Skip to main content

OpenCV : Threshold (Binary Image)

1. What is binary Image?

It is an image made up of two colors, black and white. It is useful to tracking shape or recognizing object.

result binary images











2. How to make Binary Image?

 It can be made by using Threshold method and depending on the type of flag, you can use various binarizations.

double threshold(InputArray src, OutputArray dst, double thresh, double maxval, int type);

Type :
- THRESH_BINARY : if (src(x,y) > thresh) maxval else 0
- THRESH_BINARY_INV : if (src(x,y) > thresh) 0 else maxval
- THRESH_TRUNC : if(src(x,y) > thresh) threshold else src(x,y)
- THRESH_TOZERO : if (src(x,y) > thresh) src(x,y) else 0

- THRESH_TOZERO_INV : if(src(x,y) > thresh) 0 else src(x,y)

Mat img = imread("./res/text2.jpeg");
Mat gray, binary, binary2, binary3, otsuBinary;
cvtColor(img, gray, CV_RGB2GRAY); // convert to gray color image
// if scalar value is bigger then 128, the value changed to 255(white). else 0(black)
threshold(gray, binary, 128, 255, CV_THRESH_BINARY);
imshow("default binary", binary);


3. Adaptive Threshold

 If original image has shadow or gradients, parts of the image are all filled with white or black. This is because one threshold is applied to the entire image. At this time it is more appropriate to use adaptive threshold.
adaptiveThreshold apply the threshold to smaller image divisions.

void adaptiveThreshold(InputArray src, OutputArray dst, double maxValue, int adaptiveMethod, int thresholdType, int blockSize, double C);

adaptiveMethod :
- ADAPTIVE_THRESH_MEANC (threshold value is the mean of neighbourhood area.)
- ADAPTIVE_THRESH_GAUSSIAN_C (threshold value is the weighted sum of neighbourhood values where weights are a gaussian window)

threshold Type :
- THRESH_BINARY
- THRESH_BINARY_INV


blockSize : 3, 5, 7, 9 ...(odd number)

adaptiveThreshold(gray, binary2, 255, ADAPTIVE_THRESH_MEAN_C, THRESH_BINARY, 15, 5);
adaptiveThreshold(gray, binary3, 255, ADAPTIVE_THRESH_GAUSSIAN_C, THRESH_BINARY, 15, 5);
imshow("adaptive Thresh mean", binary2);
imshow("adaptive Thresh gaussian", binary3);


4. Otsu binary

 How can you determine the threshold value? If image has two peak of histogram, it can be found between two peaks. Otsu binary helps to calculate the value, so that you do not have to enter a threshold.

 
GaussianBlur(gray, gray, Size(7,7), 0); // remove noise by making blur
threshold(gray, otsuBinary, 0, 255, CV_THRESH_BINARY + THRESH_OTSU);
imshow("otsu binary", otsuBinary);
waitKey(0);
Labels:

Comments

Post a Comment

Popular posts from this blog

How to make gaussian blur in SparkAR

 Gaussian blur is result of blurring an image by Gaussian function. We need to use large kernel to make strong blur and this cause the high complexity. And that's why we should use multi-pass which is seperated into horizontal, vertical pass . It can reduce the complexity and make faster. In Spark AR there are many ways to make it. And I wanna introduce one of them using shader files (.sca). You can download it for free from the gumload link below. It is modulated to adjust horizontal, vertical strength easily . Hope this can help you. Loading...
Post a Comment
Read more

OpenGL SL - Translate, roate, and scale

Translate, rotate and scale are most basic element of image effect. If you didn't understand how make effect at each pixel, please read again before post "Show Video and Image" . 1. Translate Translation means each pixel move parallel on the x, y-axis. In other words, if you want to translate 0.25 on the x-axis, a pixel should point to a distance of -0.25 from the original. Translate ex) void mainImage( out vec4 fragColor, in vec2 fragCoord ){     vec2 uv = fragCoord/iResolution.xy;     if (uv.x < 0.5 ){         // translate to x ( -0.25 )         uv.x += 0.25 ;     } else {         // translate to x ( +0.25 )         uv.x -= 0.25 ;     }     fragColor = texture (iChannel0, uv); } result) 2. Scale Scale also has a  similar concept to parallel translation. If you want to zoom in twice, each pixel s...
Post a Comment
Read more

OpenCV : Make sharpen image color

We sometimes take pictures where there is no light, so the pictures come out dark. This means it did not have a wide range of colors. For example, an image with a minimum value of 20 and a maximum value of 100 will be dark. So we need to wide the color range of the image 0 to 255. The full code is as follows. #include <iostream> #include <opencv2/opencv.hpp> using namespace cv ; using namespace std ; void meanVal ( Mat * inputArray ){ double minVal , maxVal ; minMaxIdx( * inputArray , & minVal , & maxVal ); //get min, max value of matrix double ratio = ( maxVal - minVal ) / 255 ; * inputArray = ( * inputArray - minVal ) / ratio ; // wide image range cout << minVal << "," << maxVal << endl; } int main () { Mat img = imread( "./res/woman.png" ); imshow( "origin" , img ); vector < Mat > bgr_img ; split( img , bgr_img ); // split matrix to 3...
Post a Comment
Read more