detecting object code?

8 visualizaciones (últimos 30 días)
niloofar zolghadr
niloofar zolghadr el 28 de Jun. de 2019
Comentada: niloofar zolghadr el 29 de Jun. de 2019
Hi, I have watched this video https://www.youtube.com/watch?v=ddXvs1Wp95A in you tube . and I followed all the codes that he described for detecting object but one part,describing rectangle, (because this line was long and not readable in video) was not readable for me .I know it may seem easy but my profession field is architecture and I am absolutely beginner in this field so If it is possible I ask for the code in (line 14 in the function findtemplate) which is about describing a rectangular.
Thank you so much

Respuesta aceptada

Image Analyst
Image Analyst el 28 de Jun. de 2019
Looks like he's using normalized cross correlation as a template finding technique. See my attached demo.
  3 comentarios
Image Analyst
Image Analyst el 28 de Jun. de 2019
You need to pass in [x, y, width, height] which is what the BoundingBox measurement gives.:
r=regionprops(bwlabel(bw), 'BoundingBox');
hold on;
for k = 1 : length( r )
thisBB = r(k).BoundingBox;
rectangle('Position', thisBB);
end
niloofar zolghadr
niloofar zolghadr el 29 de Jun. de 2019
You are the best.Thank you.

Iniciar sesión para comentar.

Más respuestas (1)

niloofar zolghadr
niloofar zolghadr el 29 de Jun. de 2019
Editada: niloofar zolghadr el 29 de Jun. de 2019
@ Image Analyst oh, you have answered my question actually there is no error but I dont know why it does not detect my template window .I mean I have scattered bounding box all over the image .Do you know what is wrong with this algorithm?(detectobj and findtemplate are algorithms and hibbard is the image and window is template) By the way I think you are so expert is there any way that I can contact you easier forexample email or Telegram ? because my field is architecture and I am not so familiar with image processing.If there is anyway which I can contact you easier please let me know ...thank you so much
  2 comentarios
Image Analyst
Image Analyst el 29 de Jun. de 2019
You may need to lower your threshold or try a different technique, like deep learning.
niloofar zolghadr
niloofar zolghadr el 29 de Jun. de 2019
@ Image Analyst I have tried different threshold before but it was useless.the result of algorithm is so disappointing. I could not post my image in that time because I have uploaded more than 10 images.but now I have posted it .If it is possible for you can you check those algorithms that I have sent with these images?and by the way I am trying to seperate the windows of this image if you know any code I would be so thankful if you share it here.
Thank you again

Iniciar sesión para comentar.

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by