JPH04340177A - Detection of edge - Google Patents

Abstract

PURPOSE:To set up a proper threshold for binarization by applying the comparatively small quantity of operation to a binarized image. CONSTITUTION:A straight line to be a candidate including edges for a binarized image between them is detected (step 101). A pair of windows holding the straight line are set up and the number of black picture elements included in each window is counted (step 102). Whether the straight line includes an edge or not is judged based upon the number of black picture elements in respective windows (steps 103, 104). At the time of judging no inclusion of an edge, a threshold for binarizing is changed (steps 106, 107). A new binarized image is formed based upon the changed threshold and the same processing is repeated for the newly formed binary image.

Description

[Detailed description of the invention]

0001

[Industrial Application Field] The present invention utilizes the density difference between the object and the background to determine the edge of the boundary between the object and the background based on a binary image that is binarized using a predetermined threshold value. The present invention relates to an edge detection method for detecting.

0002

[Prior Art] Conventionally, as an edge detection method for detecting an edge that is a boundary line between an object and a background, a differential value for each pixel of a grayscale image is determined, and the differential value is binarized based on the determined differential value. A method for correcting the threshold value for performing
It is disclosed in Japanese Patent No. 9882.

This method is based on the knowledge that when the density difference between the object and the background is small, the maximum value of the differential value is small, and when the density difference is large, the maximum value of the differential value is also large. In other words, by setting the threshold for binarization higher as the maximum value of the differential value becomes larger,
The aim is to obtain an appropriate binary image that makes it easy to distinguish between the object and the background.

0004

The above method has the advantage that an appropriate threshold value can be set depending on the contrast between the object and the background. However, at least for areas that include edges, it is necessary to find differential values for all pixels, so it is necessary to perform a large amount of calculations on grayscale images in which each pixel has multiple values, and the amount of calculations is extremely large. There was a problem.

[0005] The present invention aims to solve the above-mentioned problems, and is capable of processing binary images with a relatively small amount of calculations.
The present invention aims to provide an edge detection method that allows setting of an appropriate threshold value for value conversion.

[0006]

[Means for Solving the Problems] In order to achieve the above object, the present invention provides an edge detection method for detecting an edge that is a boundary line between an object and a background based on a binary image. After detecting a straight line, set a pair of windows across this straight line, determine whether the straight line includes an edge based on the number of white pixels or black pixels included in each window, and detect the edge. If it is determined that it is not included, the threshold value for binarization is changed to create a binary image again, and the above process is performed on the newly created binary image.

[0007]

[Operation] According to the above method, after detecting a straight line that is a candidate for containing an edge in a binary image, by detecting the distribution of white pixels or black pixels on both sides of this straight line, the straight line includes an edge. If the threshold is set appropriately and the above straight line includes an edge, there will be a large difference in the number of white pixels or black pixels across the above straight line. , it is possible to accurately determine whether the straight line includes an edge. Additionally, if it is determined that no edges are included, the threshold value is changed and the binary image is created again, and the above process is then repeated, so the threshold value can be changed until the threshold value reaches an appropriate value. Therefore, it is possible to converge to an appropriate threshold value using a relatively small amount of calculation, which involves only calculations for binary images.

[0008]

[Example] As shown in FIG. 2, an image containing an object is C
The image is captured by an image input device 1 such as a CD camera. The image signal, which is an analog signal output from the image input device 1, is converted into a digital signal through the A/D conversion circuit 2, and then binarized in the binarization processing section 3 and stored in the binary image memory 4. be done. The threshold value in the binarization processing section 3 is set by the threshold value setting section 5.

In the present invention, an appropriate threshold value in the binarization processing unit 3 is set based on the binary image stored in the binary image memory 4. As shown in FIG. A candidate straight line (hereinafter referred to as an edge candidate) including an edge that is a boundary line between the image and the background is detected (steps 100 and 101). That is, in the edge detection section 6,
Regarding the binary image stored in the binary image memory 4, FIG.
As shown in FIG. 3, a mask M is set in advance as a region in which edges are to be detected, and edge candidates are detected for pixels P in the mask M. For example, to detect edge candidates in the vertical direction, a search is performed sequentially in the horizontal direction, and first, the pixel values of the leftmost vertical line L1 in the mask M are sequentially determined from the top downward. . Here, set the value of the white pixel (white circle) to 0, and the value of the black pixel (
When two or more lines L1 to L5 in which three or more black pixels are connected are connected (in this case, lines L3 and L4 satisfy the condition), the value of The leftmost line L of the lines L3 and L4
3 is an edge candidate. To detect edge candidates in the horizontal direction, sequential searches may be performed in the vertical direction. Also, here we assume that the inside of the object becomes a black pixel, but
If the inside of the object is a white pixel, edge candidates may be detected by connecting white pixels. Note that the size of the mask M, the number of connected pixels P, the number of connected lines L1 to L5, etc. are not limited to those in the embodiment.

After detecting the edge candidate E as described above, the counting section 7 sets a pair of windows W1 and W2 in the mask M with the edge candidate E sandwiched therebetween, as shown in FIG. Both windows W1 and W2 are set to have the same shape and include the same number of pixels P.
1, the number of white pixels or black pixels in W2 is counted (step 102 in FIG. 1), respectively. For example, if the inside of the object is a black pixel, each window W1, W
2, the number of black pixels on the object side with respect to the edge candidate E will be significantly greater than the number of black pixels on the background side. By finding the difference or ratio of the number of black pixels in , it is possible to determine whether edge candidate E is a true edge.

That is, the determining unit 8 compares the difference or ratio of the number of black pixels or white pixels in each window W1, W2 counted by the counting unit 7 with a predetermined reference value, and determines the difference or ratio between the numbers and the reference value. Based on the relationship, it is determined whether the edge candidate E is a true edge. Here, the number of black pixels in windows W1 and W2 is w1,
w2, and two types of reference values α and β are set. As shown in steps 103 and 104 in FIG. 1, if w2≧α, it is determined that the light amount is sufficient, and if w2−w1≧β, the contrast is also large, so the detected edge candidate E is a true edge. It is determined that (step 105). On the other hand, if w2<α, it is determined that the amount of light is insufficient, and the threshold setting unit 5 is controlled to lower the threshold for binarization (step 106). 3 is changed, imaging is performed again, and the above process is repeated. Furthermore, even if w2≧α holds true, if w2−w1<β, it is determined that the amount of light is too large, and the threshold value setting unit 5 is controlled to increase the threshold value (step 107), and imaging is performed again. and repeat the above process. Here, the binarization processing section 3
The threshold value is set in multiple steps with a predetermined width, and the threshold value is raised and lowered one step at a time.

[0012] In the above embodiment, when the threshold value is changed, the image input device 1 performs imaging again, but the captured image is stored in the grayscale image memory, and when the threshold value is changed, the grayscale image is Binarization may be applied to the grayscale image stored in the image memory. Furthermore, the binarization processing section 3, threshold value setting section 5, edge detection section 6, counting section 7, and determination section 8 can be easily realized using a microcomputer.

[0013]

Effects of the Invention As described above, the present invention detects a straight line that is a candidate including an edge in a binary image, and then detects the distribution of white pixels or black pixels on both sides of this straight line. is determining whether or not it includes an edge, and if the threshold is set appropriately and the above straight line includes an edge, there will be a large difference in the number of white pixels or black pixels across the above straight line. By this occurrence, it is possible to accurately determine whether the above-mentioned straight line includes an edge. Additionally, if it is determined that no edges are included, the threshold value is changed and the binary image is created again, and the above process is then repeated, so the threshold value can be changed until the threshold value reaches an appropriate value. This has the advantage that an appropriate threshold value can be determined with a relatively small amount of calculations that involve only calculations for binary images. In other words, the amount of light around the edge can be determined and the threshold can be corrected for changes in contrast due to changes in the environment. It can be detected.

[Brief explanation of drawings]

FIG. 1 is an explanatory diagram of the operation of the embodiment.

FIG. 2 is a block diagram of an embodiment.

FIG. 3 is an explanatory diagram of the operation of the edge detection section of the embodiment.

FIG. 4 is an explanatory diagram of the operation of the counting section of the embodiment.

[Explanation of symbols]

1 Image input device 2 A/D conversion circuit 3 Binarization processing section 4 Binary image memory 5 Threshold setting section 6 Edge detection section 7 Counting section 8 Judgment section

Claims (1)

[Claims] Claim 1: In an edge detection method that detects an edge that is a boundary line between an object and a background based on a binary image, after detecting a straight line that is a candidate including an edge, a pair of windows is formed with this straight line in between. is set, and it is determined whether the above straight line includes an edge based on the number of white pixels or black pixels included in each window. If it is determined that the straight line does not include an edge, the threshold for binarization is changed. An edge detection method characterized in that the binary image is re-created by the user, and the above-mentioned processing is performed on the newly created binary image.