# Detect corners of different scales

**URL:** <https://discourse.julialang.org/t/detect-corners-of-different-scales/6740>\
**Category:** General Usage\
**Tags:** machinevision\
**Created:** [October 28, 2017, 12:11pm UTC](https://discourse.julialang.org/t/detect-corners-of-different-scales/6740 "2017-10-28T12:11:48Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![yakir12](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/yakir12/32/297_2.png) [@yakir12](https://discourse.julialang.org/u/yakir12)\
**Post date:** [October 28, 2017, 12:11pm UTC](https://discourse.julialang.org/t/detect-corners-of-different-scales/6740/1 "2017-10-28T12:11:48Z")

</div>

Are there any settings in `Images.imcorner` that “tune” the scale or radius of the corners being detected? I tried setting `blockSize` [here](https://github.com/JuliaImages/Images.jl/blob/9ee8146a52aac267cf5bbe80dd82840619f0c543/src/corner.jl#L169) but I can’t get it to work…

For example, I’d like to detect the corners of the large squares in the following image, not the corners of all the small notches:

 ![screenshot_2017-10-28_14:03:53](https://global.discourse-cdn.com/julialang/original/3X/3/2/32470e771622297abfe38677bdf925ff771e6507.png)

Possible solutions are:

- reduce the image size: but that would reduce the accuracy of the corners’ locations.
- try not to get those notches in the segmentation process, but that seems too sensitive to changes in the original images.

This relates to [my previous question](https://discourse.julialang.org/t/detecting-crosses-in-grids/6720)…

Thanks!