Logo Detection Method

Abstract

A logo detection method is disclosed. According to the grayscale image of the current display screen, the pixel brightness of each feature point of at least two groups of feature points is obtained, and according to the pixel brightness of each feature point and its corresponding preset brightness value interval, whether the feature point is a standard feature point is determined, and according to the number of standard feature points and the total number of the at least two groups of feature points, whether the current display screen displays a target logo is determined. The detection and judgment steps of the logo detection method of the present invention are simple and effective, and can quickly and in real time determine whether the current display screen has an effective display of the target logo.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and in particular to a logo detection method. Background Art

[0002] In published videos or pictures, the display screen or image generally includes a logo to indicate the ownership of the video or picture.

[0003] Among them, for video information, the displayed picture is constantly changing, and the logo may not be displayed or displayed incorrectly. Therefore, it is necessary to be able to detect in real time whether the logo in the displayed picture is effectively displayed. Summary of the invention

[0004] In view of the above problems, an object of the present invention is to provide a logo detection method, so as to conveniently detect in real time whether a logo in a video image display screen is effectively displayed.

[0005] According to one aspect of the present invention, there is provided a logo detection method, comprising:

[0006] Get the grayscale image of the current display;

[0007] According to the grayscale image, obtaining pixel brightness of each feature point of at least two groups of feature points;

[0008] Determine whether the feature point is a standard feature point based on the pixel brightness of each feature point and its corresponding preset brightness value interval;

[0009] It is determined whether the current display screen displays the target logo according to the number of standard feature points and the total number of the at least two groups of feature points.

[0010] Optionally, the step of judging whether the current display screen displays the target logo according to the number of standard feature points and the total number of the at least two groups of feature points includes:

[0011] When the difference between the number of the standard feature points and the total number of the at least two groups of feature points is less than or equal to the error threshold, it is determined that the current display screen displays the target logo.

[0012] Optionally, also include

[0013] The at least two groups of feature points are obtained according to the shape of the target logo.

[0014] Optionally, it also includes:

[0015] At least two groups of preset brightness value intervals are obtained according to the grayscale range of the display picture, and the brightness value intervals correspond one-to-one to each group of feature points of the at least two groups of feature points.

[0016] Optionally, the step of obtaining the at least two groups of feature points according to the shape of the target logo includes:

[0017] Acquire a display image including a target logo;

[0018] Acquire a grayscale image of the display screen;

[0019] Selecting a plurality of feature points in the grayscale image according to the shape of the target logo;

[0020] Obtaining brightness values ​​of the plurality of feature points;

[0021] The plurality of feature points are grouped according to the brightness values ​​of the plurality of feature points to obtain the at least two groups of feature points.

[0022] Optionally, the step of obtaining at least two groups of preset brightness value intervals according to the grayscale range of the display picture includes:

[0023] The preset brightness value interval is obtained according to the obtained pixel brightness of each feature point of the at least two groups of feature points and the grayscale range of the grayscale image of the display screen including the target logo.

[0024] Optionally, the position of the target logo in the display image and the brightness of each pixel are fixed, and the outline of the target logo is fixed.

[0025] The logo detection method provided by the present invention obtains the pixel brightness of each feature point of at least two groups of feature points based on the grayscale image of the current display screen, determines whether the feature point is a standard feature point based on the pixel brightness of each feature point and its corresponding preset brightness value range, and determines whether the current display screen displays the target logo based on the number of standard feature points and the total number of the at least two groups of feature points. The detection and judgment steps are simple and effective, and can quickly and in real time determine whether the current display screen effectively displays the target logo.

[0026] Set the error threshold to avoid misjudgment and improve the reliability of detection.

[0027] The preset brightness value interval is determined according to the grayscale range of the grayscale image of the display screen, which can be applied to the logo detection of the display screen of any grayscale standard, thereby improving the practicability of the logo detection method of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] The above and other objects, features and advantages of the present invention will become more apparent through the following description of the embodiments of the present invention with reference to the accompanying drawings, in which:

[0029] Figure 1A and Figure 1B A schematic diagram showing feature point selection of a logo detection method according to an embodiment of the present invention is shown;

[0030] Figure 2 A flow chart of feature point selection of a logo detection method according to an embodiment of the present invention is shown;

[0031] Figure 3 A detection flow chart of a logo detection method according to an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0032] Various embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. In each of the accompanying drawings, the same elements are represented by the same or similar reference numerals. For the sake of clarity, the various parts in the accompanying drawings are not drawn to scale.

[0033] The specific implementation of the present invention is further described in detail below in conjunction with the drawings and examples.

[0034] Figure 1A and Figure 1B A schematic diagram of feature point selection of a logo detection method according to an embodiment of the present invention is shown. Figure 1A and Figure 1B The pixel values ​​of are inverted to mark the selected feature points, wherein, to clearly show the pixel points of the present invention, Figure 1B The area within each indicated box in corresponds to a pixel, and the wireframe corresponds to the spacing between pixels.

[0035] like Figure 1A and Figure 1B As shown, in this embodiment, the logo "BTV" is taken as an example. According to the outline of the logo, 14 feature points are selected, wherein the feature points are divided into two groups according to the gray value of each feature point. The first group includes pixel points 1, 3, 5, 8, 9, and 14, and the second group includes pixel points 2, 4, 6, 7, 10, 11, 12, and 13. Figure 1A For grayscale reference, the brightness range of the pixels in the first group is [0, 120], and the brightness range of the pixels in the second group is [230, 255]. The second group is within the logo range, and the first group is outside the logo range. Each feature point is located at a vertex or inflection point of the logo pattern.

[0036] The feature points are selected adaptively according to the actual shape of the target logo, and are also selected according to possible display errors of the target logo to ensure the accuracy of the detection. This embodiment takes a logo in the form of letters as an example, but the present invention has no particular limitation on the logo shape and can be applied to target logos of any shape.

[0037] Figure 2 A flow chart of feature point selection of a logo detection method according to an embodiment of the present invention is shown.

[0038] like Figure 2As shown, the feature point selection process of the logo detection method according to the embodiment of the present invention includes:

[0039] Step S101: Acquire a logo image.

[0040] Get the display showing the target logo.

[0041] Step S102: extracting the grayscale image of the logo picture.

[0042] The obtained logo picture is converted into a grayscale image to obtain the grayscale image of the logo picture, so as to perform subsequent processing according to the brightness value of each pixel in the image.

[0043] Step S103: Select N feature points.

[0044] N feature points are selected according to the contour of the target logo and its position in the image, where N ≥ 3.

[0045] Step S104: Obtain the brightness value of the feature point.

[0046] According to the obtained grayscale image and the positions of the selected N feature points, the brightness values ​​of the N feature points are obtained.

[0047] Step S105: grouping the feature points according to their brightness values.

[0048] The feature points are divided into two groups according to their brightness values. The grouping standard is based on the grayscale range of the image. For example, the grayscale range is [0, 255], and the brightness of the pixels used to display the logo is generally much different from that of other pixels. You can choose to group them using a grayscale of 200 as a threshold to obtain a first group of feature points and a second group of feature points. The brightness of the pixels in the first group of feature points is less than 200, corresponding to the pixels outside the logo, and the brightness of the pixels in the second group of feature points is greater than 200, corresponding to the pixels inside the logo.

[0049] Step S106: Determine the brightness value interval of each group.

[0050] According to the brightness of each pixel of the first group of feature points and the second group of feature points, a first brightness value interval of each pixel of the first group of feature points and a second brightness value interval of each pixel of the second group of feature points are obtained. The brightness value interval is used as a basis for feature point judgment of actual logo detection, wherein the first group is an interval from the highest brightness to 0, and the second group is an interval from the lowest brightness to 255. The brightness value interval determined in this way can be adjusted according to the grayscale range of the display screen, so as to be applicable to display screens of various brightnesses, thereby improving the universality of the logo detection method of the present invention.

[0051] Step S107: Determine the number of feature points where errors are allowed in logo detection

[0052] The allowed error threshold E is set and confirmed according to the number of selected feature points, where 0≤E<N, N is the total number of selected feature points, and N is greater than or equal to 3.

[0053] Figure 3 A detection flow chart of a logo detection method according to an embodiment of the present invention is shown.

[0054] like Figure 3 As shown, the logo detection method of the embodiment of the present invention includes:

[0055] Step S210: extracting a grayscale image.

[0056] Extract a grayscale image of the displayed image.

[0057] Step S220: extracting the pixel brightness of the feature point.

[0058] The feature point positions are determined according to the target logo, and the brightness of the pixels at each feature point position is extracted.

[0059] Step S230: Determine whether the brightness value is within the group brightness range.

[0060] According to the preset brightness value interval, it is determined whether the brightness value of the pixel point at each feature point position is within the brightness value interval of the corresponding group.

[0061] According to the judgment of step S230, if it is within the brightness value interval of the corresponding group, step S231 is determined: the feature point is a standard feature point; if it is not within the brightness value interval of the corresponding group, step S232 is determined: the feature point is not a standard feature point.

[0062] For example, the second brightness interval is [235, 255], and the first brightness interval is [0, 120]. The brightness values ​​of the first group of feature points are determined to be within the range of 235 to 255, and the brightness values ​​of the second group of feature points are determined to be within the range of 0 to 120. Feature points whose brightness values ​​are within their respective intervals are determined as standard feature points.

[0063] Step S240: Determine whether the feature point traversal is completed.

[0064] It is determined whether all feature points have been determined. If not, the process returns to step S220 to extract the pixel brightness of the next feature point and perform determination until all feature points have been detected and determined.

[0065] Step S250: Determine whether the number of feature points that do not meet the standard is less than a threshold value. According to the determination result, step S251 is executed to determine whether it is a logo or step S252 is executed to determine whether it is not a logo.

[0066] The number of the determined standard feature points and the set threshold value determine whether the logo displayed in the current detected display image is the target logo. The threshold value is a preset error threshold value E. If the number of feature points that do not meet the standard is less than or equal to the error threshold value E, the logo is judged as Figure 1A and Figure 1B For example, the pixel brightness of pixels 1, 3, 5, 8, 9, and 14 corresponding to the first group of feature points is dark, and the pixel brightness of pixels 2, 4, 6, 7, 10, 11, 12, and 13 of the second group of feature points is bright. The bright pixels are lit and the dark pixels are dark. The combined image effectively shows the target logo "BTV".

[0067] Among them, feature points 7, 10 and 11 determine the vertex area of ​​the bright display area of ​​"T", feature points 8 and 9 define the turning point area of ​​"T", and combined to define the display of "T"; feature points 10, 12 and 13 roughly determine the bright display area of ​​"V", and the dark area of ​​feature point 14 determines the turning point of "V". When the pixel brightness of each feature point meets the standard, the correct display of "V" can be determined; the confirmation principle of "B" and other types of logo patterns.

[0068] The logo detection method provided by the present invention selects the first and second groups of dark and bright feature points according to the outline of the target logo, and confirms the standard brightness range of each group of feature points, and judges the number of feature points that meet the standard according to whether the brightness of the pixels of the first and second groups of feature points in the displayed image falls within the corresponding standard brightness range, and then judges whether the brightness of the feature pixels in the logo display area in the current display screen is normally combined according to the requirements, so as to judge whether the displayed logo is the target logo, or whether there is a displayed logo. And the data processing amount is small.

[0069] Among them, this embodiment sets two groups of feature points. For logos with multi-gradient brightness or color logos, the feature points can also be divided into multiple groups according to the display characteristics of the target logo, and multiple groups of brightness intervals are divided according to the pixel brightness of each group of feature points. The final judgment only needs to determine the number of feature points that meet the standards or do not meet the standards and the size of the corresponding threshold to determine whether the logo display on the current display screen is qualified.

[0070] According to the embodiments of the present invention as described above, these embodiments do not describe all the details in detail, nor do they limit the invention to the specific embodiments described. Obviously, many modifications and changes can be made based on the above description. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can make good use of the present invention and the modified use based on the present invention. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A logo detection method, characterized in that: include: Get the grayscale image of the current display; According to the grayscale image, obtaining pixel brightness of each feature point of at least two groups of feature points; Determine whether the feature point is a standard feature point according to the pixel brightness of each feature point and its corresponding preset brightness value interval; the preset brightness value interval is a plurality of groups of brightness intervals divided according to the pixel brightness of each group of feature points by dividing the feature points into multiple groups according to the display characteristics of the target logo; Determine whether the current display screen displays the target logo according to the number of standard feature points and the total number of the at least two groups of feature points, Wherein, acquiring the pixel brightness of each feature point of at least two groups of feature points according to the grayscale image comprises: Obtaining a plurality of feature points according to the shape of the target logo; Obtaining brightness values ​​of the plurality of feature points; The plurality of feature points are grouped according to the brightness values ​​of the plurality of feature points to obtain the at least two groups of feature points.

2. The logo detection method according to claim 1, characterized in that: The step of judging whether the current display screen displays the target logo according to the number of standard feature points and the total number of the at least two groups of feature points comprises: When the difference between the number of the standard feature points and the total number of the at least two groups of feature points is less than or equal to the error threshold, it is determined that the current display screen displays the target logo.

3. The logo detection method according to claim 1, characterized in that: Also includes: At least two groups of preset brightness value intervals are obtained according to the grayscale range of the display picture, and the brightness value intervals correspond one-to-one to each group of feature points of the at least two groups of feature points.

4. The logo detection method according to claim 3, characterized in that: The step of obtaining the at least two groups of feature points according to the shape of the target logo comprises: Acquire a display image including a target logo; Acquire a grayscale image of the display screen; The plurality of feature points are selected in the grayscale image according to the shape of the target logo.

5. The logo detection method according to claim 4, characterized in that: The step of obtaining at least two sets of preset brightness value intervals according to the grayscale range of the display picture includes: The preset brightness value interval is obtained according to the obtained pixel brightness of each feature point of the at least two groups of feature points and the grayscale range of the grayscale image of the display screen including the target logo.

6. The logo detection method according to claim 1, characterized in that: The position of the target logo in the display picture and the brightness of each pixel are fixed, and the outline of the target logo is fixed.

Images