JPS60179881A - Recognizing method of approximately circular outline - Google Patents

Abstract

PURPOSE:To shorten a processing time by limiting the number of times of scanning and a scanning range without tracking the whole periphery of an outline, checking the presence or absence of candidate points of the outline and recognizing the existence of an approximately circular outline. CONSTITUTION:Eight directions 0 (+X direction) - 7pi/4 around a center position candidate point C(X, Y) are previously set up as scanning directions. At the scanning of each radius direction, the differential maximum value Dmax is set up to ''0'', and while scanning any one radius direction out of said eight directions, the differential value D' of picture data is calculated. If the differential value D' is larger than the differential maximum value Dmax, the Dmax is rewritten by the D'. If the Dmax exceeds a threshold D after ending the scanning, the number (n) of candidate outline points is increased by ''1''. Hereinafter, the presence or absence of candidate outline points is checked in eight scanning directions, and when the number (n) of all the candidate outline points exceeds a previously set number No (6), it is decided that the outline of a circular substance exists in a doughnut-like area.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a substantially circular contour-like recognition method for recognizing substantially circular contours in television images at high speed.

Conventionally, the area method and the contour detection method have been used as methods for calculating the value of an object's characteristic parameters from a television image (multilevel image).

The region method divides an image into subimages (regions) of approximately equal brightness based on the assumption that each surface that makes up an object has similar brightness, and recognizes the object based on the continuity of each region. This is the way to do it. This method is effective for objects composed of flat surfaces, but becomes difficult to process when a curved surface is included.

Furthermore, since the entire image data is handled, the amount of data to be processed becomes enormous, and high speed cannot be expected.

On the other hand, the 1-contour extraction method focuses on line points (edges) on each surface that makes up the object, and extracts points where the brightness changes suddenly in the image as line points, and connects the line points. By doing so, it is converted into a line drawing (line drawing). This method is 1
This method attempts to detect lines in an image, and compared to the above-mentioned area method that attempts to detect surfaces, it requires less information when considering the detection process and its continuity, and can be expected to speed up processing. .

The g-identification procedure for circular objects using this contour extraction method is shown in the gi diagram (
To explain with reference to a) to (d), the original image taken with a television camera (Fig. (Figure 1(b)). Next, perform similar differentiation processing on each pixel in the vicinity of this point, and consider the pixel with the largest differential value to be a point continuous to the contour candidate point described above.By repeating this operation, continuous contour points can be obtained. (
When the contour points are closed (FIG. 1(d)), they are regarded as one object.

However, in such conventional contour edge extraction methods, the following factors hinder the tracking of contour candidate points: α) Blooming due to metallic luster (see Figure 2 (a)) ■) Overlapping objects (see Figure 2 Φ) ) (3) Blurred images or images due to rust, dirt, etc. on the surface of the object (4) Image disturbances due to electrical noise, etc.
As a result, there is a problem of missing objects that should be present. Also, in order to solve these problems, Ia! The algorithms for recognition become complex and real-time processing becomes almost impossible.

The present invention has been made in view of the above-mentioned circumstances, and an object of the present invention is to provide a method for recognizing a substantially circular contour line that can recognize the presence or absence of a desired contour line (substantially circular contour line) at extremely high speed.

According to this invention, a center position candidate point of a substantially circular contour is searched for, and based on the searched center position candidate point and radius, a donut shape in which the substantially circular contour is likely to exist is determined from all images. A plurality of images of the region are scanned radially from the center position candidate point, and the presence or absence of contour candidate points is checked in each scan. When the number of contour candidate points is equal to or greater than a preset number, a substantially circular contour line is formed. It is recognized that the donut-shaped area exists in the image.

The present invention will be described in detail below with reference to the accompanying drawings.

FIG. 3 is a schematic configuration diagram showing an example of an apparatus for carrying out the method for recognizing a substantially circular contour line according to the present invention, and shows a case where the ITV camera 2 is photographing a circular object 1 as a detection target object. There is. The industrial TV camera 2 photographs the circular object 1 in a predetermined field of view and outputs a video composite signal containing brightness and darkness signals of the input image to the synchronization separation circuit 3 and the A/D converter 4. The synchronization separation circuit 3 separates a synchronization signal from the input video composite signal, specifies the address of a random access memory array (RAM array) 5 based on this synchronization signal, and the A/D converter 4 The bright/dark signal of the video composite signal is converted into image data with a bright/dark state of 16 gradations, and this is written at the specified address position. In this way R.A.
For M array 5.

One screen worth of image data showing the brightness of the original image shown in FIG. 4 is saved. . In addition, X and Y of RAM array 5
Any image data can be extracted by specifying the address.

On the other hand, the memory 6 stores a main program etc. for implementing the method of the present invention, and a central processing unit (CPU)
7 executes image processing of the image data in the RAM array 5 based on the main program contents.

Next, the processing procedure of the CPU 7 will be explained according to the flowchart shown in FIG. 5 and with reference to FIGS. 6(a) to 6(e).

First, initial settings are made for the diameter of the circular object per threshold value of the differential value, the number of scans in the radial direction Ns 46, and the preset number No of contour candidate points. Here, the threshold value is a threshold value for determining contour candidate points where the brightness suddenly changes in the image data, and in this embodiment, the number of scans Ns is set to 8. Preset number N
o is set to 6.

When the initial settings are completed, a search for a candidate center position of the circular object is performed from the current image data in the RAM array 5 (FIG. 6(a)). This search for the center position candidate point is performed by differentially processing the image data in the RAM array 5 in the This is done based on the position of the candidate point.

That is, the number n of one contour candidate point is set to 0.

While scanning the current image in the X direction, calculate the differential value DI of the image data.
Calculate. It is determined whether or not this differential value DI exceeds a threshold value. If it does, the coordinate position is stored and n is incremented by 1. In this way, when n becomes 2 or more, the distance between any two points stored above is 1111L.
' (FIG. 6(b)), and it is determined whether this IE separation L approximates the initially set diameter. Then, when the distance L / approximates the diameter, a center position candidate point c (x, y) is calculated from the coordinate positions of the two points at that time (
Figure 6(C)). Note that the method of searching for the center position candidate point is not limited to this embodiment; for example, there are various methods such as finding three or more contour candidate points and calculating the center position candidate point from a circle passing through these points. Conceivable.

Next, a candidate contour point of the circular object is searched based on the candidate center position of the circular object, thereby checking the presence or absence of a contour line (circular object).

First, the number n of one contour candidate point is set to O, and scanning is performed in a preset direction in the radial direction from the center position candidate point. At this time, since the radius R (=L/2) of one circular object is known, the scanning area is the minimum allowable radius (R - ΔB)
and a circle with the maximum allowable radius (几+ΔR). Furthermore, the preset scanning direction is set to 0(
+X direction), E/4.1r/2, 31r/4.7+-
, 5/4.

31L:/2, and 7r/4 directions, for a total of 8 directions (see FIG. 6(d)).

Now, when scanning in each radial direction, the maximum differential value DlnaX
Set to 0. Next, while scanning in the radial direction of any of the eight directions, calculate the differential value DI of the image data)
IN, put out. This differential value D' is the maximum differential value Drnax
Determine whether or not □゛ is also larger, and if it is larger, rewrite the differential 1iiD' to the maximum differential value Dr11ax,
The maximum differential value within the scanning range is updated so that it is stored in the maximum differential value Drnax. After the scanning is completed, the maximum differential Il! It is determined whether Dmax exceeds a threshold, and if it does, the number n of contour candidate points is incremented by one. In this way, the presence or absence of contour candidate points is checked for each of the eight scanning directions, and if the total number n of contour candidate points is greater than or equal to the preset number No (6) of contour candidate points, the contour of the circular object is shaped like the donut. It is determined that the object exists within the area of . When the total number n of contour candidate points is less than the preset number N0 of contour candidate points, the search for the center position candidate point of the circular object is performed again.

Finally, when the existence of an outline of a circular object is recognized, a circle is approximated from the n outline candidate points at that time, and the coordinates of the center position of one circle and, if necessary, the diameter are calculated.
i'U(e)), the image processing ends.

Note that when a circular object is photographed with an ITV camera, if the circular object is largely deviated from the optical axis of the ITV camera 2, its outline will not be a perfect circle but an ellipse. Even in this case, the method of the present invention can be applied. Applicable.

In addition, not only circular objects but also at least approximately circular contours (
The method of the present invention can be applied to any object that can be used to reduce objects (ellipses or polygons that are close to circles).

Furthermore, the number of times of scanning N5Jt in the radial direction and the preset number NO serving as the threshold are not limited to this embodiment and can be set in various ways.

As explained above, according to the present invention, the presence or absence of contour candidate points is checked by limiting the number of scans S and the short range without tracing the entire circumference of the contour, thereby recognizing the existence of a substantially circular contour. Therefore, processing time can be shortened. Furthermore, since the rounding scanning range is limited when one outline candidate point is extracted, there is an advantage that it is less susceptible to the effects of dirt, rust, light, or image noise on the surface of the object.

[Brief explanation of the drawing]

Figures 1 (a) to (d) are diagrams used to explain the procedure for object recognition using the conventional contour extraction method, and Figure 2 (a)
8 and (b) are diagrams each showing an example of a factor that inhibits tracking of a conventional e4 curve, and FIG. Figure 4 is a diagram showing the brightness of the image data stored in the RAM array in Figure 3, and Figure 5 is the center of the third factor! A
FIGS. 6(a) to 6(e) are diagrams used to explain the flowchart of FIG. 5. FIGS.・l...Circular object, 2...I
Tv//17 Mera, 3... Synchronization separation circuit, 4... A
/D' converter, 5... RAM array, 6... Memory, 7... Central processing unit (CPU). (0n (b) Figure 2 (o) ” (b) Figure 5 Figure 6

Claims (4)

[Claims] (1) Search for a center position candidate point of a substantially circular contour from an input image of a predetermined field of view where the substantially circular contour exists, and based on the searched center position candidate point and its radius. An image of an area centered on the center candidate point and surrounded by a circle with a minimum allowable radius smaller than the radius r@ and a circle with a maximum allowable radius larger than the radius.
In each scan, the presence or absence of a contour candidate point where the corpse suddenly changes (brightness or darkness) is detected, and when the contour candidate point becomes larger than the predetermined footprint, the approximately circular contour is detected as described above. A method for recognizing taste (Rouchi, Hinoki Kaku) that recognizes that it exists in an image of a region. (2) The method for recognizing a substantially circular contour line according to claim 1, wherein the substantially circular contour line includes contour lines of an ellipse and a polygon close to one circle. (3) The center candidate point of the approximately circular contour is determined by searching for contour candidate points and when the distance between any two contour candidate points is approximately equal to the diameter of the approximately circular contour to be recognized. Claim No. 1 (1) Detection from two contour candidate points
) The method for recognizing a substantially circular contour line described in section ). (4) The number of radial scans is 8, and the preset number is 6.
A method for recognizing a substantially circular contour line according to claim (1).