatedTiming[LocalAdaptiveBinarize[[image], 10, PerformanceGoal -> "Speed"]]

通过 PerformanceGoal->"Speed"，降低右下角的质量：

Wolfram Language code: Table[LocalAdaptiveBinarize[ImageRotate[[image], side], 10, PerformanceGoal -> "Speed"], {side, {Top, Left, Right}}]

通过 PerformanceGoal->"Quality"，计算时间随半径增加：

Wolfram Language code: i = RandomImage[1, {2000, 1000}];
Wolfram Language code: timing1 = Table[AbsoluteTiming[LocalAdaptiveBinarize[i, r, PerformanceGoal -> "Speed"]][[1]], {r, Range[1, 40, 2]}]
Wolfram Language code: timing2 = Table[AbsoluteTiming[LocalAdaptiveBinarize[i, r, PerformanceGoal -> "Quality"]][[1]], {r, Range[1, 40, 2]}]
Wolfram Language code: ListLinePlot[{timing1, timing2}, PlotLegends -> {"Speed", "Quality"}]

Padding  (4)

默认情况下，使用 "Fixed" 填充：

Wolfram Language code: LocalAdaptiveBinarize[[image], 20, Padding -> "Fixed"]

使用零填充：

Wolfram Language code: LocalAdaptiveBinarize[[image], 20, Padding -> 0]

在图像边缘使用较小的邻域：

Wolfram Language code: LocalAdaptiveBinarize[[image], 20, Padding -> "Ignored"]

比较不同的填充设置：

Wolfram Language code: Table[ {p, LocalAdaptiveBinarize[[image], 30, Padding -> p]}, {p, {1, "Fixed", "Periodic", "Reflected", "Reversed"}}]//TextGrid

应用  (3)

通常无法正确识别不均匀光照下的特征：

Wolfram Language code: image = [image];

在这种情况下，TextRecognize 失效，并且仅返回一些字母：

Wolfram Language code: TextRecognize[image]

通过消除不均匀照明来改善文本识别的结果：

Wolfram Language code: TextRecognize[LocalAdaptiveBinarize[image, 40, {0.85, 0.07}]]

通过消除不均匀的照明提取布料纹理：

Wolfram Language code: LocalAdaptiveBinarize[[image], 10]

分割电子板：

Wolfram Language code: LocalAdaptiveBinarize[[image], 3, {.95, 0, .05}]

属性和关系  (1)

均值和标准差的参数为零时，执行一个简单的阈值：

Wolfram Language code: image = RandomImage[]; thr = 0.5; LocalAdaptiveBinarize[image, 1, {0, 0, thr}] == Binarize[image, thr]

互动范例  (1)

通过交互的方式调整系数：

Wolfram Language code: Manipulate[LocalAdaptiveBinarize[[image], 10, {a, b, c}], {{a, .6}, -2, 2}, {{b, 2}, -2, 2}, {{c, 0}, -1, 1}]
Wolfram Research (2014)，LocalAdaptiveBinarize，Wolfram 语言函数，https://reference.wolfram.com/language/ref/LocalAdaptiveBinarize.html （更新于 2016 年）.

文本

Wolfram Research (2014)，LocalAdaptiveBinarize，Wolfram 语言函数，https://reference.wolfram.com/language/ref/LocalAdaptiveBinarize.html （更新于 2016 年）.

CMS

Wolfram 语言. 2014. "LocalAdaptiveBinarize." Wolfram 语言与系统参考资料中心. Wolfram Research. 最新版本 2016. https://reference.wolfram.com/language/ref/LocalAdaptiveBinarize.html.

APA

Wolfram 语言. (2014). LocalAdaptiveBinarize. Wolfram 语言与系统参考资料中心. 追溯自 https://reference.wolfram.com/language/ref/LocalAdaptiveBinarize.html 年

BibTeX

@misc{reference.wolfram_2026_localadaptivebinarize, author="Wolfram Research", title="{LocalAdaptiveBinarize}", year="2016", howpublished="\url{https://reference.wolfram.com/language/ref/LocalAdaptiveBinarize.html}", note=[Accessed: 03-November-2026]}

BibLaTeX

@online{reference.wolfram_2026_localadaptivebinarize, organization={Wolfram Research}, title={LocalAdaptiveBinarize}, year={2016}, url={https://reference.wolfram.com/language/ref/LocalAdaptiveBinarize.html}, note=[Accessed: 03-November-2026]}