WO2016101767A1

WO2016101767A1 - Picture cropping method and device and image detecting method and device

Info

Publication number: WO2016101767A1
Application number: PCT/CN2015/095862
Authority: WO
Inventors: 张刚; 胡金辉; 韩玉刚
Original assignee: 北京奇虎科技有限公司; 奇智软件（北京）有限公司
Priority date: 2014-12-24
Filing date: 2015-11-27
Publication date: 2016-06-30

Abstract

The present invention mainly relates to the technical field of Internet, and a main aim is to adopt a proper cropping method in a timely manner for different pictures, and provided are a picture cropping method and device and image detecting method and device, the method comprising: detecting an original picture to be cropped, and detecting a characteristic element comprised in the original picture; selecting a cropping method for cropping the original picture according to the characteristic element; and cropping the original picture in the cropping method to acquire a target picture. According to the present invention, detecting the original picture before cropping the picture and detecting the characteristics element comprised in the picture, and adapting different cropping methods to crop according to a difference of characteristic elements, so as to ensure that proper cropping methods can be adopted in a timely manner for different pictures, so that the target picture acquired from cropping meets requirements.

Description

Picture cutting method and device, and image detecting method and device

Technical field

The present invention relates to the field of Internet technologies, and in particular, to a method and apparatus for picture cropping, and a picture detecting method and apparatus.

Background technique

In the Internet world, it is often necessary to crawl a large number of pictures on the Internet. Due to the different sizes of pictures on the Internet, it is difficult to meet the needs of the captured image size. In this case, the computer needs to automatically complete the cropping of the image, so that the size of the image meets the requirements.

For different images, because the content displayed in the image is different, the parts of the image that need to be cropped are different. At present, the method of cropping by using the three-part composition and the golden ratio composition is to cut the fixed position in the picture, so that it is difficult to ensure that the cropped picture is satisfactory. In particular, some pictures containing characters, using the prior art cutting methods often make the cropped characters incomplete or incomplete, so that the pictures can not be used, or have to be manually cut, low efficiency, and not widely used.

Summary of the invention

In view of the above problems, the present invention has been made in order to provide a method and apparatus for picture cropping and a picture detecting method and apparatus that overcome the above problems or at least partially solve the above problems.

According to an aspect of the present invention, a method for cropping a picture includes: detecting an original picture to be cropped, detecting a feature element included in the original picture; and selecting, according to the feature element, a The cropping mode of the original image is performed; the original image is cropped according to the cropping manner to obtain a target image.

According to another aspect of the present invention, an apparatus for image cropping is provided, comprising: a feature element detecting module, configured to detect a cut original image, and detect a feature element included in the original image; And a selection module, configured to: according to the feature element, select a cropping manner for cropping the original image; and a cropping module, configured to crop the original image according to the cropping manner to obtain a target image.

According to an aspect of the present invention, a person picture cropping method is provided, comprising:

Detecting the original picture of the person to be cropped to detect the face of the character;

The original picture is cropped to obtain a target picture, and at least the face is retained in the target picture.

According to another aspect of the present invention, a character picture cropping apparatus is provided, comprising:

a face detection module, configured to detect a cut original image having a character to detect a face of the character;

And a cropping module, configured to crop the original image to obtain a target image, wherein at least the human face remains in the target image.

According to an aspect of the present invention, a method of cropping a picture including a graphic mark is provided, comprising:

Detecting the cropped original image to detect some or all of the features of the one or more graphic markers included in the original image;

The original picture is cropped to obtain a target picture, and at least one or all of the features of the graphic mark are retained in the target picture.

According to another aspect of the present invention, an apparatus for cropping a picture including a graphical logo is provided, comprising:

a graphic mark detecting module, configured to detect the original picture to be cropped, and detect some or all features of the one or more graphic marks included in the original picture;

And a cropping module, configured to crop the original image to obtain a target image, and retain at least one or all of the features of the graphic identifier in the target image.

According to an aspect of the present invention, a method for detecting a picture mark picture is provided, which includes:

Selecting an area of the color to be extracted from the picture based on an edge position of the picture to be detected;

Extracting colors in the area;

It is determined whether the graphic flag is included in the picture according to the extracted color in the area.

According to another aspect of the present invention, a graphic mark picture detecting apparatus is provided, comprising:

An area selection module, configured to select an area of the color to be extracted from the picture based on an edge position of the picture to be detected;

a color extraction module for extracting colors in the area;

The graphic mark determining module is configured to determine, according to the extracted color in the area, whether the graphic includes a graphic mark.

According to still another aspect of the present invention, a computer program is presented comprising computer readable code that, when executed on a computing device, causes the computing device to perform the method described above.

According to still another aspect of the present invention, a computer readable medium is proposed, wherein the computer program described above is stored.

According to the above technical solution, it is known that the method and apparatus for picture cropping of the present invention have at least the following advantages:

According to the technical solution of the present invention, the original picture is first detected before the picture is cropped, and the feature elements contained therein are detected, and the different clipping methods are used according to the difference of the feature elements to ensure that the different pictures can be timely adopted. The cutting method makes the target image obtained by the cropping meet the requirements.

The above description is only an overview of the technical solutions of the present invention, and the above-described and other objects, features and advantages of the present invention can be more clearly understood. Specific embodiments of the invention are set forth below.

DRAWINGS

Various other advantages and benefits will become apparent to those skilled in the art from a The drawings are only for the purpose of illustrating the preferred embodiments and are not to be construed as limiting. Throughout the drawings, the same reference numerals are used to refer to the same parts. In the drawing:

1 shows a flow chart of a method of picture cropping in accordance with one embodiment of the present invention;

2 shows a flow chart of a method of picture cropping in accordance with one embodiment of the present invention;

3 shows a flow chart of a method of picture cropping in accordance with one embodiment of the present invention;

Figure 4 shows a block diagram of an apparatus for picture cropping in accordance with one embodiment of the present invention;

FIG. 5 is a flow chart showing a method for cropping a character picture according to an embodiment of the present invention; FIG.

FIG. 6 is a flow chart showing a method for cropping a person's picture according to an embodiment of the present invention; FIG.

FIG. 7 is a flow chart showing a method for cropping a character picture according to an embodiment of the present invention; FIG.

FIG. 8 is a schematic diagram showing the cropping of a character picture cropping method according to an embodiment of the present invention; FIG.

FIG. 9 is a flow chart showing a method for cropping a person's picture according to an embodiment of the present invention; FIG.

FIG. 10 is a block diagram showing a person picture cropping apparatus according to an embodiment of the present invention; FIG.

Figure 11 is a block diagram showing a person picture cropping apparatus according to an embodiment of the present invention;

FIG. 12 is a block diagram showing a person picture cropping apparatus according to an embodiment of the present invention; FIG.

13 shows a flow chart of a method of cropping a picture containing a graphical logo, in accordance with one embodiment of the present invention;

14 shows a flow chart of a method of cropping a picture containing a graphical logo, in accordance with one embodiment of the present invention;

15 shows a flow chart of a method of cropping a picture containing a graphical logo, in accordance with one embodiment of the present invention;

16 shows a flow chart of a method of cropping a picture containing a graphical logo, in accordance with one embodiment of the present invention;

17 illustrates an original picture cropped by a method of cropping a picture containing a graphic mark, in accordance with one embodiment of the present invention;

FIG. 18 illustrates a target picture cropped by a method of cropping a picture including a graphic mark according to an embodiment of the present invention; FIG.

19 shows a block diagram of an apparatus for cropping a picture containing a graphical logo, in accordance with one embodiment of the present invention;

20 shows a block diagram of an apparatus for cropping a picture containing a graphical logo, in accordance with one embodiment of the present invention;

21 shows a block diagram of an apparatus for cropping a picture containing a graphical logo, in accordance with one embodiment of the present invention;

22 shows a block diagram of an apparatus for cropping a picture containing a graphical logo, in accordance with one embodiment of the present invention;

Figure 23 is a flow chart showing a method of detecting a graphic mark in accordance with one embodiment of the present invention;

Figure 24 is a flow chart showing a method of detecting a graphic mark in accordance with one embodiment of the present invention;

Figure 25 is a flow chart showing a method of detecting a graphic mark in accordance with one embodiment of the present invention;

Figure 26 is a flow chart showing a method of detecting a pattern mark in accordance with one embodiment of the present invention;

Figure 27 is a flow chart showing a method of detecting a pattern mark in accordance with one embodiment of the present invention;

28 is a diagram showing the operation of a method of detecting a pattern mark according to an embodiment of the present invention;

Figure 29 is a block diagram showing a graphic mark detecting apparatus according to an embodiment of the present invention;

Figure 30 is a block diagram showing a graphic mark detecting apparatus according to an embodiment of the present invention;

Figure 31 is a block diagram showing a graphic mark detecting apparatus according to an embodiment of the present invention;

Figure 32 is a block diagram showing a graphic mark detecting apparatus according to an embodiment of the present invention.

Figure 33 is a schematic block diagram showing a computing device for performing the method according to the present invention;

Fig. 34 schematically shows a storage unit for holding or carrying program code implementing the method according to the invention.

Specific embodiment

Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While the embodiments of the present invention have been shown in the drawings, the embodiments Rather, these embodiments are provided so that this disclosure will be more fully understood and the scope of the disclosure will be fully disclosed.

As shown in FIG. 1 , an embodiment of the present invention provides a method for cropping a picture, including:

Step 110: The original picture to be cropped is detected, and the feature elements included in the original picture are detected. In the present embodiment, the type of the feature element is not limited, and includes, but is not limited to, a person's face and a graphic mark, for example, a mark of a commodity or a company.

Step 120: Select a cropping method for cropping the original image according to the feature element. In this embodiment, the clipping mode is not limited. For example, if the feature element in the photo of the person is the face, the corresponding clipping mode is the picture containing the face under the cropping; the feature element in the product logo image is the graphic function of the marking. Or the background color, copy the entire content of the image into a certain area for cropping, and fill the remaining area when copying the picture with the background color.

In step 130, the original image is cropped in a cropping manner to obtain a target image. According to the technical solution of the embodiment, the original image is first detected before the image is cropped, and the feature elements included therein are detected, and the different cropping manners are used for cropping according to different feature elements, thereby ensuring timely adoption of different images. The appropriate cutting method makes the target image obtained by the cropping meet the demand.

Further, feature elements are retained in the target image. This is because the usual feature elements are important parts of the original picture, such as the face in the picture of the person, the graphic or background color in the graphic logo picture, and it is necessary to keep it.

According to FIG. 1, the news photo of the character A is detected, and based on the face recognition technology, the face of the character A can be detected; when the photo is cropped, the area of the face of the character A is cropped from the photo, and the rest is discarded; The cropped image has the face of the character A, which is good for indicating the identity of the character A, and matches the news of the character A.

As shown in FIG. 2, an embodiment of the present invention provides a method for cropping a picture, including:

Step 210: Detect whether a human face is included in the original picture.

Step 220: When the original picture includes a human face, select a clipping mode for cropping the original picture according to the face.

Step 230: When the original picture does not include a human face, it is detected whether the original picture contains a graphic mark. In this embodiment, the face has a higher detection priority than the graphic mark. This is because the picture with the face and the graphic mark at the same time is usually the main face, so the face is preferentially detected. Cropped.

Step 240: When the original picture includes the graphic mark, select a cutting mode for cropping the original picture according to the graphic mark.

Step 250: When the original picture does not include the face and the graphic mark, the preset universal cutting mode is selected as the cutting mode for cutting the original picture. In this embodiment, the general clipping mode is not limited. Here, a clipping mode that can adapt to more types of pictures is provided: the original image is compressed by a minimum compression ratio, and the minimum scaling ratio is calculated as follows:

Min_scale=MIN(width _src /width _dst ,height _src /height _dst )

Let the width and height of the uncompressed original image be width _src , height _src ; the width and height of the target image are width _dst and height _{dst respectively} .

The pixels in the scaled image that are more than the target image range are cropped to obtain the target image. This clipping method is useful for cropping the non-critical edge portion of the original image.

It should be noted that each of the above formulas is not the only formula for implementing the present invention, and is merely an implementation of the embodiment. It is still within the scope of the present invention that the skilled person can appropriately deform the formula according to the business needs, such as adding constants or variables or coefficients.

In step 260, the original image is cropped in a cropping manner to obtain a target image.

According to Figure 2, in a photo of a star news report, a star gets on and off the vehicle, and the vehicle has a car logo. Obviously, the focus of the photo is a star; at this time, the face detection technology is first used for face detection and detection. To the face of the star, the face of the star can be cropped when cropping, and the star can be identified in the cropped image, which is compatible with the news report.

As shown in FIG. 3, an embodiment of the present invention provides a method for cropping a picture, including:

Step 310: Detect whether a human face is included in the original picture.

Step 320, counting the number of faces when detecting a face from the original picture.

In step 330, when the number of faces in the original picture is 1, the corresponding clipping mode is selected according to the face.

Step 340: When the number of faces in the original picture is greater than 1, the preset universal cutting mode is selected as the cutting mode for cropping the original picture. In the present embodiment, when the number of faces is greater than 1, the personal face is usually not the main content, for example, in the case of multi-person photos, so it is not necessary to crop each person's face.

Step 350: When the original picture does not include a human face, it is detected whether the original picture contains a graphic mark.

Step 360: When the original picture includes the graphic mark, select a cutting mode for cutting the original picture according to the graphic mark.

When the face and the graphic mark are not included in the original picture, the process also proceeds to step 340: the preset universal cutting mode is selected as the cutting mode for cropping the original picture.

In step 370, the original image is cropped in a cropping manner to obtain a target image.

According to FIG. 3, for a certain group photo, after the face detection is performed and the number of faces is found to be 20, the general cropping strategy of the foregoing embodiment is used for cropping, and the whole body part of 20 people in the cropped picture is basically retained. .

As shown in FIG. 4, an embodiment of the present invention provides an apparatus for cropping pictures, including:

The feature element detecting module 410 detects the original image to be cropped, and detects the feature element included in the original image. In the present embodiment, the type of the feature element is not limited, and includes, but is not limited to, a person's face and a graphic mark, for example, a mark of a commodity or a company.

The cropping mode selection module 420 selects a cropping method for cropping the original image according to the feature element. In this embodiment, the clipping mode is not limited. For example, if the feature element in the photo of the person is the face, the corresponding clipping mode is the picture containing the face under the cropping; the feature element in the product logo image is the graphic function of the marking. Or the background color, copy the entire content of the image into a certain area for cropping, and fill the remaining area when copying the picture with the background color.

The cropping module 430 crops the original image in a cropping manner to obtain a target image. According to the technical solution of the embodiment, the original image is first detected before the image is cropped, and the feature elements included therein are detected, and the different cropping manners are used for cropping according to different feature elements, thereby ensuring timely adoption of different images. The appropriate cutting method makes the target image obtained by the cropping meet the demand.

According to FIG. 4, the news photo of the character A is detected, and based on the face recognition technology, the face of the character A can be detected; when the photo is cropped, the area of the face of the character A is cropped from the photo, and the rest is discarded; The cropped image has the face of the character A, which is good for indicating the identity of the character A, and matches the news of the character A.

An embodiment of the present invention provides a device for cropping a picture, comprising:

The feature element detecting module 410 detects whether a face is included in the original picture.

The cropping mode selection module 420 selects a cropping method for cropping the original image according to the face when the original image includes the human face.

The feature element detecting module 410 detects whether a graphic flag is included in the original image when the original image does not include a human face. In this embodiment, the face has a higher detection priority than the graphic mark. This is because the picture with the face and the graphic mark at the same time is usually the main face, so the face is preferentially detected. Cropped.

The crop mode selection module 420 selects a crop mode for cropping the original image according to the graphic flag when the original picture includes the graphic mark.

The cropping mode selection module 420 selects a preset general cropping mode as a cropping manner for cropping the original image when the original image does not include a face and a graphic logo. In this embodiment, the general clipping mode is not limited. Here, a clipping mode that can adapt to more types of pictures is provided: the original image is compressed by a minimum compression ratio, and the minimum scaling ratio is calculated as follows:

Min_scale=MIN(width _src /width _dst ,height _src /height _dst )

The cropping module 430 crops the original image in a cropping manner to obtain a target image.

According to Figure 4, in a photo of a star news report, a star gets on and off the vehicle, and the vehicle has a car logo. Obviously, the focus of the photo is a star; at this time, the face detection technology is first used for face detection and detection. To the face of the star, the face of the star can be cropped when cropping, and the star can be identified in the cropped image, which is compatible with the news report.

The feature element detecting module 410 detects whether a face is included in the original picture, and counts the number of faces when the face is detected from the original picture.

The cropping mode selection module 420 selects a corresponding cropping mode according to the face when the number of faces in the original picture is 1.

When the number of faces in the original picture is greater than 1, the cutting mode selection module 420 selects a preset universal cutting mode as the original picture. The cutting method for cropping. In the present embodiment, when the number of faces is greater than 1, the personal face is usually not the main content, for example, in the case of multi-person photos, so it is not necessary to crop each person's face.

The feature element detecting module 410 detects whether a graphic flag is included in the original image when the original image does not include a human face.

The cropping mode selection module 420 selects a preset general cropping mode as a cropping manner for cropping the original image when the original image does not include a face and a graphic logo.

According to FIG. 4, for a certain group photo, after the face detection is performed and the number of faces is 20, the general cropping strategy of the foregoing embodiment is used for cropping, and the whole body part of 20 people in the cropped picture is basically retained. .

For different images, the content that needs to be cropped in the image is different because the content displayed in the image is different. For example, for the character image, the size and position of the characters in different images are different. At present, the prior art generally adopts a three-part composition and a golden ratio composition for cropping, and uniformly adopts a method of cutting a fixed position in a picture. The cutting method using the prior art often causes the cropped character to be incomplete or incomplete, and thus It is difficult to ensure that the cropped image is satisfactory, so that the cropped image is often unusable, or it has to be manually cropped, which is inefficient and not widely used. In this regard, the present invention provides the following scheme.

As shown in FIG. 5, an embodiment of the present invention provides a method for cropping a character image, including:

In step 510, the original picture of the person to be cropped is detected to detect the face of the person. In the present embodiment, a human face can be recognized based on the existing face recognition technology.

In step 520, the original image is cropped to obtain a target image, and at least a human face is retained in the target image. According to the technical solution of the embodiment, the human face is a key part for recognizing the character, so the target image under the cropping can still accurately reflect the user's identity; through the face detection, the image can be accurately positioned regardless of the size and position of the character in the original image. The face can be cropped as long as the face is determined, and the cropping effect is not affected by the size and position of the person in the original picture.

According to FIG. 5, the original picture is a full-body picture of a person, based on the face recognition technology, the face position can be recognized from the photo; the full-length picture is cropped, and the target picture including the face area is cropped.

As shown in FIG. 6, an embodiment of the present invention provides a method for cropping a character image, which includes:

In step 610, the original picture with the character to be cropped is detected to detect the face of the person.

Step 620: Obtain a size of the target image, and obtain a size of the face in the target image.

Step 630: The original image is scaled according to the size of the face in the target picture to adjust the size of the face in the original picture to the size of the face in the target picture.

In step 640, according to the size of the target picture, the area where the original picture is cropped is selected, and cropped to obtain the target picture. According to the technical solution of the embodiment, whether the face in the original picture is too large or too small, it can be presented in the target picture in an appropriate size.

According to FIG. 6, the original picture is a full-body photo of a person, based on the face recognition technology, the face position can be recognized from the photo; the whole body photo is scaled according to the required face size; and then the whole body photo is taken Crop, crop the target image containing the face area.

As shown in FIG. 7, an embodiment of the present invention provides a method for cropping a character image, including:

In step 710, the original picture of the person to be cropped is detected to detect the face of the person.

Step 720, detecting the size of the face in the original picture.

Step 730: Obtain a size of the target image, and calculate a size of the face in the target image according to a proportional relationship between the preset target image and the face thereof, and a size of the target image.

Step 740: Calculate the scale of scaling the original image according to the size of the face in the original picture and the size of the face in the target picture, and scale according to the scale.

Step 750: Select an area to be cropped by the original picture according to the size of the target picture, and perform cropping to obtain the target picture.

According to Figure 7, the original picture is recorded as image _src , and its width and height are respectively width _src and height _src . The target picture to be obtained by clipping is image _dst , and its width and height are width _dst and heigth _dst , respectively. Pixel. Let the width and height of the face in the original image detected are face_wdith _src and face_height _src , and the image is scaled proportionally. The width and height of the face in the compressed image are face_width, face_height, and the compressed image is compressed. Satisfaction: width _dst = 3 * face_width, according to the relationship can calculate the size of the face in the target picture, as shown in Figure 8, the slash part is the area where the face is located.

As shown in FIG. 9, an embodiment of the present invention provides a method for cropping a character image, including:

In step 910, the original picture of the person to be cropped is detected to detect the face of the person.

Step 920: Obtain a size of the target image, and obtain a size of the face in the target image.

Step 930: The original image is scaled according to the size of the face in the target picture to adjust the size of the face in the original picture to the size of the face in the target picture.

Step 940, determining the position of the cropped area according to the size of the target picture, the size of the face in the target picture, and the positional relationship of the face in the preset target picture, and cutting according to the determined position. According to the technical solution of the embodiment, the face can be limited to a desired position in the target picture.

In this embodiment, the positional relationship is not limited, and may include: along the width direction of the face, the distance between the two sides of the face and the two sides of the target picture are equal; along the height direction of the face, The distance between the top edge of the face and the top edge of the target image is less than the distance between the bottom edge of the face and the bottom edge of the target image. According to the technical solution of the embodiment, the position defined by the face is adapted to better display the face.

According to FIG. 9, in combination with the foregoing embodiment, the compressed picture is cropped so that in the cropped picture, the face is horizontally centered on the picture, and the vertical position is calculated from top to bottom (heigth _dst -face_height)/3. Specifically, as shown in Figure 8.

As shown in FIG. 10, an embodiment of the present invention provides a person picture cropping apparatus, including:

The face detection module 1010 detects the original picture of the person to be cropped to detect the face of the person. In the present embodiment, a human face can be recognized based on the existing face recognition technology.

The cropping module 1020 crops the original image to obtain a target image, and at least a human face is retained in the target image. According to the technical solution of the embodiment, the human face is a key part for recognizing the character, so the target image under the cropping can still accurately reflect the user's identity; through the face detection, the image can be accurately positioned regardless of the size and position of the character in the original image. The face can be cropped as long as the face is determined, and the cropping effect is not affected by the size and position of the person in the original picture.

According to FIG. 10, the original picture is a full-body picture of a person, based on the face recognition technology, the face position can be recognized from the photo; the full-length picture is cropped, and the target picture including the face area is cropped.

As shown in FIG. 11, an embodiment of the present invention provides a person picture cropping apparatus, which includes:

The face detection module 1110 detects the original picture of the person to be cropped to detect the face of the person.

The size obtaining module 1120 acquires the size of the target image and acquires the size of the face in the target image.

The scaling module 1130 scales the original image according to the size of the face in the target picture to adjust the size of the face in the original picture to the size of the face in the target picture.

The cropping module 1140 selects an area for cropping the original image according to the size of the target image, and performs cropping to obtain the target image. According to the technical solution of the embodiment, whether the face in the original picture is too large or too small, it can be presented in the target picture in an appropriate size.

According to FIG. 11, the original picture is a full-body photo of a person, based on the face recognition technology, the face position can be recognized from the photo; the whole body photo is scaled according to the required face size; and then the whole body photo is taken Crop, crop the target image containing the face area.

As shown in FIG. 12, an embodiment of the present invention provides a person picture cropping apparatus, including:

The face detection module 1210 detects the original picture of the person to be cropped to detect the face of the person.

The size detecting module 1220 detects the size of the face in the original picture.

The size obtaining module 1230 obtains the size of the target image, and calculates the size of the face in the target image according to the proportional relationship between the preset target image and the face thereof and the size of the target image.

The scaling module 1240 calculates a scale for scaling the original image according to the size of the face in the original image and the size of the face in the target image, and scales according to the scale.

The cropping module 1250 selects an area for cropping the original image according to the size of the target image, and performs cropping to obtain the target image.

According to Fig. 12, the original picture is recorded as image _src , and its width and height are respectively width _src and height _src . The target picture to be obtained by cropping is recorded as image _dst , and its width and height are respectively width _dst and heigth _dst , and the units are Pixel. Let the width and height of the face in the original image detected are face_wdith _src and face_height _src , and the image is scaled proportionally. The width and height of the face in the compressed image are face_width, face_height, and the compressed image is compressed. Satisfaction: width _dst = 3 * face_width, according to the relationship can calculate the size of the face in the target picture, as shown in Figure 8, the slash part is the area where the face is located.

An embodiment of the present invention provides a person picture cropping apparatus, including:

The cropping module 1140 determines the position of the cropped region according to the size of the target image, the size of the face in the target image, and the positional relationship of the face in the preset target image, and performs cropping according to the determined position. According to the technical solution of the embodiment, the face can be limited to a desired position in the target picture.

According to FIG. 11, in combination with the foregoing embodiment, the compressed picture is cropped so that in the cropped picture, the face is horizontally centered on the picture, and the vertical position is calculated from top to bottom (heigth _dst -face_height)/3. Specifically, as shown in Figure 8.

For different images, the content that needs to be cropped in the image is different because the content displayed in the image is different. For example, for the image containing the graphic logo, how to crop the image so that the target image after cropping still plays Marking is a problem. At present, the method of cutting with a three-part composition and a golden ratio composition is to cut the fixed position in the picture, and the damage to the graphic mark is great, so that it is difficult to ensure that the cut picture is satisfactory, or Manually re-cutting is not efficient. In this regard, the present invention provides the following scheme.

As shown in FIG. 13, an embodiment of the present invention provides a method for cropping a picture including a graphic mark, which includes:

Step 1310: The original picture to be cropped is detected, and some or all of the features of the one or more graphic marks included in the original picture are detected. In the present embodiment, the features of the graphic mark are not limited, and include, but are not limited to, the color and shape of the graphic mark itself, and the background color of the graphic mark.

In step 1320, the original picture is cropped to obtain a target picture, and some or all of the features of the at least one graphic mark are retained in the target picture. In this embodiment, the retained features may not be the same as the detected features. According to the technical solution of the present embodiment, when detecting that the original picture is a picture including a graphic mark, the cut target picture is required to include some or all of the technical features of at least one graphic mark; the graphic mark generally indicates a specific meaning, for example, Indicates the brand of the product, etc., so retaining the characteristics of the graphic mark can make the target image obtained by the crop still reflect a specific meaning.

According to Fig. 13, on the picture of a trademark (graphic mark); after the mark is identified, when the picture of the mark is cut, the area where the mark is located (all features) is cut, or the outline shape of the mark is extracted (partially Feature); further, when there are multiple similar trademarks in the picture, it can be tailored only according to the trademark of one of the products.

As shown in FIG. 14, an embodiment of the present invention provides a method for cropping a picture including a graphic mark, which includes:

Step 1410: The original picture to be cropped is detected, and some or all of the features of the one or more graphic marks included in the original picture are detected.

In step 1420, the size of the target picture is obtained.

In step 1430, the original image is scaled according to the size of the target image, so that the size of the original image does not exceed the size of the target image. In this embodiment, the size of the original picture is not more than the size of the target picture, which is advantageous for retaining all the features of the original picture.

In step 1440, the cropping area is drawn according to the size of the target image, and the image of the original image is copied into the cropping area, and the cropping area is cropped to obtain the target image, so as to retain all the features of all the graphic markers in the original image. According to the technical solution of the embodiment, after the original image is scaled and then cropped, all the images of the original image can be retained, that is, all the features of the graphic logo can be saved, which is beneficial to the complete reflection of the meaning of the graphic logo after the cropping. Further, when the image of the original image cannot fill the cropping area, the image of the original image is centered in the cropping area, which is beneficial to ensure that the cropped image exhibits a symmetrical effect.

As shown in FIG. 15, an embodiment of the present invention provides a method for cropping a picture including a graphic mark, which includes:

Step 1510: The original picture to be cropped is detected, and some or all of the features of the one or more graphic marks included in the original picture are detected.

In step 1520, the height and width of the original picture are detected.

Step 1530 includes the height and width of the original picture.

Step 1540, calculating a ratio of the height of the original picture to the height of the target picture, and calculating a ratio of the width of the original picture to the width of the target picture, and using the maximum value of the obtained ratio as a ratio of scaling the original picture, and Scale the original image proportionally. In this embodiment, the scaling is actually performed according to the maximum compression ratio. This method can ensure that the size of the original image is smaller than the size of the target image, and the original image size is the largest, thereby facilitating the clear display of the target image. The characteristics of the logo.

In step 1550, the cropping area is drawn according to the size of the target image, and the image of the original image is copied into the cropping area, and the cropping area is cropped to obtain the target image, so as to retain all the features of all the graphic markers in the original image.

According to Fig. 15, the original picture is recorded as image _src , and its width and height are respectively width _src and height _src . The target picture to be obtained by cropping is recorded as image _dst , and its width and height are respectively width _dst and heigth _dst , and the units are Pixel. The original image is compressed according to the maximum compression ratio, and the maximum compression ratio is calculated as: max_scale=MAX(width _src /width _dst , height _src /height _dst ).

Let the image obtained after compression be image _resize , and its width and height are width _resize and height _resize respectively:

Width _resize =width _src /max_scale;height _resize =height _src /max_scale.

As shown in FIG. 16, in an embodiment of the present invention, a method for cropping a picture including a graphic mark is provided, which includes:

Step 1610: The original picture to be cropped is detected, and some or all of the features of the one or more graphic marks included in the original picture are detected.

In step 1620, the size of the target picture is obtained.

In step 1630, the original image is scaled according to the size of the target image, so that the size of the original image does not exceed the size of the target image. In this embodiment, the size of the original picture is not more than the size of the target picture, which is advantageous for retaining all the features of the original picture.

In step 1640, the cropping area is drawn according to the size of the target image, and the image of the original image is copied into the cropping area, and the cropping area is cropped to obtain the target image, so as to retain all the features of all the graphic markers in the original image.

In step 1650, when the image of the original image cannot be filled with the cropping area, the area where the background color is to be extracted is selected from the original image and the color is extracted according to the edge position of the original image. In this embodiment, the original picture edge that usually contains the graphic mark is the background of the graphic mark.

Step 1660: Filling a portion of the cropped area that is not filled by the image of the original image according to the background color of the graphic mark retained in the target picture. According to the technical solution of the embodiment, the background color is selected for filling, so that the color of the filling is substantially consistent with the color of the edge of the copied image, and the overall effect is very good.

According to FIG. 16, in combination with the foregoing embodiment, when the image _resize pixel value is copied to the middle of the image _dst , if the image _resize size is smaller than the image _dst , that is, the width _resize <width _dst or the height _resize <heigth _dst , the missing The pixels are filled with the background color.

According to the above two embodiments, the logo image shown in Fig. 17 has an original size of 194 x 60. Now that you want to get a 100x100 image, you need to deal with it as follows: a) First detect the image as a logo image and make sure the background color is pure white. b) Calculate the maximum compression ratio, max_sacle=MAX(192/100, 60/100)=1.92. c) Compress the picture at the maximum compression ratio. The length and width of the compressed picture are 100 and 31 respectively. d) The compressed picture pixel is centered and copied into the target picture, and the missing pixels are filled with the background color to obtain a picture as shown in FIG.

As shown in FIG. 19, an embodiment of the present invention provides an apparatus for cropping a picture including a graphic mark, including:

The graphic mark detection module 1910 detects the original picture to be cropped, and detects some or all of the features of the one or more graphic marks included in the original picture. In the present embodiment, the features of the graphic mark are not limited, and include, but are not limited to, the color and shape of the graphic mark itself, and the background color of the graphic mark.

The cropping module 1920 crops the original image to obtain a target image, and retains some or all of the features of the at least one graphic identifier in the target image. In this embodiment, the retained features may not be the same as the detected features. According to the technical solution of the present embodiment, when detecting that the original picture is a picture including a graphic mark, the cut target picture is required to include some or all of the technical features of at least one graphic mark; the graphic mark generally indicates a specific meaning, for example, Indicates the brand of the product, etc., so retaining the characteristics of the graphic mark can make the target image obtained by the crop still reflect a specific meaning.

According to Fig. 19, on the picture of a trademark (graphic mark); after the mark is identified, when the picture of the mark is cut, the area where the mark is located (all features) is cut, or the outline shape of the mark is extracted (partially Feature); further, when there are multiple similar trademarks in the picture, it can be tailored only according to the trademark of one of the products.

As shown in FIG. 20, an embodiment of the present invention provides an apparatus for cropping a picture including a graphic mark, including:

The graphic mark detection module 2010 detects the original picture to be cropped, and detects some or all of the features of the one or more graphic marks included in the original picture.

The size obtaining module 2020 acquires the size of the target picture.

The scaling module 2030 scales the original image according to the size of the target image so that the size of the original image does not exceed the size of the target image. In this embodiment, the size of the original picture is not more than the size of the target picture, which is advantageous for retaining all the features of the original picture.

The cropping module 2040 divides the cropping area according to the size of the target image, and copies the image of the original image into the cropping area, and crops the cropping area to obtain the target image, so as to retain all the features of all the graphic markers in the original image. According to the technical solution of the embodiment, after the original image is scaled and then cropped, all the images of the original image can be retained, that is, all the features of the graphic logo can be saved, which is beneficial to the complete reflection of the meaning of the graphic logo after the cropping. Further, when the image of the original image cannot fill the cropping area, the image of the original image is centered in the cropping area, which is beneficial to ensure that the cropped image exhibits a symmetrical effect.

As shown in FIG. 21, an embodiment of the present invention provides an apparatus for cropping a picture including a graphic mark, including:

The graphic mark detecting module 2110 detects the original picture to be cropped, and detects some or all of the features of the one or more graphic marks included in the original picture.

The size detecting module 2120 detects the height and width of the original picture.

The size acquisition module 2130 includes the height and width of the original picture.

The scaling module 2140 calculates a ratio of the height of the original picture to the height of the target picture, and calculates a ratio of the width of the original picture to the width of the target picture, and uses the maximum value of the obtained ratio as a ratio of scaling the original picture. And scale the original image proportionally. In this embodiment, the scaling is actually performed according to the maximum compression ratio. This method can ensure that the size of the original image is smaller than the size of the target image, and the original image size is the largest, thereby facilitating the clear display of the target image. The characteristics of the logo.

The cropping module 2150 divides the cropping area according to the size of the target image, and copies the image of the original image into the cropping area, and crops the cropping area to obtain the target image, so as to retain all the features of all the graphic markers in the original image.

According to Fig. 21, the original picture is recorded as image _src , and its width and height are respectively width _src and height _src . The target picture to be obtained by cropping is recorded as image _dst , and its width and height are respectively width _dst and heigth _dst , and the units are Pixel. The original image is compressed according to the maximum compression ratio, and the maximum compression ratio is calculated as: max_scale=MAX(width _src /width _dst , height _src /height _dst ).

Width _resize =width _src /max_scale;height _resize =height _src /max_scale.

As shown in FIG. 22, an embodiment of the present invention provides an apparatus for cropping a picture including a graphic mark, including:

The graphic mark detection module 2210 detects the original picture to be cropped, and detects some or all of the features of the one or more graphic marks included in the original picture.

The size obtaining module 2220 acquires the size of the target picture.

The scaling module 2230 scales the original image according to the size of the target image so that the size of the original image does not exceed the size of the target image. In this embodiment, the size of the original picture is not more than the size of the target picture, which is advantageous for retaining all the features of the original picture.

The cropping module 2240 divides the cropping area according to the size of the target image, and copies the image of the original image into the cropping area, and crops the cropping area to obtain the target image, so as to retain all the features of all the graphic markers in the original image.

The color filling module 2250 selects an area where the background color to be extracted is selected from the original picture and performs color extraction when the image of the original picture cannot fill the cropping area, according to the edge position of the original picture. In this embodiment, the original picture edge that usually contains the graphic mark is the background of the graphic mark.

The color fill module 2250 fills a portion of the crop region that is not filled with the image of the original image according to the background color of the graphic mark retained in the target image. According to the technical solution of the embodiment, the background color is selected for filling, so that the color of the filling is substantially consistent with the color of the edge of the copied image, and the overall effect is very good.

According to FIG. 22, in combination with the foregoing embodiment, when the image _resize pixel value is copied to the middle of the image _dst , if the image _resize size is smaller than the image _dst , that is, the width _resize <width _dst or the height _resize <heigth _dst , the missing will be The pixels are filled with the background color.

In general, how the image is processed depends on what is contained in the image, so the content contained in the image needs to be detected. For example, a basic current face recognition technology can detect a face included in a person's picture. However, for a picture containing a graphic mark, for example, a product mark, a logo picture, etc., there is currently no technical solution for effectively identifying it, and the method of manual selection can only be generally adopted, and the efficiency is not high. In this regard, the present invention provides the following scheme.

As shown in FIG. 23, an embodiment of the present invention provides a method for detecting a graphic mark picture, which includes:

Step 2310: Select an area of the color to be extracted from the picture based on the edge position of the picture to be detected. In this embodiment, by observing a large number of pictures including graphic marks, it can be concluded that the color of the edge position of the graphic mark picture is generally the background color. In order to reduce the interference of the middle color of the picture with the background color, only the background color of the edge position area is extracted. .

In step 2320, the colors in the area are extracted.

Step 2330, determining whether the graphic flag is included in the picture according to the color in the extracted area. According to the technical solution of the embodiment, in the picture containing the graphic mark, usually the middle part of the picture is a graphic mark, in order to highlight the graphic mark, in addition to the graphic in the general picture The color of the background portion other than the logo is relatively single and lacks change. The technical solution of the embodiment is based on the feature of the image including the graphic mark, and the color is detected according to the edge position of the image to effectively identify whether the extracted color is the background color, and then judge Whether the picture contains a graphic logo.

According to FIG. 23, for a picture with a vehicle logo, a white vehicle logo is centered in the picture, and the background portion of the non-vehicle logo is black; the color is extracted at the edge position of the picture, and the extracted colors are all black, which is consistent with Contains the characteristics of the background color of the graphic mark, so it is judged that the picture contains the graphic mark.

As shown in FIG. 24, an embodiment of the present invention provides a method for detecting a graphic mark picture, which includes:

Step 2410: Select an area of the color to be extracted from the picture based on the edge position of the picture to be detected.

Step 2420, extracting luminance values of each pixel in the region under one or more color channels. Further, the one or more color channels include at least one of the following: a red channel, a green channel, and a blue channel. In the present embodiment, by extracting the brightness value of the multi-color channel, it is advantageous to accurately judge the difference in the difference of the extracted colors.

Step 2430: Determine, according to the color in the extracted area, whether the graphic flag is included in the picture.

As shown in FIG. 25, an embodiment of the present invention provides a method for detecting a graphic mark picture, which includes:

Step 2510: Select an area of the color to be extracted from the picture based on the edge position of the picture to be detected.

Step 2520, extracting luminance values of each pixel in the region under one or more color channels.

Step 2530, counting the number of pixels having the same brightness value under each color channel.

Step 2540, calculating the variance of the number of pixels under each color channel according to the number of pixels of each brightness value under each color channel.

Step 2550: Determine whether the color of the area is the background color of the graphic mark according to the variance of the number of pixels of the one or more channels. When the determination result is YES, the picture includes the graphic mark. According to the technical solution of the embodiment, the variance of the number of pixels is a value that can reflect the change in the color difference, and based on the change of the value, it can be accurately determined whether the extracted color is the background color.

As shown in FIG. 26, an embodiment of the present invention provides a method for detecting a graphic mark picture, which includes:

Step 2610: Select an area of the color to be extracted from the picture based on the edge position of the picture to be detected.

Step 2620, extracting luminance values of each pixel in the region under one or more color channels.

At step 2630, the number of pixels having the same brightness value under each color channel is counted.

Step 2640, calculating the proportion of the number of pixels of each brightness value under each color channel according to the number of pixels of the same brightness value under each color channel.

Step 2650: Calculate an average proportion of the number of pixels of all the brightness values under each color channel according to the proportion of the number of pixels of each brightness value under each color channel.

Step 2660: Calculate the variance of the number of pixels in each color channel according to the proportion of the number of pixels of each brightness value in each color channel and the average proportion of the number of pixels of all the brightness values in each color channel. .

Step 2670: Determine whether the color of the area is the background color of the graphic mark according to the variance of the number of pixels of the one or more channels. When the determination result is YES, the picture includes the graphic mark. According to the technical solution of the embodiment, a detailed technical solution for calculating the variance is provided.

According to Figure 26, for a picture to be detected, it needs to be processed as follows:

1. Count the number of brightness values of each channel: for the three channels R, G, and B, count the number of pixels from 0 to 255 for each brightness value, and record it as a ₀ , a ₁ , a ₂ ,... , a ₂₅₅ , the highest number of brightness values is recorded as the brightness value of the color of the channel.

2. Calculate the proportion of each brightness value. Let the total number of pixels per channel be total, then:

Calculate the proportion of the number of individual pixel values, denoted as R ₀ , R ₁ , R ₂ ,..., R ₂₅₅ , then:

R _i = (a _i /total)*100

3. Calculate the variance. For each channel, the variances of R ₀ , R ₁ , R ₂ , ..., R ₂₅₅ are calculated, respectively, as D _R , D _G , D _B . Note that the average ratio of each channel is avg, and the final variance is D _finall , then:

D _finall =(D _R +D _G +D _B )/3

4. Determine if the picture contains a graphic mark. The greater the difference above the theory, the more likely the picture contains graphical signs, the following is in 2000 The test results on the picture, the results obtained by setting different variance thresholds are as follows:

阈值Threshold	召回率Recall rate	准确率Accuracy
D＞1000D>1000	69.50％69.50%	98.40％98.40%
D＞500D>500	85.90％85.90%	96.40％96.40%
D＞300D>300	95.70％95.70%	95.20％95.20%

Therefore, in the present embodiment, the variance > 300 is used as a criterion for judging whether or not a picture contains a graphic mark.

As shown in FIG. 27, an embodiment of the present invention provides a method for detecting a picture of a graphic mark, which includes:

Step 2710: Calculate a distance between an outer edge and an inner edge of the area according to the size of the picture. In this embodiment, an area that may be the background of the graphic mark may be estimated based on the overall calculation of the picture.

Step 2720, selecting an area from the picture according to a distance between an outer edge and an inner edge of the area. In this embodiment, the area is hollow, and an outer edge of the area is coincident with an edge of the picture. In this embodiment, the area around the picture is actually selected. If the picture contains a graphic mark, the area is most likely the background of the graphic mark.

Step 2730, extract the color in the area.

Step 2740, judging whether the graphic flag is included in the picture according to the color in the extracted area.

According to FIG. 27, for a picture to be detected whether or not the graphic mark is included, an area of pixels occupying 1/5 of the original picture width and height may be selected, and a color is extracted therefrom for use in judging, as shown in FIG. The area is the selected area.

As shown in FIG. 29, an embodiment of the present invention provides a graphic mark picture detecting apparatus, including:

The area selection module 2910 selects an area of the color to be extracted from the picture based on the edge position of the picture to be detected. In this embodiment, by observing a large number of pictures including graphic marks, it can be concluded that the color of the edge position of the graphic mark picture is generally the background color. In order to reduce the interference of the middle color of the picture with the background color, only the background color of the edge position area is extracted. .

A color extraction module 2920 extracts colors in the region.

The graphic flag determining module 2930 determines whether the graphic flag is included in the image according to the color in the extracted region. According to the technical solution of the embodiment, in the picture containing the graphic mark, usually, the middle part of the picture is a graphic mark. In order to highlight the graphic mark, the color of the background part other than the graphic mark in the general picture is relatively single and lacks change. The technical solution of the example is based on the feature of the picture containing the graphic mark, and the color is extracted according to the position of the edge of the picture for detecting, which can effectively determine whether the extracted color is the background color, and thereby identify the picture containing the graphic mark.

According to FIG. 29, for a picture with a vehicle logo, a white vehicle logo is centered in the picture, and the background portion of the non-vehicle logo is black; the color is extracted at the edge position of the picture, and the extracted colors are all black, which is consistent with Contains the characteristics of the background color of the graphic mark, so it is judged that the picture contains the graphic mark.

An embodiment of the present invention provides a graphic mark picture detecting apparatus, including:

The area selection module 2910 selects an area of the color to be extracted from the picture based on the edge position of the picture to be detected.

The color extraction module 2920 extracts luminance values of each pixel in the region under one or more color channels. Further, the one or more color channels include at least one of the following: a red channel, a green channel, and a blue channel. In the present embodiment, by extracting the brightness value of the multi-color channel, it is advantageous to accurately judge the difference in the difference of the extracted colors.

The graphic flag determining module 2930 determines whether the graphic flag is included in the image according to the color in the extracted region.

As shown in FIG. 30, an embodiment of the present invention provides a graphic mark picture detecting apparatus, including:

The area selection module 3010 selects an area of the color to be extracted from the picture based on the edge position of the picture to be detected.

The color extraction module 3020 extracts luminance values of each pixel in the region under one or more color channels.

The pixel statistics module 3030 counts the number of pixels having the same brightness value under each color channel.

The variance calculation module 3040 calculates the variance of the number of pixels under each color channel based on the number of pixels of each luminance value under each color channel.

The graphic flag determining module 3050 determines whether the color of the area is the background color of the graphic mark according to the variance of the number of pixels of the one or more channels. When the determination result is YES, the graphic includes the graphic mark. According to the technical solution of the embodiment, the variance of the number of pixels is a value that can reflect the change in the color difference, and based on the change of the value, it can be accurately determined whether the extracted color is the background color.

As shown in FIG. 31, an embodiment of the present invention provides a graphic mark picture detecting apparatus, including:

The area selection module 3110 selects an area of the color to be extracted from the picture based on the edge position of the picture to be detected.

The color extraction module 3120 extracts luminance values of each pixel in the region under one or more color channels.

The pixel statistics module 3130 counts the number of pixels having the same brightness value under each color channel.

The ratio calculation module 3140 calculates the proportion of the number of pixels of each brightness value under each color channel according to the number of pixels of the same brightness value under each color channel.

The average ratio calculation module 3150 calculates an average proportion of the number of pixels of all the luminance values under each color channel according to the proportion of the number of pixels of each luminance value under each color channel.

The variance calculation module 3160 calculates the number of pixels under each color channel according to the proportion of the number of pixels of each brightness value under each color channel and the average proportion of the number of pixels of all the brightness values under each color channel. Variance.

The graphic mark determining module 3170 determines whether the color of the area is the background color of the graphic mark according to the variance of the number of pixels of the one or more channels, and when the determination result is YES, the graphic includes the graphic mark. According to the technical solution of the embodiment, a detailed technical solution for calculating the variance is provided.

According to Figure 31, for a picture to be detected, it needs to be processed as follows:

R _i = (a _i /total)*100

D _finall =(D _R +D _G +D _B )/3

4. Determine if the picture contains a graphic mark. The larger the theoretical difference is, the more likely the picture contains the graphic mark. The following is the test result on 2000 pictures. The results obtained by setting different variance thresholds are as follows:

As shown in FIG. 32, an embodiment of the present invention provides a graphic mark picture detecting apparatus, including:

The distance calculation module 3210 calculates a distance between an outer edge and an inner edge of the region according to the size of the picture. In this embodiment, an area that may be the background of the graphic mark may be estimated based on the overall calculation of the picture.

The region selection module 3220 selects an area from the picture by the distance between the outer edge and the inner edge of the area. In this embodiment, the area is hollow, and an outer edge of the area is coincident with an edge of the picture. In this embodiment, the area around the picture is actually selected. If the picture contains a graphic mark, the area is most likely the background of the graphic mark.

The color extraction module 3230 extracts colors in the area.

The graphic flag determining module 3240 determines whether the graphic flag is included in the image according to the color in the extracted region.

According to FIG. 32, for a picture to be detected whether or not the graphic mark is included, an area of pixels occupying 1/5 of the original picture width and height may be selected, and a color is extracted therefrom for use in determining, as shown in FIG. The area is the selected area.

The algorithms and displays provided herein are not inherently related to any particular computer, virtual system, or other device. Various general purpose systems can also be used with the teaching based on the teachings herein. The structure required to construct such a system is apparent from the above description. Moreover, the invention is not directed to any particular programming language. It is to be understood that the invention may be embodied in a variety of programming language, and the description of the specific language has been described above in order to disclose the preferred embodiments of the invention.

In the description provided herein, numerous specific details are set forth. However, it is understood that the embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures, and techniques are not shown in detail so as not to obscure the understanding of the description.

Similarly, the various features of the invention are sometimes grouped together into a single embodiment, in the above description of the exemplary embodiments of the invention, Figure, or a description of it. However, the method disclosed is not to be interpreted as reflecting the intention that the claimed invention requires more features than those recited in the claims. Rather, as the following claims reflect, inventive aspects reside in less than all features of the single embodiments disclosed herein. Therefore, the claims following the specific embodiments are hereby explicitly incorporated into the embodiments, and each of the claims as a separate embodiment of the invention.

Those skilled in the art will appreciate that the modules in the devices of the embodiments can be adaptively changed and placed in one or more devices different from the embodiment. The modules or units or components of the embodiments may be combined into one module or unit or component, and further they may be divided into a plurality of sub-modules or sub-units or sub-components. In addition to such features and/or at least some of the processes or units being mutually exclusive, any combination of the features disclosed in the specification, including the accompanying claims, the abstract and the drawings, and any methods so disclosed, or All processes or units of the device are combined. Each feature disclosed in this specification (including the accompanying claims, the abstract and the drawings) may be replaced by alternative features that provide the same, equivalent or similar purpose.

In addition, those skilled in the art will appreciate that, although some embodiments described herein include certain features that are included in other embodiments and not in other features, combinations of features of different embodiments are intended to be within the scope of the present invention. Different embodiments are formed and formed. For example, in the following claims, any one of the claimed embodiments can be used in any combination.

The various component embodiments of the present invention may be implemented in hardware, or in a software module running on one or more processors, or in a combination thereof. Those skilled in the art will appreciate that some or all of the functionality of some or all of the components of the apparatus for cropping pictures in accordance with embodiments of the present invention may be implemented in practice using a microprocessor or digital signal processor (DSP). The invention can also be implemented as a device or device program (e.g., a computer program and a computer program product) for performing some or all of the methods described herein. Such a program implementing the invention may be stored on a computer readable medium or may be in the form of one or more signals. Such signals may be downloaded from an Internet website, provided on a carrier signal, or provided in any other form.

For example, Figure 33 schematically illustrates a block diagram of a computing device for performing the method in accordance with the present invention. The computing device conventionally includes a processor 3310 and a computer program product or computer readable medium in the form of a memory 3320. The memory 3320 may be an electronic memory such as a flash memory, an EEPROM (Electrically Erasable Programmable Read Only Memory), an EPROM, a hard disk, or a ROM. Memory 3320 has a storage space 3330 for program code 3331 for performing any of the method steps described above. For example, the storage space 3330 for program code may include respective program codes 3331 for respectively implementing various steps in the above method. The program code can be read from or written to one or more computer program products. These computer program products include program code carriers such as hard disks, compact disks (CDs), memory cards or floppy disks. Such a computer program product is typically a portable or fixed storage unit as described with reference to FIG. The storage unit may have a storage segment, a storage space, and the like arranged similarly to the storage 3320 in the computing device of FIG. The program code can be compressed, for example, in an appropriate form. Typically, the storage unit comprises computer readable code 3331' for performing the steps of the method according to the invention, ie code that can be read by a processor such as, for example, 3310, which when executed by the computing device causes the calculation The device performs the various steps in the methods described above.

It is to be noted that the above-described embodiments are illustrative of the invention and are not intended to be limiting, and that the invention may be devised without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses shall not be construed as a limitation. The word "comprising" does not exclude the presence of the elements or steps that are not recited in the claims. The word "a" or "an" The invention can be implemented by means of hardware comprising several distinct elements and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means can be embodied by the same hardware item. The use of the words first, second, and third does not indicate any order. These words can be interpreted as names.

In addition, it should be noted that the language used in the specification has been selected for the purpose of readability and teaching, and is not intended to be construed or limited. Therefore, many modifications and changes will be apparent to those skilled in the art without departing from the scope of the invention. The disclosure of the present invention is intended to be illustrative, and not restrictive, and the scope of the invention is defined by the appended claims.

The present invention is applicable to computer systems/servers that can operate with numerous other general purpose or special purpose computing system environments or configurations. Examples of well-known computing systems, environments, and/or configurations suitable for use with computer systems/servers include, but are not limited to, personal computer systems, Server computer system, thin client, thick client, handheld or laptop device, microprocessor based system, set top box, programmable consumer electronics, network personal computer, small computer system, mainframe computer system, and any system including any of the above Distributed cloud computing technology environments, and more.

The computer system/server can be described in the general context of computer system executable instructions (such as program modules) being executed by a computer system. Generally, program modules may include routines, programs, target programs, components, logic, data structures, and the like that perform particular tasks or implement particular abstract data types. The computer system/server can be implemented in a distributed cloud computing environment where tasks are performed by remote processing devices that are linked through a communication network. In a distributed cloud computing environment, program modules may be located on a local or remote computing system storage medium including storage devices.

"an embodiment," or "an embodiment," or "an embodiment," In addition, it is noted that the phrase "in one embodiment" is not necessarily referring to the same embodiment.

Claims

A method of cropping a picture, comprising:

Detecting the cropped original image and detecting the feature elements included in the original image;

Selecting a cropping manner for cropping the original image according to the feature element;

The original picture is cropped according to the cropping manner to obtain a target picture.
The method of claim 1 wherein the feature element comprises a person's face and/or graphical indicia.
The method according to any one of claims 1-2, wherein the original image to be cropped is detected, and the feature elements included in the original image are detected, which specifically includes:

Detecting whether the face is included in the original picture;

When the face is not included in the original picture, it is detected whether the graphic flag is included in the original picture.
The method of any of claims 1-3, further comprising:

When the face and the graphic mark are not included in the original picture, a preset universal cutting mode is selected as a cutting mode for cropping the original picture.
The method according to any one of claims 1 to 4, further comprising: before selecting a cropping manner for cropping the original image according to the picture element, further comprising:

Counting the number of faces when the face is detected from the original picture;

And selecting, according to the picture element, a cutting manner for cutting the original image, specifically:

When the number of the faces in the original picture is 1, the corresponding clipping mode is selected according to the face.
The method of any of claims 1-5, further comprising:

When the number of the faces in the original picture is greater than 1, the preset universal cutting mode is selected as a cutting mode for cropping the original picture.
A method according to any one of claims 1 to 6, wherein

The feature element is retained in the target picture.
A picture cropping device, comprising:

a feature element detecting module, configured to detect the original picture to be cropped, and detect a feature element included in the original picture;

a cutting mode selection module, configured to select a cutting mode for cutting the original image according to the feature element;

And a cropping module, configured to crop the original image according to the cutting manner to obtain a target image.
The apparatus of claim 8 wherein the feature element comprises a person's face and/or a graphical logo.
The apparatus according to any one of claims 8-9, wherein the feature element detecting module detects whether the face is included in the original picture; and when the face is not included in the original picture, Detecting whether the graphic flag is included in the original picture.
The device according to any one of claims 8 to 10, wherein the cropping mode selection module selects a preset general clipping mode as a pair when the human face and the graphic logo are not included in the original picture. The original image is cropped in a cropped manner.
The apparatus according to any one of claims 8 to 11, wherein the feature element detecting module further counts the number of the faces when the face is detected from the original picture; the cutting mode When the number of the faces in the original picture is 1, the selection module selects a corresponding clipping mode according to the face.
The apparatus according to any one of claims 8 to 12, wherein, when the number of the faces in the original picture is greater than 1, the cutting mode selection module selects a preset universal cutting mode as the original The way the image is cropped.
Apparatus according to any of claims 8-13, wherein said feature element is retained in said target picture.
A character picture cropping method, comprising:

Detecting the original picture of the person to be cropped to detect the face of the character;

The original picture is cropped to obtain a target picture, and at least the face is retained in the target picture.
The method according to claim 15, wherein the original picture is cropped to obtain a target picture, and at least the face is retained in the target picture, which specifically includes:

Obtaining a size of the target picture, and acquiring a size of the face in the target picture;

And scaling the original image according to the size of the face in the target picture to adjust a size of the face in the original picture to a size of the face in the target picture;

According to the size of the target picture, an area for cropping the original picture is selected, and cropped to obtain the target picture.
The method according to claim 16, wherein the original picture of the person to be cropped is detected to detect the person The face also includes:

Detecting a size of the face in the original picture;

And scaling the original image according to the size of the face in the target image, specifically:

Calculating a scale for scaling the original image according to a size of the face in the original picture and a size of the face in the target picture, and performing scaling according to the ratio.
The method of claim 16, wherein the obtaining the size of the face in the target picture comprises:

Calculating a size of the face in the target picture according to a preset ratio relationship between the target picture and the face therein, and a size of the target picture.
The method according to claim 16, wherein the area of the original picture is selected according to the size of the target picture, and the area is cropped to obtain the target picture, which specifically includes:

Determining a location of the cropped region according to a size of the target image, a size of the face in the target image, and a preset positional relationship of the face in the target image, and determining The position is cropped.
The method of claim 19, wherein the positional relationship between the face and the target picture comprises:

A distance between two side edges of the human face and two side edges of the target picture is equal along a width direction of the face;

a distance between a top edge of the face and a top edge of the target image is smaller than a bottom edge of the face and a bottom edge of the target image along a height direction of the face The distance between them.
A character picture cropping device includes:

a face detection module, configured to detect a cut original image having a character to detect a face of the character;

And a cropping module, configured to crop the original image to obtain a target image, wherein at least the human face remains in the target image.
The device according to claim 21, further comprising:

a size obtaining module, configured to acquire a size of the target image, and obtain a size of the face in the target image;

a scaling module, configured to scale the original image according to a size of the face in the target picture to adjust a size of the face in the original picture to the face in the target picture size of;

The cropping module selects an area for cropping the original picture according to the size of the target picture, and performs cropping to obtain the target picture.
The apparatus of claim 22, further comprising:

a size detecting module, configured to detect a size of the face in the original picture;

The scaling module calculates a scale for scaling the original image according to the size of the face in the original picture and the size of the face in the target picture, and performs scaling according to the ratio.
The device according to claim 22, wherein

The size obtaining module calculates a size of the face in the target picture according to a preset proportional relationship between the target picture and the face therein, and a size of the target picture.
The device according to claim 22, wherein

The cropping module determines the location of the cropped region according to the size of the target image, the size of the face in the target image, and the preset positional relationship of the face in the target image. And crop according to the determined position.
The apparatus according to claim 25, wherein the positional relationship between the face and the target picture comprises:

A distance between two side edges of the human face and two side edges of the target picture is equal along a width direction of the face;

a distance between a top edge of the face and a top edge of the target image is smaller than a bottom edge of the face and a bottom edge of the target image along a height direction of the face The distance between them.
A method of cropping a picture containing a graphic logo, comprising:

Detecting the cropped original image to detect some or all of the features of the one or more graphic markers included in the original image;

The original picture is cropped to obtain a target picture, and at least one or all of the features of the graphic mark are retained in the target picture.
The method of claim 27, wherein the cropping the original image to obtain a target image comprises:

Obtaining a size of the target image;

And scaling the original picture according to a size of the target picture, such that a size of the original picture does not exceed a size of the target picture;

And cropping the cropped area according to the size of the target picture, and copying the image of the original picture into the cropping area, and cropping the cropped area to obtain the target picture, to retain all the original pictures All the features of the graphic logo.
The method of claim 28, further comprising:

Detecting a size of the original picture, including a height and a width of the original picture;

Obtaining the size of the target image, specifically including:

Obtaining the height and width of the target image;

And scaling the original image according to the size of the target image, specifically:

Calculating a ratio of a height of the original picture to a height of the target picture, and calculating a ratio of a width of the original picture to a width of the target picture, and using a maximum value of the obtained ratio as the original The scale at which the picture is scaled, and the original picture is scaled by the scale.
The method of claim 28, further comprising:

When the image of the original picture cannot fill the cropping area, the portion of the cropping area that is not filled by the image of the original image is filled according to the background color of the graphic mark retained in the target picture. .
The method according to claim 30, wherein before the filling of the portion of the cropped region that is not filled by the image of the original image, according to the background color of the graphic mark retained in the target image, :

According to the edge position of the original picture, the area where the background color is to be extracted is selected from the original picture and color extraction is performed.
The method of any one of claims 28 to 31, further comprising:

When the image of the original picture cannot fill the crop area, the image of the original picture is centered in the crop area.
An apparatus for cropping a picture containing a graphic logo, comprising:

a graphic mark detecting module, configured to detect the original picture to be cropped, and detect some or all features of the one or more graphic marks included in the original picture;

And a cropping module, configured to crop the original image to obtain a target image, and retain at least one or all of the features of the graphic identifier in the target image.
The apparatus of claim 33, further comprising:

a size obtaining module, configured to acquire a size of the target image;

a scaling module, configured to scale the original image according to a size of the target image, such that a size of the original image does not exceed a size of the target image;

a cropping module, configured to draw a cropping area according to a size of the target image, and copy an image of the original image into the cropping area, and crop the cropping area to obtain the target image, to retain the All features of all of the graphic markers in the original image.
The device of claim 34, further comprising:

a size detecting module, configured to detect a size of the original picture, including a height and a width of the original picture;

The size obtaining module acquires a height and a width of the target picture;

The scaling module calculates a ratio of a height of the original picture to a height of the target picture, and calculates a ratio of a width of the original picture to a width of the target picture, and uses a maximum value of the obtained ratio as The scale of the original picture is scaled, and the original picture is scaled according to the ratio.
The device of claim 34, further comprising:

a color filling module, when the image of the original picture cannot fill the cropping area, according to the background color of the graphic mark retained in the target picture, and the extracted color is not described in the cropping area The portion of the original image that is filled with the image is filled.
The device according to claim 36, wherein

The color filling module selects an area where the background color is to be extracted from the original picture according to an edge position of the original picture and performs color extraction.
A device according to any one of claims 34 to 37, wherein

The cropping module centers the image of the original image in the cropping area when the image of the original picture cannot fill the cropping area.
A graphic mark picture detecting method, comprising:

Selecting an area of the color to be extracted from the picture based on an edge position of the picture to be detected;

Extracting colors in the area;

It is determined whether the graphic flag is included in the picture according to the extracted color in the area.
The method of claim 39, wherein extracting colors in the area comprises:

A luminance value for each pixel in the region under one or more color channels is extracted.
A method according to any of claims 39-40, wherein the one or more color channels comprise at least one of the following:

Red channel, green channel, blue channel.
The method according to any one of claims 39 to 41, wherein determining whether the picture is included in the picture according to the color in the extracted area comprises:

Counting the number of pixels with the same brightness value under each color channel;

Calculating the variance of the number of pixels under each color channel according to the number of pixels of each brightness value under each color channel;

Determining whether the color of the area is the background color of the graphic mark according to the variance of the number of pixels of the one or more channels, and determining that the result is YES means that the picture includes the graphic mark.
The method according to any one of claims 39 to 42, wherein the variance of the pixels under each color channel is calculated according to the number of pixels of each brightness value under each color channel, and specifically includes:

Calculating the proportion of the number of pixels of each brightness value under each color channel according to the number of pixels of the same brightness value under each color channel;

Calculating the average proportion of the number of pixels of all luminance values under each color channel according to the proportion of the number of pixels of each brightness value under each color channel;

The variance of the number of pixels under each color channel is calculated based on the ratio of the number of pixels of each luminance value under each color channel and the average proportion of the number of pixels of all luminance values under each color channel.
A method according to any one of claims 39-43, wherein

The area is hollow and the outer edge of the area coincides with the edge of the picture.
The method according to any one of claims 39 to 44, wherein selecting an area of the color to be extracted from the picture based on an edge position of the picture to be detected includes:

Calculating a distance between an outer edge and an inner edge of the area according to a size of the picture;

An area is selected from the picture by the distance between the outer edge and the inner edge of the area.
A graphic mark picture detecting device comprising:

An area selection module, configured to select an area of the color to be extracted from the picture based on an edge position of the picture to be detected;

a color extraction module for extracting colors in the area;

The graphic mark determining module is configured to determine, according to the extracted color in the area, whether the graphic includes a graphic mark.
The device according to claim 46, wherein

The color extraction module extracts luminance values for each pixel in the region under one or more color channels.
Apparatus according to any of claims 46-47, wherein the one or more color channels comprise at least one of the following:

Red channel, green channel, blue channel.
The apparatus of any of claims 46-48, further comprising:

a pixel statistics module, configured to count the number of pixels having the same brightness value under each color channel;

a variance calculation module, configured to calculate a variance of the number of pixels under each color channel according to the number of pixels of each brightness value under each color channel;

The graphic mark determining module determines whether the color of the area is the background color of the graphic mark according to the variance of the number of pixels of the one or more channels, and if the determination result is yes, whether the picture contains the graphic Sign.
The apparatus according to claims 46-49, further comprising:

a ratio calculation module, configured to calculate a proportion of the number of pixels of each brightness value under each color channel according to the number of pixels of the same brightness value under each color channel;

An average ratio calculation module for calculating an average proportion of the number of pixels of all luminance values under each color channel according to the proportion of the number of pixels of each luminance value in each color channel;

The variance calculation module calculates the number of pixels under each color channel according to the ratio of the number of pixels of each brightness value under each color channel and the average proportion of the number of pixels of all brightness values under each color channel. Variance.
A device according to any one of claims 46 to 50, wherein

The area is hollow and the outer edge of the area coincides with the edge of the picture.
The apparatus according to claims 46-51, further comprising:

a distance calculation module, configured to calculate a distance between an outer edge and an inner edge of the area according to a size of the picture;

The area selection module selects an area from the picture according to a distance between an outer edge and an inner edge of the area.
[Corrected in accordance with Rule 91 15.12.2015]
A computer program comprising computer readable code, when the computer readable code is run on a computing device, causing the computing device to perform the method of cropping a picture according to any one of claims 1-7, Or causing the computing device to perform the character picture cropping method according to any one of claims 15-20, or causing the computing device to perform the pair of containing graphics according to any one of claims 27-32 A method of cropping a picture of the logo or causing the computing device to perform the graphical logo picture detection method according to any one of claims 39-45.
[Corrected in accordance with Rule 91 15.12.2015]
A computer readable medium storing the computer program of claim 53.