OpenCV Tutorials

Image effects are created by applying a convolution matrix (also known as kernel) to the image. This is a small matrix that slides through the original image and does a mathematical operation on the pixels for each position of the sliding window. Depending on the convolution matrix you get a different image effect. OpenCV has many of these effects integrated including guassian blur, laplacian and sobel.

Python code to do image effects: gaussian blur, laplacian, sobel:

import numpy as np
import cv2

img = cv2.imread('face.jpg')


# converting to gray scale
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)

# gaussian blur
blur = cv2.GaussianBlur(gray,(3,3),0)

# convolute with proper kernels
laplacian = cv2.Laplacian(img,cv2.CV_64F)
sobelx = cv2.Sobel(img,cv2.CV_64F,1,0,ksize=5)  # x
sobely = cv2.Sobel(img,cv2.CV_64F,0,1,ksize=5)  # y

cv2.imshow("Grayscale", gray)
cv2.imshow("Gaussian blur", blur)
cv2.imshow("Laplacian", laplacian)
cv2.imshow("SobelX", sobelx)
cv2.imshow("SobelY", sobely)

cv2.waitKey(0)

import numpy as np

import cv2

img = cv2.imread('face.jpg')

# converting to gray scale

gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)

# gaussian blur

blur = cv2.GaussianBlur(gray,(3,3),0)

# convolute with proper kernels

laplacian = cv2.Laplacian(img,cv2.CV_64F)

sobelx = cv2.Sobel(img,cv2.CV_64F,1,0,ksize=5) # x

sobely = cv2.Sobel(img,cv2.CV_64F,0,1,ksize=5) # y

cv2.imshow("Grayscale", gray)

cv2.imshow("Gaussian blur", blur)

cv2.imshow("Laplacian", laplacian)

cv2.imshow("SobelX", sobelx)

cv2.imshow("SobelY", sobely)

cv2.waitKey(0)

Output:

Image effects

We can use a technique called template matching to find a template of size $w \times h$ in an input image. We use the function cv2.matchTemplate() which slides through the image and compares the overlapped patches of size $w \times h$ against the input image using a statistical comparison method.

The template image can be anywhere inside the input image but it cannot be rotated, this is a limitation of template matching. A full description of the function and formulas can be found here: http://docs.opencv.org/modules/imgproc/doc/object_detection.html?highlight=matchtemplate#matchtemplate

Output:
In this example I searched for the template image in the red box;

Image source: http://filmfabrikasi.com/wp-content/uploads/2012/08/x-men_3.jpg

Template matching in Python:

import numpy as np
import cv2

image = cv2.imread('wallpaper.jpg')
imageColor = cv2.imread('wallpaper.jpg')
template = cv2.imread('template.jpg')

image = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
template = cv2.cvtColor(template, cv2.COLOR_BGR2GRAY)
h,w = template.shape

result = cv2.matchTemplate(image,template, cv2.TM_CCOEFF)
min_val, max_val, min_loc, max_loc = cv2.minMaxLoc(result)
top_left = max_loc
bottom_right = (top_left[0] + w, top_left[1] + h)

cv2.rectangle(imageColor,top_left, bottom_right,(0,0,255),6)

cv2.imshow("Result", imageColor)
cv2.waitKey(0)

import numpy as np

import cv2

image = cv2.imread('wallpaper.jpg')

imageColor = cv2.imread('wallpaper.jpg')

template = cv2.imread('template.jpg')

image = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)

template = cv2.cvtColor(template, cv2.COLOR_BGR2GRAY)

h,w = template.shape

result = cv2.matchTemplate(image,template, cv2.TM_CCOEFF)

min_val, max_val, min_loc, max_loc = cv2.minMaxLoc(result)

top_left = max_loc

bottom_right = (top_left[0] + w, top_left[1] + h)

cv2.rectangle(imageColor,top_left, bottom_right,(0,0,255),6)

cv2.imshow("Result", imageColor)

cv2.waitKey(0)

Object detection with template matching in C++

#include <opencv2>
#include <opencv2>
#include <opencv2>
#include <opencv2>
#include <iostream>
#include <stdio.h>

using namespace cv;
using namespace std;

/// Global Variables
Mat img; Mat templ; Mat result;
const char* image_window = "Source Image";
const char* result_window = "Result window";

int match_method;

/// Function Headers
void MatchingMethod( int, void* );

/**
* @function main
*/
int main( int, char** argv )
{
  /// Load image and template
  img = imread( "wallpaper.jpg", 1 );
  templ = imread( "logo.jpg", 1 );

  // Match
  MatchingMethod( 0, 0 );

  waitKey(0);
  return 0;
}

/**
* @function MatchingMethod
* @brief Trackbar callback
*/
void MatchingMethod( int, void* )
{
  /// Source image to display
  Mat img_display;
  img.copyTo( img_display );

  /// Create the result matrix
  int result_cols = img.cols - templ.cols + 1;
  int result_rows = img.rows - templ.rows + 1;

  result.create( result_cols, result_rows, CV_32FC1 );

  /// Do the Matching and Normalize
  matchTemplate( img, templ, result, match_method );
  normalize( result, result, 0, 1, NORM_MINMAX, -1, Mat() );

  /// Localizing the best match with minMaxLoc
  double minVal; double maxVal; Point minLoc; Point maxLoc;
  Point matchLoc;

  minMaxLoc( result, &minVal, &maxVal, &minLoc, &maxLoc, Mat() );


  /// For SQDIFF and SQDIFF_NORMED, the best matches are lower values. For all the other methods, the higher the better
  if( match_method == TM_SQDIFF || match_method == TM_SQDIFF_NORMED )
    { matchLoc = minLoc; }
  else
    { matchLoc = maxLoc; }

  /// Show me what you got
  rectangle( img, matchLoc, Point( matchLoc.x + templ.cols , matchLoc.y + templ.rows ), Scalar(0,0,255), 4, 8, 0 );
  imshow( result_window, img );

  return;
}</stdio.h></iostream></opencv2></opencv2></opencv2></opencv2>

#include

using namespace cv;

using namespace std;

/// Global Variables

Mat img; Mat templ; Mat result;

const char* image_window = "Source Image";

const char* result_window = "Result window";

int match_method;

/// Function Headers

void MatchingMethod( int, void* );

/**

* @function main

int main( int, char** argv )

{

/// Load image and template

img = imread( "wallpaper.jpg", 1 );

templ = imread( "logo.jpg", 1 );

// Match

MatchingMethod( 0, 0 );

waitKey(0);

return 0;

}

/**

* @function MatchingMethod

* @brief Trackbar callback

void MatchingMethod( int, void* )

{

/// Source image to display

Mat img_display;

img.copyTo( img_display );

/// Create the result matrix

int result_cols = img.cols - templ.cols + 1;

int result_rows = img.rows - templ.rows + 1;

result.create( result_cols, result_rows, CV_32FC1 );

/// Do the Matching and Normalize

matchTemplate( img, templ, result, match_method );

normalize( result, result, 0, 1, NORM_MINMAX, -1, Mat() );

/// Localizing the best match with minMaxLoc

double minVal; double maxVal; Point minLoc; Point maxLoc;

Point matchLoc;

minMaxLoc( result, &minVal, &maxVal, &minLoc, &maxLoc, Mat() );

/// For SQDIFF and SQDIFF_NORMED, the best matches are lower values. For all the other methods, the higher the better

if( match_method == TM_SQDIFF || match_method == TM_SQDIFF_NORMED )

{ matchLoc = minLoc; }

else

{ matchLoc = maxLoc; }

/// Show me what you got

rectangle( img, matchLoc, Point( matchLoc.x + templ.cols , matchLoc.y + templ.rows ), Scalar(0,0,255), 4, 8, 0 );

imshow( result_window, img );

return;

}

Image effects with Python

Object detection using Template Matching