CN106204542A - Visual identity method and system - Google Patents

Abstract

The invention relates to the technical field of visual recognition, and discloses a visual recognition method and system. In the present invention, the visual recognition method includes: obtaining the known conditions of the grayscale image of the object to be recognized and the target circle; wherein the known conditions include the position of the first center of the circle, the first radius, the polarity of the target circle, and the first deviation value and the second deviation value; carry out edge detection on the grayscale image to obtain the image contour of the grayscale image; perform polarity screening on the image contour according to known conditions to obtain the first contour; determine the second circle center position and the second circle center position according to the first contour Radius; wherein, the second center position is the determined center position of the target circle, and the second radius is the determined radius of the target circle. The embodiment of the present invention also provides a visual recognition system. Compared with the prior art, the speed and accuracy of the circle center and radius extraction can be improved when extracting the circle on the image of the object to be recognized.

Description

Visual recognition method and system

technical field

The invention relates to the technical field of visual recognition, in particular to a visual recognition method and system.

Background technique

A layer of keyboard soft membrane is arranged under the upper cover of the membrane keyboard, and an elastic keycap (usually a rubber cap) is arranged on the position of each key. The position of the elastic keycap relative to the key must not be offset, otherwise , which may result in invalid keystrokes.

When manufacturing the membrane keyboard, it is necessary to detect whether the elastic key cap is offset relative to the corresponding key. When testing, first take a photo of the front of the membrane keyboard to obtain a grayscale image (a pair of two-dimensional plan); wherein, in the grayscale image, the outline of the area covered by the key is a rounded quadrilateral, and the elastic keycap covers The outline of the area is a circle; then, using visual recognition technology, the circle corresponding to the elastic keycap is extracted from the grayscale image; finally, whether the extracted circle is offset relative to the rounded quadrilateral is detected.

In the process of realizing the present invention, the inventor found that there are the following problems in the prior art: on the one hand, there are few methods for extracting circles around keys (keys) in keyboard soft membrane detection, and, from the angle of circle extraction, Said that there are two main methods that may be used at present: the first one is based on the Hough transform circle extraction method, but the gradient-based method currently commonly used in this type of method, when the circle is relatively small, the error is relatively large, and the extraction If it is not accurate, the position of the extracted circle center will be offset. The second method is to extract the circle by fitting. This method extracts the circle with a large amount of calculation, resulting in a slow extraction speed.

Contents of the invention

The purpose of the embodiments of the present invention is to provide a visual recognition method and system, so that the efficiency and accuracy of circle center and radius extraction can be improved when extracting the circle on the image of the object to be recognized.

In order to solve the above-mentioned technical problems, the embodiment of the present invention provides a visual recognition method, including: acquiring the grayscale image of the item to be recognized and the known conditions of the target circle; wherein the known conditions include the position of the center of the first circle, The first radius, the polarity of the target circle, the first deviation value and the second deviation value, the target circle is a circle to be extracted, the first deviation value is the center deviation value of the target circle, the The second deviation value is the radius deviation value of the target circle, the first center position is the estimated position of the target circle center, and the first radius is the estimated radius of the target circle; for the grayscale image performing edge detection to obtain the image contour of the grayscale image; performing polarity screening on the image contour according to the known conditions to obtain a first contour; determining a second circle center position and a second radius according to the first contour; Wherein, the second center position is the determined center position of the target circle, and the second radius is the determined radius of the target circle.

Embodiments of the present invention also provide a visual recognition system, including: an acquisition module, a detection module, a screening module, and a determination module; the acquisition module is used to acquire the grayscale image of the item to be identified and the known conditions of the target circle ; Wherein, the known conditions include the first center position, the first radius, the polarity of the target circle, the first deviation value and the second deviation value, the target circle is a circle to be extracted, and the first The deviation value is the center deviation value of the target circle, the second deviation value is the radius deviation value of the target circle, the first center position is the estimated position of the target circle center, and the first radius is The estimated radius of the target circle; the detection module is used to perform edge detection on the grayscale image to obtain the image profile of the grayscale image; the screening module is used to perform edge detection on the grayscale image according to the known conditions. Polarity screening is performed on the image contour to obtain the first contour; the determining module is configured to determine a second center position and a second radius according to the first contour; wherein the second center position is the determined center of the target circle position, the second radius is the determined radius of the target circle.

Compared with the prior art, the embodiment of the present invention obtains the image contour by performing edge detection on the grayscale image of the object to be recognized, and then performs polarity screening on the image contour according to known conditions, so that the contour of the circle to be extracted can be obtained, Thus, the position and radius of the center of the circle to be extracted can be determined. In the calculation process, since the estimated position of the center of the circle to be extracted and its deviation, the estimated radius and its deviation are known, using these known conditions can first obtain the fuzzy The outline of the image can then be accurately screened to finally determine the exact outline of the circle to be extracted. It is simple, fast, and the amount of calculation is small, and it is accurate, so that when extracting the circle on the image of the object to be recognized, the center and radius of the circle can be extracted. efficiency and accuracy.

In addition, determining the second circle center position and the second radius according to the first contour specifically includes: taking the pixel point on the first contour as the center of the circle, and taking any radius within the range of the radius of the target circle The value is to draw a circle with a radius; the value range of the radius is calculated according to the first radius and the second deviation value; within the range of the center position of the target circle, the number of intersections of circles with the same radius is counted; wherein, The center position range of the target circle is calculated according to the first center position and the first deviation value; the radius of the circle with the most number of intersections is determined as the second radius, and the position of the intersection point is determined as the The position of the second circle center. In this way, only the intersections of all circles within the center position range of the target circle are counted, regardless of the number of intersections of circles outside the range, which can increase the speed of selecting the second circle center position.

In addition, in drawing a circle with a pixel point on the first contour as the center and using any radius value within the radius value range of the target circle as a radius, it specifically includes: only drawing the center of the target circle Arcs within the position range; within the center position range of the target circle, counting the number of intersections of circles with the same radius specifically includes: counting the number of intersections of arcs of circles with the same radius; The radius of the circle is determined as the second radius, and the position of the intersection point is determined as the position of the second center of the circle, which specifically includes: determining the radius of the circle corresponding to the arc with the largest number of intersections as the second radius . Only draw the arc within the range of the center position of the target circle, and automatically avoid counting the number of intersections of circles outside the range when counting, which can improve the speed of selecting the second center position.

In addition, before obtaining the known conditions of the grayscale image of the item to be identified and the target circle, it also includes: establishing a first coordinate system with the first center position as the origin, and taking the preset point in the grayscale image as the origin Establishing a second coordinate system; determining the second center position and the second radius according to the first contour specifically includes: determining the second center position and the second radius in the first coordinate system; After determining the second circle center position and the second radius according to the first contour, it also includes: according to the transformation relationship between the first coordinate system and the second coordinate system, obtaining the second circle center position at the Coordinates in the second coordinate system described above. By establishing a coordinate system with the estimated position of the center of the target circle as the origin, the complexity of calculation is reduced, simple and easy to implement, and at the same time, the calculation of the center position of the target circle is more accurate.

In addition, performing edge detection on the grayscale image to obtain the image profile of the grayscale image specifically includes: performing edge detection on the grayscale image by using a Canny operator. Canny operator has strong denoising ability and better edge detection effect.

Description of drawings

Fig. 1 is a flow chart of a visual recognition method according to a first embodiment of the present invention;

Fig. 2 is a flow chart of a visual recognition method according to a second embodiment of the present invention;

Fig. 3 is a gray scale image of the soft membrane of the keyboard according to the second embodiment of the present invention;

Fig. 4 is a contour diagram of a single button extracted from a grayscale image according to a second embodiment of the present invention;

Fig. 5 is a second outline diagram according to the second embodiment of the present invention;

Fig. 6 is a schematic diagram of selecting a pixel point and its corresponding inner and outer points on the second contour map according to the second embodiment of the present invention;

FIG. 7 is a first outline diagram according to a second embodiment of the present invention;

Fig. 8 is a schematic diagram of drawing a circle with the pixel point on the first contour as the center and any radius value within the radius value range of the circle to be extracted according to the second embodiment of the present invention;

9 is a schematic diagram of calculating the effective angle of each pixel according to the third embodiment of the present invention;

FIG. 10 is a flowchart of a visual recognition method according to a fourth embodiment of the present invention;

Fig. 11 is a schematic structural diagram of a visual recognition system according to a fifth embodiment of the present invention;

Fig. 12 is a schematic structural diagram of a visual recognition system according to a sixth embodiment of the present invention.

detailed description

In order to make the object, technical solution and advantages of the present invention clearer, various embodiments of the present invention will be described in detail below in conjunction with the accompanying drawings. However, those of ordinary skill in the art can understand that, in each implementation manner of the present invention, many technical details are provided for readers to better understand the present application. However, even without these technical details and various changes and modifications based on the following implementation modes, the technical solution claimed in this application can also be realized.

The first embodiment of the present invention relates to a visual recognition method. The visual recognition method in this embodiment is suitable for extracting circles in objects to be recognized. The process is shown in FIG. 1 , and the details are as follows:

In step 101: Acquiring the grayscale image of the object to be identified and the known conditions of the target circle. In the field of keyboard manufacturing, the item to be identified can be the soft membrane of the keyboard. Specifically, the known conditions in this embodiment include the position of the first center of the circle, the first radius, the polarity of the target circle, the first deviation value and the second deviation value. Wherein, the target circle is the circle to be extracted, the first deviation value is the center deviation value of the target circle, the second deviation value is the radius deviation value of the target circle, the first center position is the estimated position of the center of the target circle, and the first radius is Inferred radius of the target circle.

In step 102: performing edge detection on the grayscale image to obtain an image profile of the grayscale image.

In step 103: performing polarity screening on image contours according to known conditions to obtain a first contour.

In step 104: a second center position and a second radius are determined according to the first contour; wherein, the second center position is the determined center position of the circle to be extracted, and the second radius is the determined radius of the circle to be extracted.

Compared with the prior art, this embodiment obtains the image contour by performing edge detection on the grayscale image of the object to be recognized, and then performs polarity screening on the image contour according to known conditions, so that the contour of the circle to be extracted can be obtained, so that The position and radius of the center of the circle to be extracted can be determined. In the calculation process, since the estimated position and deviation, estimated radius and deviation of the center of the circle to be extracted are known, a blurred image can be obtained first by using these known conditions Contour, and then you can carry out precise screening, and finally determine the accurate contour of the circle to be extracted, which is simple, fast, with a small amount of calculation, and accurate, so that the efficiency of center and radius extraction can be improved when extracting the circle on the image of the object to be recognized and accuracy.

The second embodiment of the present invention relates to a visual recognition method, the process of which is shown in Figure 2, specifically as follows:

In step 201, the grayscale image of the object to be identified and the known conditions of the target circle are acquired. In this implementation manner, an example is described in which the item to be identified is a soft membrane of a keyboard. Specifically, the known conditions in this embodiment include the position of the first center of the circle, the first radius, the polarity of the target circle, the first deviation value and the second deviation value. Wherein, the target circle is the circle to be extracted, the first deviation value is the center deviation value of the target circle, the second deviation value is the radius deviation value of the target circle, the first center position is the estimated position of the center of the target circle, and the first radius is Inferred radius of the target circle. In practical applications, the image of the item to be identified can be obtained by taking pictures or scanning the actual object, and the grayscale processing is performed on the image to obtain the grayscale image of the item to be identified, as shown in Figure 3, where 31 is the area corresponding to a button .

In step 202, an edge detection is performed on the grayscale image to obtain an image profile of the grayscale image. Specifically, in this embodiment, the contour in Fig. 3 can be detected by the edge detection operator-Canny (canny) operator, in preparation for extracting the circle on the key (key), wherein, the contour of part 31 in Fig. 3 The enlarged view is shown in Figure 4.

In step 203, a second profile is obtained according to known conditions. Specifically, the second contour is the contour in the target area, and the target area is an annular area, and the radius of the inner circle of the annular area is equal to the first radius minus the first fixed value (the first fixed value can be based on the first deviation value and The second deviation value is determined, specifically, the first fixed value is greater than the sum of the first deviation value and the second deviation value), and the outer circle radius of the annular area is equal to the first radius plus the second fixed value (the second fixed value It may be determined according to the first deviation value and the second deviation value, specifically, the second fixed value is greater than the sum of the first deviation value and the second deviation value); the first fixed value and the second fixed value may be the same. For example, it can be assumed in Figure 4 that the center of the circle is (x, y) and the radius is r. In this embodiment, the first deviation value is assumed to be 5, the second deviation value is assumed to be 2, and the first fixed value and the second fixed value Both can be 10, then the radius of the inner circle of the second contour is r-10, and the radius of the outer circle of the second contour is r+10, and this part of the contour is displayed separately, as shown in FIG. 5 .

In step 204, a pixel point on the second contour is selected to obtain an inner point and an outer point. Specifically, both the inner point and the outer point are located on the line connecting the pixel point on the second contour and the pixel point at the first circle center position, more specifically, the inner point and the outer point are respectively located at the first circle center position The pixel point is the center of the circle, and the distance from the pixel point on the second contour to the pixel point of the first center position is the inside and outside of the circle of the radius, as shown in Figure 6 (the O point is the center of the circle among the figure, and the A point is The inner point, the B point is the outer point, and the C point is the pixel point on the second contour).

In step 205, it is judged whether the gray value of the inner point is smaller than the gray value of the outer point. If yes, go to step 206, otherwise go to step 207. Specifically, if the judgment result in this step is yes, indicating that the gray value of point A is smaller than the gray value of point B, then enter step 206; otherwise, it means that the gray value of point A is greater than the gray value of point B, then Go to step 207.

In step 206, the pixels corresponding to the inner point and the outer point are removed from the second contour. Specifically, when the gray value of point A is smaller than the gray value of point B, pixel C is cleared.

In step 207, the pixel points corresponding to the inner point and the outer point on the second contour are reserved. Generally speaking, when the gray value of point A is greater than the gray value of point B, keep C pixel.

In step 208, it is determined whether there are unselected pixel points on the second contour. If so, return to step 204, otherwise enter step 209. Specifically, if the judgment result in this step is yes, it means that there are unselected pixels on the second contour, then return to step 204 and continue to select pixels on the second contour for judgment, otherwise, it means that all pixels on the second contour If all the points have been selected, go to step 209.

In step 209, all the retained pixel points form the first contour. The first contour in this embodiment is the final remaining contour (composed of pixels) after passing the polarity screening, as shown in FIG. 7 .

In step 210, a circle is drawn with the pixel point on the first contour as the center and any radius value within the radius value range of the circle to be extracted. In this way, the pixel point on the first contour is used as the center of the circle, and a circle is continuously drawn with any radius value within the radius value range of the circle to be extracted, and the effect diagram shown in FIG. 8 is obtained.

In step 211 , count the number of intersections of circles with the same radius within the range of the center position of the circle to be extracted. Wherein, the range of the center position of the circle to be extracted can be calculated according to the first center position and the first deviation value, for example, the first deviation value can be 5 pixels, and the position of the circle center is recorded as s[x,y], for example , if the estimated position of the center of the target circle is [0,0], then the value range of x is [-5,5], and the value range of y is [-5,5] (the precision is 1 pixel).

In step 212, the radius of the circle with the most number of intersections is determined as the second radius, and the position of the intersection point is determined as the second circle center position.

In this embodiment, only the intersections of all circles within the center position range of the target circle are counted, and the number of intersections of circles outside the range is not considered, so that the speed of selecting the second circle center position can be improved.

A third embodiment of the present invention relates to a visual recognition method. The third embodiment has been improved on the basis of the second embodiment. The improvement is that: in this embodiment, the pixel point on the first contour is used as the center of the circle, and the radius of the circle to be extracted is arbitrary within the value range. When a radius value is used to draw a circle, only draw the arc within the range of the center of the circle to be extracted.

Specifically, in this embodiment, when counting the number of intersections of circles with the same radius, only the number of intersections of arcs of circles with the same radius may be counted. The radius of the circle corresponding to the arc with the most number of intersections is determined as the second radius, and the position of the intersection point is determined as the second center position.

In this embodiment, as shown in Figure 9, the effective angles a2 and a1 of each pixel point on the outline in Figure 7 are first calculated (the area between the two rays L1 and L2 starting from the outline point just includes a rectangular area , the angle between L1 and the positive direction of the coordinate axis x is a1, and the angle between L2 and the positive direction of the coordinate axis x is a2); the black dot in the figure is the estimated position of the center of the target circle (known conditions ), the rectangle is the range of the center of the target circle (the actual center of the circle may be any point in the area, according to the deviation of ±5 above, the length of the rectangle is 11 pixels).

In practical applications, when counting the number of intersections of arcs of circles with the same radius, a three-dimensional array with an initial value of 0 can be generated according to the deviation of the center of the target circle and the deviation of the radius. In this embodiment, the center of the target circle The deviation can be 5 pixels, and the radius deviation can be 2 pixels. Assuming that the three-dimensional array is s[x,y,r], the respective value ranges of x, y, and r are [x-5, x+5], [y -5, y+5], [r-2, r+2], where the coordinates of the estimated position of the center of the target circle are (x, y), and the estimated radius is r.

According to the accuracy of the center and radius to be extracted, set the following quantities: angle step stepa, x step size stepx, y step size stepy, r step size stepr; x, y, r step size stepx, stepy respectively , stepr, it is assumed that stepx, stepy, and stepr are all 1 in this embodiment. The deviation of r is recorded as △R, the deviation of x is recorded as △X, and the deviation of y is recorded as △Y; traverse each pixel shown in Figure 7 (assuming the coordinates are x0, y0), and count the circles of the circles with the same radius The number of arc intersections, the specific calculation code is as follows:

Among them, the angle between L1 and L2 in Figure 9 is divided into Indexa parts, ceil is the smallest integer that returns greater than or equal to the specified expression, inDxa is the independent variable, and the value range is from 0 to indexa, and inDxr is the independent variable , the value range is from 0 to 2*△R.

According to the above statistical results, the radius of the circle corresponding to the arc with the most number of intersections is determined as the second radius, and the position of the intersection point is determined as the second center position.

This embodiment can only draw the arc within the range of the center position of the target circle, and automatically avoid counting the number of intersections of the circles outside the range during statistics, which can increase the speed of selecting the second center position.

A fourth embodiment of the present invention relates to a visual recognition method. The fourth embodiment has been improved on the basis of the first embodiment. The improvement is that in this embodiment, the coordinate system can be established with different center positions of the circles to be extracted, which reduces the complexity of calculation and is simple. , easy to implement, and at the same time make the calculation of the center position of the target circle more accurate. The process flow of the visual recognition method in this embodiment is shown in FIG. 10 , and the details are as follows.

In step 1001: a first coordinate system is established with the first circle center position as the origin. Specifically, the first center position in this embodiment is the estimated position of the center of the target circle.

In step 1002: a second coordinate system is established with the preset point in the grayscale image as the origin. Specifically, the preset point in this embodiment can be the center of a circle on any key (key) in the grayscale image in FIG.

In step 1003: Acquiring the grayscale image of the object to be identified and the known conditions of the target circle. Specifically, the known conditions in this embodiment include the position of the first center of the circle, the first radius, the polarity of the target circle, the first deviation value and the second deviation value. Wherein, the target circle is the circle to be extracted, the first deviation value is the center deviation value of the target circle, the second deviation value is the radius deviation value of the target circle, the first center position is the estimated position of the center of the target circle, and the first radius is Inferred radius of the target circle. In practical applications, the image of the item to be identified can be obtained by taking a photo or scanning the object, and the image is processed in grayscale to obtain a grayscale image of the item to be identified, as shown in FIG. 3 .

In step 1004: performing edge detection on the grayscale image to obtain the image contour of the grayscale image. Specifically, in this embodiment, the contour in FIG. 3 can be detected by the edge detection operator—canny operator, in preparation for extracting the circle on the key. The contour of the key after extraction is shown in FIG. 4 .

In step 1005: performing polarity screening on the image contour according to known conditions to obtain the first contour. Specifically, the first contour can be obtained according to steps 203 to 209 in the second embodiment, and the obtained first contour is shown in FIG. 7 .

In step 1006: in the first coordinate system, determine the position of the second circle center and the second radius. Specifically, the values of the second circle center position and the second radius in this embodiment can be obtained according to steps 210 to 212 in the second embodiment.

In step 1007: According to the transformation relationship between the first coordinate system and the second coordinate system, the coordinates of the second circle center position in the second coordinate system are acquired. Specifically, according to the positional relationship between the origin of the second coordinate system and the origin of the first coordinate system, the coordinate value of the second circle center position in the first coordinate system can be converted into the coordinate value in the second coordinate system.

In this embodiment, by establishing a coordinate system with the estimated position of the center of the target circle as the origin, the complexity of calculation is reduced, which is simple and easy to implement, and at the same time makes the calculation of the center position of the target circle more accurate.

The division of steps in the above methods is only for the sake of clarity of description. During implementation, they can be combined into one step or some steps can be split and decomposed into multiple steps. As long as they contain the same logical relationship, they are all within the scope of protection of this patent. ; Adding insignificant modifications or introducing insignificant designs to the algorithm or process, but not changing the core design of the algorithm and process are all within the scope of protection of this patent.

The fifth embodiment of the present invention relates to a visual recognition system, as shown in FIG. 11 , including: an acquisition module 1 , a detection module 2 , a screening module 3 and a determination module 4 .

The acquisition module 1 is used to acquire the grayscale image of the item to be identified and the known conditions of the target circle; where the known conditions include the first circle center position, the first radius, the polarity of the target circle, the first deviation value and the second Deviation value, the target circle is the circle to be extracted, the first deviation value is the deviation value of the center of the target circle, the second deviation value is the radius deviation value of the target circle, the first center position is the estimated position of the center of the target circle, and the first radius is the estimated radius of the target circle.

The detection module 2 is configured to perform edge detection on the grayscale image to obtain an image profile of the grayscale image.

The screening module 3 is configured to perform polarity screening on image contours according to known conditions to obtain the first contour.

The determining module 4 is configured to determine a second center position and a second radius according to the first contour; wherein, the second center position is the determined center position of the target circle, and the second radius is the determined radius of the target circle.

It is not difficult to find that this embodiment is a system embodiment corresponding to the first embodiment, and this embodiment can be implemented in cooperation with the first embodiment. The relevant technical details mentioned in the first embodiment are still valid in this embodiment, and will not be repeated here in order to reduce repetition. Correspondingly, the relevant technical details mentioned in this implementation manner can also be applied in the first implementation manner.

It is worth mentioning that all the modules involved in this embodiment are logical modules. In practical applications, a logical unit can be a physical unit, or a part of a physical unit, or multiple physical units. Combination of units. In addition, in order to highlight the innovative part of the present invention, units that are not closely related to solving the technical problems proposed by the present invention are not introduced in this embodiment, but this does not mean that there are no other units in this embodiment.

The sixth embodiment of the present invention relates to a visual recognition system. The determination module 4 in this embodiment specifically includes: a circle drawing sub-module 41 , a statistics sub-module 42 and a comparison sub-module 43 , as shown in FIG. 12 .

The circle drawing sub-module 41 is used to draw a circle with the pixel point on the first contour as the center and any radius value within the radius value range of the target circle; the radius value range is calculated according to the first radius and the second deviation value get.

The statistics sub-module 42 is used to count the number of intersections of circles with the same radius within the center position range of the target circle; wherein, the center position range of the target circle is calculated according to the first center position and the first deviation value.

Comparison sub-module 43 is used to compare the number of intersections of circles with different radii in the statistical results of the statistical sub-module 42, determine the radius of the circle with the largest number of intersections as the second radius, and determine the position of the intersection point as the second center position .

Since the second embodiment corresponds to the present embodiment, the present embodiment can be implemented in cooperation with the second embodiment. The relevant technical details mentioned in the second embodiment are still valid in this embodiment, and the technical effects that can be achieved in the second embodiment can also be achieved in this embodiment, and in order to reduce repetition, details are not repeated here. Correspondingly, the relevant technical details mentioned in this embodiment mode can also be applied in the second embodiment mode.

The seventh embodiment of the present invention relates to a visual recognition system. This embodiment has been improved on the basis of the sixth embodiment, and the improvements are as follows:

In this embodiment, the circle drawing sub-module 41 can only draw arcs within the range of the center position of the target circle.

The statistical sub-module 42 can also be used to count the number of intersections of arcs of circles with the same radius.

The comparison sub-module 43 can also be used to compare the number of intersections of arcs of circles with different radii in the statistical results of the statistics sub-module 42, and determine the radius of the circle corresponding to the arc with the most number of intersections as the second radius.

This embodiment is a system embodiment corresponding to the third embodiment, and this embodiment can be implemented in cooperation with the third embodiment. The relevant technical details mentioned in the third embodiment are still valid in this embodiment, and will not be repeated here in order to reduce repetition. Correspondingly, the relevant technical details mentioned in this embodiment mode can also be applied in the third embodiment mode.

Those skilled in the art can understand that all or part of the steps in the method of the above-mentioned embodiments can be completed by instructing related hardware through a program. The program is stored in a storage medium and includes several instructions to make a device (which can be a single-chip , chip, etc.) or a processor (processor) executes all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-OnlyMemory), random access memory (RAM, Random Access Memory), magnetic disk or optical disk and other media that can store program codes.

Those of ordinary skill in the art can understand that the above-mentioned embodiments are specific examples for realizing the present invention, and in practical applications, various changes can be made to it in form and details without departing from the spirit and spirit of the present invention. scope.

Claims (10)

1. A visual recognition method, characterized in that, comprising: Acquiring the grayscale image of the item to be identified and the known conditions of the target circle; wherein the known conditions include the position of the first center of the circle, the first radius, the polarity of the target circle, the first deviation value and the second deviation value , the target circle is a circle to be extracted, the first deviation value is the center deviation value of the target circle, the second deviation value is the radius deviation value of the target circle, and the first center position is The estimated position of the center of the target circle, the first radius being the estimated radius of the target circle; performing edge detection on the grayscale image to obtain the image profile of the grayscale image; performing polarity screening on the image profile according to the known conditions to obtain a first profile; A second center position and a second radius are determined according to the first profile; wherein, the second center position is the determined center position of the target circle, and the second radius is the determined radius of the target circle. 2. The visual recognition method according to claim 1, wherein determining the second center position and the second radius according to the first outline specifically includes: Taking the pixel point on the first contour as the center, draw a circle with any radius value in the radius value range of the target circle; the radius value range is based on the first radius and the second radius value. The deviation value is calculated; Within the center position range of the target circle, count the number of intersections of circles with the same radius; wherein, the center position range of the target circle is calculated according to the first center position and the first deviation value; The radius of the circle with the most number of intersections is determined as the second radius, and the position of the intersection point is determined as the second center position. 3. The visual recognition method according to claim 2, wherein, taking the pixel point on the first contour as the center, drawing a circle with any radius value within the radius value range of the target circle , including: Only draw the arcs within the range of the center position of the target circle; Within the range of the center position of the target circle, counting the number of intersections of circles with the same radius specifically includes: Count the number of intersections of arcs of circles with the same radius; In said determining the radius of the circle with the most number of intersections as the second radius, and determining the position of the intersection point as the position of the second center of the circle, it specifically includes: The radius of the circle corresponding to the arc with the largest number of intersections is determined as the second radius. 4. The visual recognition method according to claim 1, characterized in that, before obtaining the grayscale image of the item to be recognized and the known conditions of the target circle, it also includes: Establishing a first coordinate system with the first center position as the origin, and establishing a second coordinate system with the preset point in the grayscale image as the origin; In determining the position of the second circle center and the second radius according to the first contour, it specifically includes: In the first coordinate system, determine the second center position and the second radius; After determining the second circle center position and the second radius according to the first contour, it also includes: According to the transformation relationship between the first coordinate system and the second coordinate system, the coordinates of the second circle center position in the second coordinate system are acquired. 5. The visual recognition method according to claim 1, characterized in that, performing polarity screening on the image contour according to the known conditions to obtain the first contour, specifically includes: The second contour is obtained according to the known conditions; wherein, the second contour is a contour in the target area, the target area is an annular area, and the radius of the inner circle of the annular area is equal to the first radius minus the first radius A fixed value; the first fixed value is greater than the sum of the first deviation value and the second deviation value, and the outer circle radius of the annular area is equal to the first radius plus the second fixed value; the The second fixed value is greater than the sum of the first deviation value and the second deviation value; For each pixel point on the second contour, obtain an inner point and an outer point; wherein, the inner point and the outer point are both located at the pixel point on the second contour and the first circle center On the connection line of the pixel points at the position, the inner point and the outer point are respectively located with the pixel point at the first center position as the center of the circle, and the pixel point on the second contour to the pixel at the first center position The distance between the points is the inside and outside of the circle of radius; If the gray value of the inner point is smaller than the gray value of the outer point, then remove the inner point and the pixel corresponding to the outer point from the second contour; if the gray value of the inner point If the value is greater than the gray value of the outer point, the pixel points corresponding to the inner point and the outer point on the second contour are retained, wherein all the retained pixel points form the first contour. 6. The visual recognition method according to claim 1, wherein, in performing edge detection on the grayscale image to obtain the image profile of the grayscale image, specifically comprising: Edge detection is performed on the grayscale image using the Canny operator. 7. The visual recognition method according to claim 1, wherein the first deviation value is 5 pixels, and the second deviation value is 2 pixels. 8. A visual recognition system, comprising: an acquisition module, a detection module, a screening module and a determination module; The acquisition module is used to acquire the grayscale image of the item to be identified and the known conditions of the target circle; wherein the known conditions include the first circle center position, the first radius, the polarity of the target circle, the first A deviation value and a second deviation value, the target circle is a circle to be extracted, the first deviation value is a center deviation value of the target circle, and the second deviation value is a radius deviation value of the target circle, The first center position is the estimated position of the center of the target circle, and the first radius is the estimated radius of the target circle; The detection module is configured to perform edge detection on the grayscale image to obtain an image profile of the grayscale image; The screening module is configured to perform polarity screening on the image profile according to the known conditions to obtain a first profile; The determining module is configured to determine a second center position and a second radius according to the first profile; wherein, the second center position is the center position of the determined target circle, and the second radius is the determined target circle of the radius. 9. The visual recognition system according to claim 8, wherein the determination module specifically comprises: a circle drawing submodule, a statistics submodule and a comparison submodule; The circle drawing sub-module is used to draw a circle with the pixel point on the first contour as the center and any radius value within the radius value range of the target circle; the radius value range is based on the The first radius and the second deviation value are calculated; The statistical sub-module is used to count the number of intersections of circles with the same radius within the center position range of the target circle; wherein, the center position range of the target circle is based on the first circle center position and the first The deviation value is calculated; The comparison submodule is used to compare the number of intersections of circles with different radii in the statistical results of the statistical submodule, determine the radius of the circle with the largest number of intersections as the second radius, and determine the position of the intersection as The second center position. 10. The visual recognition system according to claim 9, wherein the circle drawing sub-module only draws an arc within the range of the center position of the target circle; The statistical submodule is used to count the number of times of intersections of circular arcs of circles with the same radius; The comparison submodule is used to compare the number of intersections of arcs of circles with different radii in the statistical results of the statistical submodule, and determine the radius of the circle corresponding to the arc with the largest number of intersections as the second radius .