[OpenCV4] cv :: adaptiveThreshold() 自适应阈值函数

 

官方文档：参考链接

1.函数原型

2.解释

2.1 头文件

#include <opencv2/imgproc.hpp>

2.2 功能

Applies an adaptive threshold to an array.

对于图片施加自适应阈值操作。

The function transforms a grayscale image to a binary image according to the formulae:

该函数可以将一个灰度图根据下面的公式转换为二值图：

• THRESH_BINARY

  

   

• THRESH_BINARY_INV

   

  

  where T(x,y) is a threshold calculated individually for each pixel (see adaptiveMethod parameter).
• T(x,y) 对于每一个像素单独进行计算

The function can process the image in-place.

该函数直接对原图像进行操作。

2.3 参数说明

Parameters

src	Source 8-bit single-channel image.
dst	Destination image of the same size and the same type as src.
maxValue	Non-zero value assigned to the pixels for which the condition is satisfied
adaptiveMethod	Adaptive thresholding algorithm to use, see AdaptiveThresholdTypes. The BORDER_REPLICATE | BORDER_ISOLATED is used to process boundaries.
thresholdType	Thresholding type that must be either THRESH_BINARY or THRESH_BINARY_INV, see ThresholdTypes.
blockSize	Size of a pixel neighborhood that is used to calculate a threshold value for the pixel: 3, 5, 7, and so on.
C	Constant subtracted from the mean or weighted mean (see the details below). Normally, it is positive but may be zero or negative as well.

src：8bit单通道输入图像

dst：与src相同大小和类型的输出图像

maxValue：非0值，在满足特定情况的时候被赋值给特定像素

adaptiveMethod：自适应阈值方法，OpenCV给出了两种标准自适应阈值方法：

1.ADAPTIVE_THRESH_MEAN_C （Python: cv.ADAPTIVE_THRESH_MEAN_C）

the threshold value T(x,y) is a mean of the blockSize×blockSize neighborhood of (x,y) minus C

这种阈值是(x, y) 点对应的Block*Block大小的邻域的平局值

 

2.ADAPTIVE_THRESH_GAUSSIAN_C （Python: cv.ADAPTIVE_THRESH_GAUSSIAN_C）

the threshold value T(x,y) is a weighted sum (cross-correlation with a Gaussian window) of the blockSize×blockSize neighborhood of (x,y) minus C . The default sigma (standard deviation) is used for the specified blockSize . See getGaussianKernel

这种阈值是有权重的加权和（在高斯滑窗中的互相关）在像素 (x, y)的blockSize*blockSize的邻域滑窗中每个像素都要-C

默认的sigma值（标准差）被在特定的块大小。计算公式如下：（ksize为kernel size）

sigma

Gaussian standard deviation. If it is non-positive, it is computed from ksize as sigma = 0.3*((ksize-1)*0.5 - 1) + 0.8.

thresholdType：阈值类型（ THRESH_BINARY和THRESH_BINARY_INV中的一个）

blockSize：用来做自适应kernel的块大小（3，5，7...）

C：最终要从得到的均值或加权均值减去的值，这个值可以是正，0，负

 

2.4 自适应阈值类型

ADAPTIVE_THRESH_MEAN_C 均值

ADAPTIVE_THRESH_GAUSSIAN_C 高斯

 

3.Demo

后面再补