提取ORB特征点时是不是应该按照灰度图读取图片？

[cowant]

第七章的代码读取图片时都是按彩色图读取的，但是FAST关键点不应该是基于灰度的算法吗？这里不明白为什么不是按照灰度图读取。在pose_estimation_2d2d.cpp的程序里如果按照灰度图读取图片，那么得到的R,t参数与按照彩色读取是不一样的:

• 原来的读取参数为 CV_LOAD_IMAGE_COLOR

 Mat img_1 = imread(argv[1], CV_LOAD_IMAGE_COLOR);
 Mat img_2 = imread(argv[2], CV_LOAD_IMAGE_COLOR);

R is 
[0.9985534106102478, -0.05339308467584758, 0.006345444621108698;
 0.05321959721496264, 0.9982715997131746, 0.02492965459802003;
 -0.007665548311697523, -0.02455588961730239, 0.9996690690694516]
t is 
[-0.8829934995085544;
 -0.05539655431450562;
 0.4661048182498402]

• 修改为cv::IMREAD_GRAYSCALE

 Mat img_1 = imread(argv[1], cv::IMREAD_GRAYSCALE);
 Mat img_2 = imread(argv[2], cv::IMREAD_GRAYSCALE);

R is 
[0.9956586755557854, -0.06056285980694839, 0.07068197650393594;
 0.05735576134499068, 0.9972677254086454, 0.04655534876684903;
 -0.07330837899609778, -0.04229921831748662, 0.9964118915883602]
t is 
[-0.9753492080965818;
 -0.2054788588410324;
 0.08045098405089755]

[gaoxiang12]

OpenCV本身接口支持彩色图像，内部应该会再转换一次，这个不影响。但是彩色图提的点和灰度图提的点会有一些区别，导致R,t计算结果也存在区别。