matlab disparity函数

matlab生成视差图函数disparity参数选择对生成的视差图效果有很大的影响，这里将matlab帮助中的参数进行列举，方便之后修改查验。

`'Method'` — Disparity estimation algorithm
`'SemiGlobal'` (default) | `'BlockMatching'`

模式可选择BM和SGBM;

`'DisparityRange'` — Range of disparity
[`0 64`] (default) | two-element vector

视差图范围，默认是[0 64]

该值的设置取决于两相机的间距和物距综合影响，大基线距和近处测量时应增大视差范围，但范围差值要求可以被16整除。

`'BlockSize'` — Square block size
`15` (default) | odd integer

窗口大小，范围是5~255，需要为奇数

`'ContrastThreshold'` — Contrast threshold range
`0.5` (default) | scalar value

置信度区间，（0，1]

`'UniquenessThreshold'` — Minimum value of uniqueness
`15` (default) | non-negative integer

视差唯一性百分比，视差窗口范围内最低代价是次低代价的(1 + uniquenessRatio/100)倍时，最低代价对应的视差值才是该像素点的视差，否则该像素点的视差为 0 。

当一个像素的唯一性值较低时，计算出的视差不太可靠。

`'DistanceThreshold'` — Maximum distance for left-to-right image checking
`[]` (disabled) (default) | non-negative integer

左右图像的最大阈值

matlab帮助中是这么写的，也就是当设置这个值时就能直接实现左右一致性检验了？目前还不清楚实现出来的效果怎么样

The distance threshold specifies the maximum distance between a point inI1 and the same point found from I2. The function finds the distance and marks the pixel in the following way:

Let p1 be a point in image I1.

Step 1: The function searches for point p1's best match in image I2 (left-to-right check) and finds point p2.

Step 2: The function searches for p2's best match in image I1 (right-to-left check) and finds point p3.

If the search returns a distance between p1 and p3 greater thanDistanceThreshold, the function marks the disparity for the point p1 as unreliable.

`'TextureThreshold'` — Minimum texture threshold
`0.0002` (default) | scalar value

最小纹理阈值，This parameter applies only when you set Method to'BlockMatching'.

程序如下：

clear;
% close all;
I1 = imread('E:left\l1.png');
% I1 = imresize(I1, 0.2);
I2 = imread('E:\right\r1.png');
% I2 = imresize(I2, 0.2);
% figure, imshow(I1);

% figure
% imshow(stereoAnaglyph(I1,I2));
% title('Red-cyan composite view of the stereo images');
%
disparityRange = [0 128];
disparityMap = disparity(rgb2gray(I1),rgb2gray(I2),'BlockSize',...
15,'DisparityRange',disparityRange);
%
%disparityMap = disparity(rgb2gray(I1),rgb2gray(I2));

figure
imshow(disparityMap,disparityRange);
%imshow(disparityMap);
title('Disparity Map');
colormap jet
colorbar