CN103500439B - Drawing method is beaten based on image processing techniques - Google Patents

Abstract

The invention discloses a drawing method based on image processing technology, which is specifically implemented according to the following steps: step 1, preprocessing the original image; step 2, extracting the outline of the image preprocessed in step 1; step 3, Compress the image after the contour extraction in step 2; step 4, print the compressed image obtained in step 3 on the drawing board through the dotting device, and complete the drawing of the original image. By integrating a variety of image processing technologies, the image is collected, processed and analyzed, and the processed image is printed on the drawing board through the dot device; this method can effectively extract the outline of the image, and compared with manual drawing, The time is greatly shortened, the efficiency is improved, manpower is liberated, the collection, processing and analysis of image information are realized, and the problem of low efficiency of the existing manual painting method is solved.

Description

Drawing method based on image processing technology

technical field

The invention belongs to the technical field of image processing and automatic control, and relates to a drawing method based on image processing technology.

Background technique

Engraving on stone slabs is one of the typical traditional cultures in China, and line drawing on stone inscriptions is a typical representative. Since the Han Dynasty, it has gradually appeared on epitaphs, door lintels, and column foundations. However, with the continuous development of society and the continuous advancement of science and technology, since the appearance of the world's first printer in 1885, printed matter has rapidly spread to all aspects of the world. Although prints account for most of the decorations, slate paintings still occupy a place in decorations as a work of art, and with the improvement of living standards and spiritual levels, human beings are more and more fond of artistic decorations, usually However, artificial slate painting requires manpower to complete, which is time-consuming and labor-intensive, and the work efficiency is low, which is not conducive to the liberation of manpower.

Contents of the invention

The purpose of the present invention is to provide a drawing method based on image processing technology to solve the problem of low efficiency of the existing manual drawing method.

The technical scheme adopted in the present invention is, based on the image processing technology drawing method, specifically implement according to the following steps,

Step 1, preprocessing the original image;

Step 2, performing contour extraction on the image preprocessed in step 1;

Step 3, compressing the image after the contour extraction in step 2;

In step 4, the compressed image obtained in step 3 is printed on the drawing board by the dotting device, and the drawing of the original image is completed.

The present invention is also characterized in that,

The specific method of step 1 is as follows: first, the original image is grayscaled to obtain a grayscale image; then the grayscale image is filtered with a median filter; finally, the filtered image is histogram equalized, that is The preprocessing of the original image is completed.

The specific method of step 2 is to calculate the gradient map of the image preprocessed in step 1; then calculate the distance function image of the gradient map; then calculate the external constraints and internal constraints of the distance function image to obtain the constraint image; Construct the gradient map; finally, through the watershed algorithm, the reconstructed gradient map is segmented to obtain a black and white binary image, that is, the contour extraction of the preprocessed image is completed.

The specific method of image compression in step 3 is,

3.1) Let the length and width of the binarized image obtained in step 2 be m and n respectively, the size compression ratios of length and width be m1 and n1 respectively, p is the side length of the drawing board to be drawn, and p The unit is mm, according to m1=[m/(p/2)]+1 and n1=[n/(p/2)]+1, the size compression ratio of length and width is calculated;

3.2) Let the size of the compressed small matrix be z×z, if m1>n1, then z is m1; if m1<n1, then z is n1; through the calculation of m/z and n/z, the remainders e and f are obtained respectively , the binarized image is cut off e row and f column respectively, and the cropped image is obtained;

3.3) calculate s and t respectively by s=(m-e)/z and t=(n-f)/z, then the cropped image that step 3.2) obtains is divided into width is s, the matrix that is long is t, the matrix of this matrix The elements are then divided into small matrices of z×z;

3.4) If the total number of pixels with a pixel value of 0 in the small matrix is greater than the total number of pixels with a pixel value of 1, set the small matrix to 0, if the total number of pixels with a pixel value of 0 in the small matrix is less than the total number of pixels with a pixel value of 1 , then the small matrix is set to 1, and the new matrix formed at last is the compressed image.

The specific method of painting in step 4 is to scan the compressed image processed in step 3 line by line. If the pixel value of the small matrix is 1, no dots will be placed at the position of the small matrix; if the pixel value of the small matrix is 1 0, then at the position of the small matrix, dots are made on the drawing board by the dotting device until all the small matrices are scanned and dotted, that is, the original image is finished.

The beneficial effect of the present invention is that it integrates multiple image processing techniques, and through the collection, processing and analysis of pictures, the processed image is printed on the drawing board through the dotting device; the method can effectively extract the outline of the image, Moreover, compared with manual drawing, it greatly shortens the time, improves efficiency, liberates manpower, realizes the collection, processing and analysis of image information, and solves the problem of low efficiency of the existing manual drawing method.

detailed description

The present invention will be described in detail below in combination with specific embodiments.

The present invention provides a method for drawing pictures based on image processing technology, which is specifically implemented according to the following steps,

Step 1, preprocessing the original image

First, the original image is grayscaled to obtain a grayscale image. The original image includes color pictures and line drawings. Usually, the grayscaled image contains a lot of salt and pepper noise, and the noise is small; so the grayscale image is used 3× 3. The median filter of the template performs filtering processing; in order to enhance the characteristics of the target object, the filtered image is finally subjected to histogram equalization processing, that is, the preprocessing of the original image is completed.

The preprocessing of the original image is because the image will inevitably be damaged to some extent and various noise pollution in the process of transmission and storage, which will cause the image to lose its essence or deviate from people's needs, so a series of Preprocessing operations to remove the image from the effect.

It can be seen from the experiment that the brightness value between [0.2, 0.8] in the image is mapped to [0, 1], and the image after histogram equalization can get better results; The features of the target object are not obvious, which is not conducive to the contour extraction of the image. In order to extract the contour of the target object more accurately, it is necessary to normalize the image to improve the features caused by lighting and other reasons.

Step 2, perform contour extraction on the image preprocessed in step 1

Calculate the gradient map of the image preprocessed in step 1; then calculate the distance function image of the gradient map; then calculate the external constraints and internal constraints of the distance function image to obtain the constrained image; then reconstruct the gradient map for the constrained image; finally pass the watershed Algorithm, segment the reconstructed gradient image to obtain a black and white binary image, that is, complete the contour extraction of the preprocessed image.

The purpose of contour extraction is to describe the object with the least number of pixels, so it is very important to extract the contour of the image. The essence of image contour extraction is to detect the edge of the image. Commonly used edge detection operators include Roberts operator, Sobel algorithm Operator, Prewitt operator, Canny operator, etc., are usually very sensitive to noise and easy to lose edge information when using these operators to calculate, so the watershed algorithm is chosen in this invention.

Step 3, compress the image after the contour extraction in step 2

3.1) Suppose the length and width of the binarized image obtained in step 2 are m pixels and n pixels respectively, the compression ratios of length and width are m1 and n1 respectively, and p is the edge of the drawing board to be drawn Length, the unit of p is mm; according to m1=[m/(p/2)]+1 and n1=[n/(p/2)]+1, the size compression ratio of length and width is calculated;

3.2) Let the size of the compressed small matrix be z×z, if m1>n1, then z is m1; if m1<n1, then z is n1; through the calculation of m/z and n/z, the remainders e and f are obtained respectively , the binarized image is cut out of e row pixels and f column pixels respectively to obtain the cropped image;

In order to prevent the image from being deformed after compression and the size of the compressed image does not exceed the relative size of the drawing board of the printer, the larger number of m1 and n1 is selected as the length and width of the partition matrix for the size of the partition matrix.

3.3) Calculate s and t respectively by s=(m-e)/z and t=(n-f)/z, then the cropped image obtained in step 3.2) is divided into a matrix with a width of s pixels and a length of t pixels , the elements of the matrix are divided into small z×z matrices;

In order to be able to divide the image to be printed into a matrix of integer multiples, it is necessary to adjust the length and width of the printed image, and calculate the remainders e and f through m/z and n/z, e and f are respectively to be cut and printed The number of pixel rows and pixel columns of the image matrix, since e and f are smaller than z, will not affect the overall content of the image.

3.4) If the total number of pixels with a pixel value of 0 in the small matrix is greater than the total number of pixels with a pixel value of 1, set the small matrix to 0, if the total number of pixels with a pixel value of 0 in the small matrix is less than the total number of pixels with a pixel value of 1 , then the small matrix is set to 1, and the new matrix formed at last is the compressed image.

In step 4, the compressed image obtained in step 3 is printed on the drawing board through the dotting device, and the drawing of the original image is completed.

The specific method of painting is,

Carry out line-by-line scanning of the compressed image processed in step 3, if the pixel value of the small matrix is 1, no dots will be placed at the position of the small matrix; if the pixel value of the small matrix is 0, then at the position of the small matrix , use the dotting device to dot on the drawing board until all the small matrices are scanned and dotted, that is, the original image is finished.

Example

Step 1. First, grayscale a color picture with a size of 420 pixels×406 pixels to obtain its grayscale image; then filter the grayscale image with a median filter of a 3×3 template; The image is processed by histogram equalization, that is, the preprocessing of the original image is completed.

Step 2, calculate the gradient map of the image preprocessed in step 1; then reconstruct the gradient map of the constrained image; finally, divide the reconstructed gradient map through the watershed algorithm to obtain a black and white binary image, that is, complete the Contour extraction of the preprocessed image.

Step 3, compress the image after the contour extraction in step 2

3.1) The length and width of the binarized image obtained in step 2 have dimensions of 420 pixels and 406 pixels respectively, and the side length of the drawing board is 350mm, so the compression ratio of length and width are both 3;

3.2) If the size of the compressed small matrix is z×z, then z=3, e=0, f=1, then the binarized image is cut off by 1 column to obtain the cropped image;

3.3) After cropping, the image is divided into a matrix with a width of 140 pixels and a length of 135 pixels, and the elements of the matrix are further divided into small matrices of 3*3;

3.4) If the total number of pixels with a pixel value of 0 in the small matrix is greater than the total number of pixels with a pixel value of 1, set the small matrix to 0, if the total number of pixels with a pixel value of 0 in the small matrix is less than the total number of pixels with a pixel value of 1 , then the small matrix is set to 1, and the new matrix formed at last is the compressed image.

Step 4, the compressed image processed in step 3 is scanned line by line, if the pixel value of the small matrix is 1, no dots are placed at the position of the small matrix; if the pixel value of the small matrix is 0, then in the small matrix At the location, dots are made on the drawing board by the dotting device until all the small matrices are scanned and dotted, and the original image is finished.

The painting method based on image processing technology integrates multiple image processing technologies. Through the collection, processing and analysis of pictures, the processed image is printed on the drawing board through the dot device; this method can effectively extract the image Compared with manual drawing, it greatly shortens the time, improves efficiency, liberates manpower, realizes the collection, processing and analysis of image information, and solves the problem of low efficiency of the existing manual drawing method.

Claims (2)

1. The method for painting based on image processing technology is characterized in that, it is specifically implemented according to the following steps, Step 1, preprocessing the original image; Step 2, performing contour extraction on the image preprocessed in step 1; Step 3, compressing the image after the contour extraction in step 2; Step 4, the compressed image obtained through step 3 is punched on the drawing board by the dotting device, and the drawing of the original image is completed; The specific method of the step 1 is, First, the original image is grayscaled to obtain a grayscale image; then the grayscale image is filtered with a median filter; finally, the filtered image is processed by histogram equalization, that is, the preprocessing of the original image is completed. ; The specific method of the step 2 is, Calculate the gradient map of the image preprocessed in step 1; then calculate the distance function image of the gradient map; then calculate the external constraints and internal constraints of the distance function image to obtain the constrained image; then reconstruct the gradient map for the constrained image; finally pass the watershed Algorithm, segmenting the reconstructed gradient image to obtain a black and white binary image, that is, the contour extraction of the preprocessed image is completed; The specific method of image compression in the step 3 is, 3.1) Suppose the length and width of the binarized image obtained in step 2 are m pixels and n pixels respectively, the compression ratios of length and width are m1 and n1 respectively, and p is the edge of the drawing board to be drawn Length, the unit of p is millimeter, according to m1=[m/(p/2)]+1 and n1=[n/(p/2)]+1, calculate the size compression ratio of length and width; 3.2) Let the size of the compressed small matrix be z×z, if m1>n1, then z is m1; if m1<n1, then z is n1; through the calculation of m/z and n/z, the remainders e and f are obtained respectively , the binarized image is cut out of e row pixels and f column pixels respectively to obtain the cropped image; 3.3) Calculate s and t respectively by s=(m-e)/z and t=(n-f)/z, then the cropped image obtained in step 3.2) is divided into a matrix with a width of s pixels and a length of t pixels , the elements of the matrix are divided into small z×z matrices; 3.4) If the total number of pixels with a pixel value of 0 in the small matrix is greater than the total number of pixels with a pixel value of 1, set the small matrix to 0, if the total number of pixels with a pixel value of 0 in the small matrix is less than the total number of pixels with a pixel value of 1 , then the small matrix is set to 1, and the new matrix formed at last is the compressed image. 2. the drawing method based on image processing technology as claimed in claim 1, is characterized in that, the specific method of drawing in the described step 4 is, Carry out line-by-line scanning of the compressed image processed in step 3, if the pixel value of the small matrix is 1, no dots will be placed at the position of the small matrix; if the pixel value of the small matrix is 0, then at the position of the small matrix , use the dotting device to dot on the drawing board until all the small matrices are scanned and dotted, that is, the original image is finished.