# \[Project\] Fast and rotation invariant template matching **URL:** \ **Category:** Uncategorized\ **Created:** [May 24, 2022, 12:36pm UTC](https://forum.opencv.org/t/project-fast-and-rotation-invariant-template-matching/8895 "2022-05-24T12:36:03Z")\ **Posts on this page:** 10\ **Page:** 1
**Author:** ![den8017981](https://sea2.discourse-cdn.com/flex020/user_avatar/forum.opencv.org/den8017981/32/5276_2.png) [@den8017981](https://forum.opencv.org/u/den8017981)\ **Post date:** [May 24, 2022, 12:36pm UTC](https://forum.opencv.org/t/project-fast-and-rotation-invariant-template-matching/8895/1 "2022-05-24T12:36:03Z")
Lots of people have complained about `cv::matchTemplate ()` because it is not a rotation invariant function, which means it is not practical. Therefore, I improved and combined this function with a search strategy, and create a project.Also, this project passed some cases in real world assembly lines. [Details and the source code of this project](https://github.com/DennisLiu1993/Fastest_Image_Pattern_Matching) The formula of similarity is as followed: ![image](https://us1.discourse-cdn.com/flex020/uploads/opencv/original/2X/4/4d1b7bd86224eed635dd927b746e8b59e94da443.jpeg) Some practical tests: ![image](https://us1.discourse-cdn.com/flex020/uploads/opencv/original/2X/6/6a13d6461fb079c852c03534f8cc0182266a730b.png) ![image](https://us1.discourse-cdn.com/flex020/uploads/opencv/original/2X/4/48732586f51752d07b76866a600f6e2150732082.jpeg) ![image](https://us1.discourse-cdn.com/flex020/uploads/opencv/original/2X/a/ab9f0b72d5bb15022c8baec84d99ea815f30e6ec.png) [It can also used as OCR](https://www.youtube.com/watch?v=lM0NK6xVNfg) If you think my project useful and practical, give me a star on [github](https://github.com/DennisLiu1993/Fastest_Image_Pattern_Matching) Current Status: I am doing researching on scale invariant version, perhaps next month can get a positive result. If you have any test data that can verify scale issue, please contact me. ---
**Author:** ![crackwitz](https://sea2.discourse-cdn.com/flex020/user_avatar/forum.opencv.org/crackwitz/32/14_2.png) [@crackwitz](https://forum.opencv.org/u/crackwitz)\ **Post date:** [May 24, 2022, 5:51pm UTC](https://forum.opencv.org/t/project-fast-and-rotation-invariant-template-matching/8895/2 "2022-05-24T17:51:27Z")
what does the code even do with the pyramid? all I can tell from the code is that it tries `matchTemplate` on a number of rotated versions of the template. that shouldn’t be fast at all, considering a single `matchTemplate` call on any usefully large image takes a while. and **please** stop posting that stuff to every thread you can find. ---
**Author:** ![den8017981](https://sea2.discourse-cdn.com/flex020/user_avatar/forum.opencv.org/den8017981/32/5276_2.png) [@den8017981](https://forum.opencv.org/u/den8017981)\ **Post date:** [May 25, 2022, 1:18am UTC](https://forum.opencv.org/t/project-fast-and-rotation-invariant-template-matching/8895/3 "2022-05-25T01:18:46Z")
1. what does the code even do with the pyramid? **speed up the whole process by reducing possible searching area** 2. on a number of rotated versions of the template. **not the rotated templates but the rotated source, which is the key element of this method** 3. that shouldn’t be fast at all, considering a single `matchTemplate` call on any usefully large image takes a while. **I just don’t get it. I made some tests in comparison with commercial image processing libraries, and the results demonstrated its speed.** **Of course, this method will become slow when image size increases** 4. stop posting **OK** ---
**Author:** ![berak](https://avatars.discourse-cdn.com/v4/letter/b/85f322/32.png) [@berak](https://forum.opencv.org/u/berak)\ **Post date:** [May 25, 2022, 10:39am UTC](https://forum.opencv.org/t/project-fast-and-rotation-invariant-template-matching/8895/4 "2022-05-25T10:39:40Z")
how hard would it be, to seperate the algorithm from the gui app ? as it is now, only VS users can even build it ---
**Author:** ![den8017981](https://sea2.discourse-cdn.com/flex020/user_avatar/forum.opencv.org/den8017981/32/5276_2.png) [@den8017981](https://forum.opencv.org/u/den8017981)\ **Post date:** [May 25, 2022, 11:40am UTC](https://forum.opencv.org/t/project-fast-and-rotation-invariant-template-matching/8895/5 "2022-05-25T11:40:51Z")
1. separating from gui is easy. only the part of showing results will be affected 2. vector can be replaced with some containers in other language 3. Uncertainty: direct access to Mat’s elements ---
**Author:** ![den8017981](https://sea2.discourse-cdn.com/flex020/user_avatar/forum.opencv.org/den8017981/32/5276_2.png) [@den8017981](https://forum.opencv.org/u/den8017981)\ **Post date:** [June 3, 2022, 4:00pm UTC](https://forum.opencv.org/t/project-fast-and-rotation-invariant-template-matching/8895/6 "2022-06-03T16:00:15Z")
20220602 update: Using SIMD version of image convolution to replace matchTemplate () with parameter CV\_TM\_CCORR For many cases, it significantly speed up the process. (For small template, it is slight slow) See [this commit](https://github.com/DennisLiu1993/Fastest_Image_Pattern_Matching/commit/359696dd1a9150715864020c8f3883fbda630862) ---
**Author:** ![FahNos](https://sea2.discourse-cdn.com/flex020/user_avatar/forum.opencv.org/fahnos/32/6303_2.png) [@FahNos](https://forum.opencv.org/u/FahNos)\ **Post date:** [October 19, 2022, 8:15am UTC](https://forum.opencv.org/t/project-fast-and-rotation-invariant-template-matching/8895/7 "2022-10-19T08:15:31Z")
@den8017981 Could you explain about this formular: double dAngleStep = atan (2.0 / max (pTemplData-\>vecPyramid[iTopLayer].cols, pTemplData-\>vecPyramid[iTopLayer].rows)) \* R2D; Thank you, ---
**Author:** ![den8017981](https://sea2.discourse-cdn.com/flex020/user_avatar/forum.opencv.org/den8017981/32/5276_2.png) [@den8017981](https://forum.opencv.org/u/den8017981)\ **Post date:** [October 19, 2022, 8:28am UTC](https://forum.opencv.org/t/project-fast-and-rotation-invariant-template-matching/8895/8 "2022-10-19T08:28:00Z")
![image](https://us1.discourse-cdn.com/flex020/uploads/opencv/original/2X/a/a1ca6296e837394c39b3415ce8fcae4d622319c9.png) This image can explain, just a little math. ---
**Author:** ![FahNos](https://sea2.discourse-cdn.com/flex020/user_avatar/forum.opencv.org/fahnos/32/6303_2.png) [@FahNos](https://forum.opencv.org/u/FahNos)\ **Post date:** [October 24, 2022, 7:12am UTC](https://forum.opencv.org/t/project-fast-and-rotation-invariant-template-matching/8895/9 "2022-10-24T07:12:43Z")
@den8017981 Could you explain more about this function: void CMatchToolDlg::MatchTemplate() - SIMD: inline int IM\_Conv\_SIMD() Thank you ---
**Author:** ![den8017981](https://sea2.discourse-cdn.com/flex020/user_avatar/forum.opencv.org/den8017981/32/5276_2.png) [@den8017981](https://forum.opencv.org/u/den8017981)\ **Post date:** [October 25, 2022, 3:33am UTC](https://forum.opencv.org/t/project-fast-and-rotation-invariant-template-matching/8895/10 "2022-10-25T03:33:56Z")
@ use this [document](https://www.intel.com/content/dam/develop/external/us/en/documents/18072-347603.pdf) provided by intel as a reference