Main Content

Find Location of Object in Image Using Template Matching

This example uses:

Open Script

Read the reference image and the template image into the MATLAB workspace.

img = imread('stopSignTest.jpg');
tempImg = imread('vipwarningsigns_stop_template.png');

Display the images. The reference image is a scene containing the stop sign board. The template is a low spatial resolution image of the stop sign board.

figure
imshow(img)
title('Reference Image')
figure
imshow(tempImg)
title('Template')

Open the Simulink® model.

modelname = 'ex_blktemplatematching.slx';
open_system(modelname)

The model reads the images by using the Image From Workspace block. To perform template matching, you must first convert the input color images to intensity images by using the Color Space Conversion block. Then, find the location of the template image in the reference image by using the Template Matching block with these parameter values:

  • Match Metric - Sum of absolute differences

  • Output - Best match location

  • Search method - Three-step

The Template Matching block outputs a location in the reference image for which the pixel regions around it best matches with the template image.

Run the model.

out = sim(modelname);

Read the output value.

location = out.simout;

Draw a circle to highlight the region around the best matching pixel location. Display the results.

img = insertShape(img,'circle',[location(1) location(2) 20]);
figure
imshow(img);
hold on
plot(location(1),location(2),'*r')
title('Results of Template Matching')