opencv特征点

代码在百度网盘链接里面:

 

链接:https://pan.baidu.com/s/14ennXmtHnZBBDMOEk_jjEQ 
提取码:8888 

1.Harris角点提取

具体原理不细讲,只是运行看一些效果,具体以后要使用到的话再回来看看。

执行效果:

通过调节阈值可以增加或减少角点提取的个数。

2.fast关键点

执行结果:

3.KAZE

执行效果:

特征点寻找所花费时间(ms):19.689875

opencv特征点_第1张图片

 执行效果:

 ./kaze1 1.png 2.png 
extract AKAZE cost = 0.0927352 seconds. 
Gtk-Message: 21:21:13.884: Failed to load module "canberra-gtk-module"
match AKAZE cost = 0.00147178 seconds. 
-- Max dist : 174.000000 
-- Min dist : 9.000000 

opencv特征点_第2张图片

opencv特征点_第3张图片

4.AKAZE

./akaze 
Gtk-Message: 21:24:55.141: Failed to load module "canberra-gtk-module"
特征点寻找所花费时间(ms):49.478649

 opencv特征点_第4张图片

opencv特征点_第5张图片

opencv特征点_第6张图片

opencv特征点_第7张图片

5.BRISK

执行效果:

 ./brisk 
Gtk-Message: 21:29:14.116: Failed to load module "canberra-gtk-module"
BRISK 执行时间为(ms):49.439546

6.shi_Tomasi

执行效果:

./shi_Tomasi 

Gtk-Message: 21:36:33.528: Failed to load module "canberra-gtk-module"
** Number of corners detected: 13
** Number of corners detected: 23
** Number of corners detected: 13
** Number of corners detected: 3
** Number of corners detected: 1533
** Number of corners detected: 10
** Number of corners detected: 20
** Number of corners detected: 21
** Number of corners detected: 25
** Number of corners detected: 26
** Number of corners detected: 28
** Number of corners detected: 29
** Number of corners detected: 34
** Number of corners detected: 35
** Number of corners detected: 49
** Number of corners detected: 59
** Number of corners detected: 62
** Number of corners detected: 66
** Number of corners detected: 75
** Number of corners detected: 79
** Number of corners detected: 83
** Number of corners detected: 84
** Number of corners detected: 85
** Number of corners detected: 91
** Number of corners detected: 92
** Number of corners detected: 93
** Number of corners detected: 95
** Number of corners detected: 98
** Number of corners detected: 99
** Number of corners detected: 100
** Number of corners detected: 103
** Number of corners detected: 105
** Number of corners detected: 109
** Number of corners detected: 110
** Number of corners detected: 111
** Number of corners detected: 112
** Number of corners detected: 113
** Number of corners detected: 114
** Number of corners detected: 118
** Number of corners detected: 117
** Number of corners detected: 118
** Number of corners detected: 120
** Number of corners detected: 119
** Number of corners detected: 106
** Number of corners detected: 105
** Number of corners detected: 97
** Number of corners detected: 87
** Number of corners detected: 74
** Number of corners detected: 57
** Number of corners detected: 40
** Number of corners detected: 36
** Number of corners detected: 30
** Number of corners detected: 27
** Number of corners detected: 26
** Number of corners detected: 22
** Number of corners detected: 20
** Number of corners detected: 16
** Number of corners detected: 13
** Number of corners detected: 9
** Number of corners detected: 7
** Number of corners detected: 6
** Number of corners detected: 5
** Number of corners detected: 4
** Number of corners detected: 3
** Number of corners detected: 2
** Number of corners detected: 1
** Number of corners detected: 1533

 7.GMS

 paper:https://openaccess.thecvf.com/content_cvpr_2017/papers/Bian_GMS_Grid-based_Motion_CVPR_2017_paper.pdfhttps://openaccess.thecvf.com/content_cvpr_2017/papers/Bian_GMS_Grid-based_Motion_CVPR_2017_paper.pdf

理论可以文章看这篇:CVPR2017-图像特征匹配-GMS:基于网格的运动统计的快速且极度鲁棒的图像特征匹配算法 - 知乎

执行结果:

./gms_match_demo 

Get total 613 matches.

opencv特征点_第8张图片

8.ORB

./feature_extraction 1.png 2.png 

use FAST_feature_matches found 231keypoints


ORB Feature detector time: 0.00716571seconds
ORB Descriptor Extractor time: 0.0115678seconds
ORB Descriptor match time: 0.00047365seconds
-- Max dist : 94.000000 
-- Min dist : 4.000000 
ORB-- All matches : 500 
ORB-- filter match : 33 
use ORB_feature_matches found500keypoints


BRISK Feature detector time: 0.0232805seconds
BRISK Descriptor Extractor time: 0.0149624seconds
BRISK Descriptor match time: 0.0020376seconds
BRESK-- Max dist : 183.000000 
BRESK-- Min dist : 16.000000 
BRESK-- All matches : 1117 
BRESK-- filter match : 44 
use BRISK_feature_matches found 1117keypoints


AKAZE Feature detector time: 0.0478079seconds
AKAZE Descriptor Extractor time: 0.0374116seconds
AKAZE Descriptor match time: 0.00125573seconds
AKAZE-- Max dist : 174.000000 
AKAZE-- Min dist : 9.000000 
AKAZE-- All matches : 773 
AKAZE-- filter match : 73 
use akaze_feature_matches found 773keypoints

此外还有SIFT、SURF特征点

你可能感兴趣的:(视觉SLAM十四讲,opencv,计算机视觉,c++)