# FAST Algorithm for Corner Detection. Is it Machine learning?

**URL:** <https://forum.opencv.org/t/fast-algorithm-for-corner-detection-is-it-machine-learning/933>\
**Category:** Uncategorized\
**Created:** [January 19, 2021, 8:19am UTC](https://forum.opencv.org/t/fast-algorithm-for-corner-detection-is-it-machine-learning/933 "2021-01-19T08:19:39Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![MH\_Yip](https://sea2.discourse-cdn.com/flex020/user_avatar/forum.opencv.org/mh_yip/32/610_2.png) [@MH\_Yip](https://forum.opencv.org/u/MH_Yip)\
**Post date:** [January 19, 2021, 8:19am UTC](https://forum.opencv.org/t/fast-algorithm-for-corner-detection-is-it-machine-learning/933/1 "2021-01-19T08:19:39Z")

</div>

Hi everyone, Thank you for reading this post.

I have read the tutorial post of [FAST](https://docs.opencv.org/master/df/d0c/tutorial_py_fast.html). It seems that there is two procedure to detect corners. One is not machine-learned and One is machine-learned (Decision tree). I don’t know which one they use. I went to the [source code](https://github.com/opencv/opencv/blob/master/modules/features2d/src/fast.cpp). It does not seems to load any model.

When I use the `fast.detect(img,None)`, which procedure it use? The non-machine-learned one OR the machine-learned one.

Thank you for your help.