CN111986277A - Image local comparison method, device, equipment and medium based on gray value - Google Patents

Abstract

The present application provides a method, device, device and medium for local image comparison based on gray value. By performing Gaussian blur processing and gray value processing on the obtained original image; The scale is reduced to a preset range, and according to the reduced gray value of each pixel point, it is divided into multiple gray value images of the same order corresponding to different integer gray values; binarize each gray value image of the same order process, and combine the pixels with the same gray value to form one or more connected domains; build a minimum rectangular frame for each connected domain; binarize the pixels in each minimum rectangular frame using the P-Hash algorithm to obtain a binary integer string for judging whether there are at least partially similar images in the image database by comparing the binary integer strings. The present application is convenient for storage and comparison, greatly saves the comparison processing speed and the memory required for storage, and further improves the accuracy and efficiency of local image comparison.

Description

Method, Apparatus, Equipment and Medium for Local Image Comparison Based on Gray Value

technical field

The present invention relates to the technical field of image processing, in particular to a method, device, device and medium for local comparison of images based on gray values.

Background technique

At present, more and more scenes require the comparison of pictures, such as the common identification of embezzled or plagiarized patterns. As the plagiarism of pictures or patterns becomes more and more concealed, it is often necessary to locally compare the colors and shapes of the patterns in the pictures.

In the traditional image comparison process, it is necessary to compare the images to be compared with the images in the database one by one. There are also many, which results in a lot of comparison speed, and the comparison speed is very slow. In addition, the data storage of pictures in the picture database is also very large, and local or subtle plagiarism patterns are often not found.

SUMMARY OF THE INVENTION

In view of the above-mentioned shortcomings of the prior art, the purpose of the present application is to provide a method, apparatus, device and medium for local comparison of images based on gray values, so as to solve at least one problem existing in the prior art.

In order to achieve the above purpose and other related purposes, the present application provides a gray value-based local image comparison method, the method includes: performing Gaussian blurring and gray value processing on the acquired original image; The gray value of the point is proportionally reduced to the preset range, and according to the reduced gray value of each pixel point, it is divided into multiple gray value images of the same level corresponding to different integer gray values; The value image is binarized, and the pixels with the same gray value are combined to form one or more connected domains; one or more connected domains containing independent complete patterns are selected according to the screening conditions, and for each connected domain Constructing the minimum rectangular frames used to cover all the independent complete patterns respectively; using the P-hash algorithm to binarize the pixels in each of the minimum rectangular frames to obtain a binary integer string for use in the image database It is judged whether there are at least partially similar images through the comparison of binary integer strings.

In an embodiment of the present application, the grayscale value of each pixel is proportionally reduced to a predetermined range, and is divided into a plurality of corresponding different integer grayscales according to the reduced grayscale value of each pixel. The same-level gray value image of the value, including: reducing the gray value of each pixel from the range 0-255×255 to the range 0-N in equal proportion; for the reduced gray value, the non-integer gray value The value is rounded to an integer gray value; corresponding pixels are selected according to the gray values belonging to the same integer in 0-N, so as to obtain the same-level gray value image of N+1 pixels only containing the same integer gray value .

In an embodiment of the present application, performing binarization processing on each gray value image of the same level, and combining pixel points with the same gray value to form one or more connected domains, includes: making each of the In the same-level gray-value image, the gray value of the selected same integer is 1, and the gray value of the gray value that does not belong to the selected same integer is 0, so that the gray value of each of the same-level gray value images is It is only 0 or 1; the pixel points with the gray value of 0 are respectively combined to form one or more connected domains; and/or the pixel points with the gray value of 1 are combined to form one or more connected domains.

In an embodiment of the present application, the screening conditions include any one or more of the following combinations: 1) the number of pixels in the connected domain is greater than 121; 2) the area of the connected domain accounts for at most the entire 3) The ratio of the longest length to the longest width in the connected domain is between 0.2 and 5.

In an embodiment of the present application, performing binarization processing on the pixels in each of the minimum rectangular frames by using the P-hash algorithm to obtain a binary integer string includes: dividing the minimum rectangular frame by n× The n grid is divided into n×n small blocks on average; calculate the sum of the gray values of all pixels in each small block; average the sum of the non-0 gray values in all the small blocks, and make each non-0 small block The sum of the gray value of , is compared with the average value; if it is less than the average value, the value corresponding to the non-0 small block is recorded as 0; if it is greater than or equal to the average value, the non-0 small block is recorded as 0. The corresponding value is denoted as 1; according to the method of gradually diffusing from the center position of the minimum rectangular frame to the outer ring, the values corresponding to the small blocks at the center position and the values corresponding to the small blocks on the outer ring of each layer from the inside to the outside are counted in turn. corresponding value to obtain a binary integer string corresponding to the same-level gray value image.

In an embodiment of the present application, the writing format of the binary integer string includes: respectively corresponding to the center position and sections of the binary integer string counted by the outer circle of each layer from the inside to the outside, corresponding to the In the outer order, the segments of the binary integer strings are sequentially arranged from the high order to the low order; the high order to the low order corresponds to right to left or left to right; The value corresponding to the block starts from the beginning, counts the values corresponding to each small block on the outer ring in a clockwise direction, and writes it in the order from low to high.

In an embodiment of the present application, determining whether there are locally similar images by comparing binary integer strings in the image database includes: setting a comparison line for the binary integer strings, so that the two The high-order segment of the binary integer string on the right or left side of the comparison line in the value integer string is compared with the value of the binary integer segment corresponding to each integer in the image database; wherein, the comparison line can be based on The comparison situation or scene needs are adjusted to adjust the comparison accuracy; if the comparison of the high-order segment of the binary integer string is the same, it is determined that the corresponding binary integer string is similar; if m corresponding grayscale values of the same level If the comparisons of the binary integer strings of the images are respectively similar, it is determined that there is an image in the image database that is partially similar to the original image.

In order to achieve the above purpose and other related purposes, the present application provides a gray value-based image local comparison device, the device includes: a first processing module for performing Gaussian blurring on the acquired original image, and grayscale Value processing; reduce the gray value of each pixel to the preset range in equal proportion, and divide it into multiple gray value images of the same level corresponding to different integer gray values according to the reduced gray value of each pixel ; The second processing module is used to binarize the gray value images of the same order, and combine the pixels with the same gray value to form one or more connected domains; Connected domains of independent complete patterns, and for each of the connected domains, respectively construct a minimum rectangular frame for covering all the independent complete patterns; binarize the pixels in each of the minimum rectangular frames using the P-Hash algorithm Processing to obtain a binary integer string for determining whether there are at least partially similar images in the image database by comparing the binary integer strings.

In order to achieve the above purpose and other related purposes, the present application provides a computer device, the device includes: a memory and a processor; the memory is used to store computer instructions; the processor executes the computer instructions to implement the above method. .

In order to achieve the above object and other related objects, the present application provides a computer-readable storage medium storing computer instructions, and when the computer instructions are executed, the above-mentioned method is executed.

To sum up, a method, device, device and medium for local image comparison based on gray value provided by the present application, by subjecting the acquired original image to Gaussian blurring and gray value processing; The gray value of the pixel is reduced to the preset range in equal proportion, and according to the reduced gray value of each pixel, it is divided into multiple gray value images of the same level corresponding to different integer gray values; The image is binarized, and the pixels with the same gray value are combined to form one or more connected domains; one or more connected domains containing independent complete patterns are selected according to the screening conditions, and for each connected domain, one or more connected domains are selected. Constructing a minimum rectangular frame for covering all the independent complete patterns; binarizing the pixels in each of the minimum rectangular frames using the P-Hash algorithm to obtain a binary integer string for passing in the image database The comparison of binary integer strings determines whether there are at least partially similar images.

Has the following beneficial effects:

The present application binarizes the image features based on the gray value, and converts the features of the local details of the image into binary value strings through the segmentation of the connected domain, which is convenient for storage and comparison, and greatly saves the comparison processing speed and the memory required for storage. , which improves the accuracy and efficiency of local image comparison.

Description of drawings

FIG. 1 is a schematic flowchart of a method for local image comparison based on gray value in an embodiment of the present application.

2A-2C are schematic diagrams of scenarios in which the pixel points in each of the smallest rectangular frames are binarized by the P-hash algorithm to obtain a binary integer string according to an embodiment of the present application.

FIG. 3 is a schematic block diagram of a local image comparison apparatus based on gray value in an embodiment of the present application.

FIG. 4 is a schematic structural diagram of a computer device in an embodiment of the present application.

Detailed ways

The embodiments of the present application are described below through specific specific examples, and those skilled in the art can easily understand other advantages and effects of the present application from the contents disclosed in this specification. The present application can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present application. It should be noted that the following embodiments and features in the embodiments may be combined with each other under the condition of no conflict.

It should be noted that the drawings provided in the following embodiments are only used to illustrate the basic concept of the present application in a schematic way, although the drawings only show the components related to the present application rather than the number, shape and number of components in actual implementation. The dimensions are drawn, but the type, quantity and proportion of each component can be arbitrarily changed in actual implementation, and the component layout may also be more complicated.

Throughout the specification, when a part is said to be "connected" to another part, this includes not only the case of "direct connection" but also the case of "indirect connection" with other elements interposed therebetween. In addition, when it says that a certain part "includes" a certain constituent element, unless there is particularly no description to the contrary, it does not exclude other constituent elements, but means that other constituent elements may also be included.

The terms first, second, and third mentioned herein are used for the purpose of describing various parts, components, regions, layers and/or sections and are not intended to be limiting. These terms are only used to distinguish one part, component, region, layer or section from another part, component, region, layer or section. Thus, recitation of a first part, component, region, layer or section below may refer to a second part, component, region, layer or section without departing from the scope of the present application.

Also, as used herein, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context dictates otherwise. It should be further understood that the terms "comprising", "comprising" indicate the presence of a stated feature, operation, element, component, item, kind, and/or group, but do not exclude one or more other features, operations, elements, components, The existence, appearance or addition of items, categories, and/or groups. The terms "or" and "and/or" as used herein are to be construed to be inclusive or to mean any one or any combination. Thus, "A, B or C" or "A, B and/or C" means "any of the following: A; B; C; A and B; A and C; B and C; A, B and C" . Exceptions to this definition arise only when combinations of elements, functions, or operations are inherently mutually exclusive in some way.

In the traditional image comparison process, it is necessary to compare the images to be compared with the images in the database one by one. For example, the common feature vector extracted from the image is compared, but the feature vector or pixel memory contained in a There are a lot of them, which results in a lot of comparison speeds. The comparison speed is very slow, and local or subtle plagiarism patterns are often not found. In addition, the data storage of pictures in the picture database is also very large.

In order to solve the above-mentioned problems, the present application proposes a method, device, equipment and medium for local image comparison based on gray value, which converts the feature value of the picture into an integer form, which is convenient for storage and comparison, and greatly saves the comparison speed and speed. memory required for storage.

As shown in FIG. 1 , a schematic flowchart of a method for local image comparison based on gray value in an embodiment of the present application is shown. As shown, the method includes:

Step S101: Perform Gaussian blur processing and gray value processing on the acquired original image.

In this embodiment, Gaussian blurring is performed on the acquired original image mainly for denoising.

The Gaussian Blur, also called Gaussian smoothing, is a processing effect widely used in image processing software such as Adobe Photoshop, GIMP, and Paint.NET, and is usually used to reduce image noise and reduce the level of detail. Gaussian smoothing is also used in the preprocessing stage in computer vision algorithms to enhance images at different scales (see Scale Space Representation and Scale Space Implementation). From a mathematical point of view, the Gaussian blurring process of an image is the convolution of the image with the normal distribution. Since the normal distribution is also known as the Gaussian distribution, this technique is called Gaussian blur.

Preferably, this step further includes: pruning four RGBA channels of the original image into three RGB channels.

Usually, an original image has four RGBA channels, which represent the color space of Red (red) Green (green) Blue (blue) and Alpha respectively, but it is actually just additional information of the RGB model. The color used is usually RGB, and the Alpha channel is generally used as an opacity parameter. Therefore, this application removes the Alpha channel that has little or no impact in the image comparison process to reduce the number of channels, thereby reducing the comparison. data used.

In this embodiment, the gray value processing in this step is mainly to convert the RGB three-channel pixels of each pixel into gray values, such as converting the pixel coordinates (x, y, rgb) of a certain pixel into gray. Degree coordinates (x,y,g).

Step S102: Reduce the gray value of each pixel to a preset range in equal proportions, and divide it into a plurality of gray value images of the same level corresponding to different integer gray values according to the reduced gray value of each pixel .

In an embodiment of the present application,

A. Reduce the gray value of each pixel from the range 0-255×255 to the range 0-N in equal proportions.

For example, after converting the three color channels of RGB to grayscale values, the grayscale value range is (0-255×255). Assuming that N is 15, a coefficient G is set here, G=255×255÷15=4335. Therefore, by dividing the gray value of each pixel in the original image by the coefficient G, the gray value of each pixel can be scaled down to a range of 0-15.

B. For the reduced grayscale value, round the non-integer grayscale value to an integer grayscale value.

For example, after dividing the gray value of each pixel by the coefficient G, the obtained gray value is either a non-integer with a decimal or an integer. Therefore, the present application converts the non-integer into a uniform by rounding. is an integer.

C. Select corresponding pixel points according to the gray value of the same integer in 0-N, so as to obtain N+1 gray value images of the same level that only contain the same integer gray value of the pixel points.

For example, by selecting the pixels whose gray value is adjusted to 0, and setting the corresponding other unselected pixels to be empty, an image of the same gray value whose gray value only contains 0 can be obtained. By analogy, assuming that N is 15, 16 such grayscale images of the same order can be obtained. But it should be noted that N here is not limited to 15, and N can also take a number below 10 and below 100. For example, when the pixel size of the original image is not large, it can take 1 digit, and when the pixel size of the original image is very large, it can also take any two digits below 100.

For another example, assuming that the original image is a face image, the situation presented by dividing it into multiple grayscale images of the same level corresponding to different integer grayscale values may be: pixel points in a grayscale image of the same level It is presented as the outline of the face, the pixels in another grayscale image of the same level are presented as eyes, and the pixels in another grayscale image of the same level are presented as hair, etc. Each gray value image of the same level is formed into sub-images of different angles, so as to greatly refine the local characteristics of the image, and even in the case of misappropriation or plagiarism of small image elements, a good comparison effect can be obtained.

Step S103 : Binarize each gray value image of the same level, and combine pixel points with the same gray value to form one or more connected domains.

This step specifically includes:

A. Let the grayscale value of each of the same-level grayscale value images belonging to the selected same integer be 1, and let the grayscale value of the grayscale value not belonging to the selected same integer be 0, so that each of the same-level grayscale values The grayscale value of the value image is only 0 or 1;

B. Respectively combine pixel points with a grayscale value of 0 and pixels with a grayscale value of 1 to form one or more connected domains.

Taking the same-level gray value image from the above example, assuming that it only contains the same integer gray value of N, the following formula can be used to perform binarization processing, so that the value of all pixels in the image is 1 or 0 .

Among them, g' represents the gray value after proportional reduction.

After the gray value of the gray value image of the same level becomes only 0 or 1, the pixels with the gray value of 0 can be connected, and the pixels with the gray value of 1 can be connected, so as to obtain an or Multiple connected domains that contain independent complete patterns.

It should be noted that in this step, the connected domain formed by the same gray value actually displays and highlights any descriptive and meaningful small pattern or complete pattern. Therefore, each connected domain theory has certain characteristics. Compare value.

Generally speaking, a gray value image of the same level can obtain dozens or even hundreds of connected domains containing independent complete patterns, and each connected domain can be represented as a valuable subgraph.

Step S104 : Select one or more connected domains including independent complete patterns according to the screening conditions, and construct a minimum rectangular frame for covering all the independent complete patterns for each connected domain.

It should be noted that the patterns of one or more connected domains obtained from the same-level gray value image may be large or small, tall, short, fat and thin, or tend to be plump, and have various shapes. Therefore, not any connected domain satisfies the requirements of the subsequent calculation of the method described in this application. Therefore, the present application selects suitable connected domains containing independent complete patterns through preset screening conditions.

In an embodiment of the present application, the screening conditions include any one or more of the following combinations:

1) The number of pixels in the connected domain is greater than 121. Preferably, the length and width are preferably greater than 11.

It should be noted that the purpose of requiring the number of pixels to be greater than 121 is that in step S105, the same-level gray value image needs to be divided by an n×n grid, and in this application, 11×11 is It is more preferable to divide the grid, so it is hoped that the number of pixels in the connected domain is not less than 121. In addition, the size of the original image is not scaled in this application, so the size of the original image is theoretically relatively large, and the number of pixels in the connected domain is limited to 121 here, which actually limits the selected independent pixels. Minimum size of a complete pattern. Because, in practice, the number of pixels that are too small contains fewer valuable image elements, and it also increases the number of comparisons and prolongs the processing time.

2) The area of the connected domain accounts for at most one third of the entire gray value image of the same level.

The purpose of this filter condition is to try to make the image more detailed in order to improve the accuracy of finding similar elements in complex images. Moreover, an area that is too large is likely to be the background, which is not suitable for finding local similar patterns in this application. Preferably, the P hash (Phash) operation is performed after the entire image is reduced to 11x11 size, which makes it easier to find the overall similarity. However, if the comparison area is too large, it will interfere with finding different pictures.

3) The ratio of the longest length to the longest width in the connected domain is between 0.2 and 5.

It should be noted that the ratio of the longest length to the longest width described here is the aspect ratio of the corresponding connected domain, because a rectangular frame covering the connected domain needs to be constructed later, so the more the shape of the connected domain is. The better the saturation, i.e. round or square. Therefore, in the present application, it is expected that the aspect ratio of the connected domain is limited between 0.2 and 5, that is, the length is not greater than 5 times the width, or less than 1/5 of the width.

Step S105: Perform binarization processing on the pixels in each of the minimum rectangular frames using the P-Hash algorithm to obtain a binary integer string, which is used to judge whether there is an at least partial integer string in the image database by comparing the binary integer string. similar images.

In an embodiment of the present application, performing binarization processing on the pixels in each of the smallest rectangular frames by using the P hash algorithm to obtain a binary integer string includes:

A. Divide the smallest rectangular frame into n×n small blocks on average according to n×n grids.

Preferably, the n is preferably singular, such as 1, 3, 5, 7, 9, 11, 13 and the like. The reason is that, n is an n×n grid formed by a singular number, so that the divided small blocks can form a layer-by-layer outer ring extending outward from the center. For example, a 5×5 grid is divided into 25 blocks, which can form a center block, an inner circle near the center, and an outermost outer circle.

It should be noted that, in actual operation, the minimum rectangular frame is preferably not less than 11×11 in an n×n grid. In the present application, the grid less than 11×11 is used for illustration only, and it also corresponds to the connected domain filtering condition, where the number of pixels in the connected domain is greater than 121. Preferably, the length and width are preferably greater than 11.

B. Calculate the sum of the gray values of all pixels in each small block.

Since the size of the original image is not limited, the size of the connected domain obtained from the corresponding gray value images of the same level is not small. Therefore, although it is divided into small blocks, each small block may contain multiple Each pixel in the small block may be 1 or 0. Therefore, to count the sum of the gray values of all pixels in each small block is to add the values of 0 or 1 corresponding to each pixel. .

As shown in Figure 2A, it is assumed that the minimum rectangular frame is divided into 25 blocks according to 5 × 5 grids. The sum of the gray values of the small blocks in a connected domain in the figure is 2, and there are 2 × 2 pixels in this small block. Points are 0, 1, 0, 1, respectively, so the sum of the grayscale values of this small block is 2.

It should be noted that although the edge of the connected domain shown in Figure 2A only accounts for half of the small block whose gray value sum is 2, it should be noted that when the connected domain is constructed, two gray value images of the same level are used for After the value processing, the pixels whose gray value is 1 are connected. Therefore, if the connected domain is enlarged to display the size of each pixel, the edges of the connected domain are all composed of pixel units, so the actual The above-mentioned connected domain will not only occupy half of the pixels, but the entire pixel will be used as a component of the edge, and Figure 2A only displays the size to the size of the small block, so it appears to occupy half of the small block. .

C. Average the sum of the non-0 grayscale values in all the small blocks, and compare the sum of the grayscale values of each non-0 small block with the average value.

Count the sum of the gray values of all the pixels in the small block. If the gray value of all the pixels in the small block is 0, then the sum of the gray values is also 0. For example, the edge of the rectangle is far from the connected domain. In the area of it often occurs that the sum of the gray values is 0. Then calculate the average value of the sum of the gray value of these non-zero small blocks, and then compare the value of the sum of the gray value that is not 0 with the obtained average value.

D. If it is less than the average value, the value corresponding to the non-zero block is recorded as 0; if it is greater than or equal to the average value, the value corresponding to the non-0 small block is recorded as 1;

For example, as shown in Figure 2B, the figure is a 3×3 rectangular block, the sum of the gray value of each small block is shown in the figure, wherein, there are 7 small blocks whose gray value sum is non-0, The total sum of the gray value of these 7 small blocks is 14, and the average is 2. Then compare the sum of the grayscale values of these 7 small blocks with 2, the value corresponding to the small block smaller than 2 is recorded as 0, and the value corresponding to the small block greater than or equal to 2 is recorded as 1.

E. According to the method of gradually diffusing from the center position of the minimum rectangular frame to the outer circle, count the values corresponding to the small blocks at the center position and the values corresponding to the small blocks on the outer circles of each layer from the inside to the outside in turn, so that Obtain the binary integer string corresponding to the same-level gray value image.

It should be noted that, corresponding to this embodiment, steps A-E here are similar to the P hash algorithm in the art.

In an embodiment of the present application, the writing format of the binary integer string includes:

1) Corresponding to the central position and the sections of the binary integer string counted by the outer circles of each layer from the inside to the outside, corresponding to the order from the inside to the outside, arranging the sections of each binary integer string from the high order to the low order. ; The high-to-low corresponds to right-to-left or left-to-right;

and / or,

2) The value corresponding to the small block in the upper left corner of the outer ring of each layer starts from the value corresponding to the small block in the upper left corner, counts the values corresponding to each small block on the outer ring in a clockwise direction, and writes in the order from low to high.

In this embodiment, the value closer to the center point has a higher corresponding weight and is placed at a high position. The high-to-low may correspond to right-to-left or left-to-right.

For example, as shown in FIG. 2C , the binary integer counted by the center position is 1, and the corresponding segment is set as A; the binary integer counted by the outer ring adjacent to the center position is 00100101, Let its corresponding section be B; the binary integer counted by the outermost outer ring is 01110011011101110, and let its corresponding section be C. Assuming that the high-order to low-order can correspond to from right to left, then the obtained binary integer string corresponding to the same-level gray value image after the processing of steps A-E shown in FIG. 2C is:

01110011011101110/00100101/1;

Among them, "/" is only used to distinguish different sections, and does not belong to the value or character in the binary integer string.

In an embodiment of the present application, the judging whether there are locally similar images in the image database through the comparison of binary integer strings includes:

A. Set an alignment line for the binary integer string, so that the high-order segment of the binary integer string on the right or left side of the alignment line in the binary integer string corresponds to the binary integer corresponding to each integer in the image database. The value of the integer value segment is compared; wherein, the comparison line can be adjusted according to the comparison situation or scene requirements, so as to adjust the comparison accuracy;

It should be noted that, in the writing format of the binary integer string as exemplified above, it is not difficult to see that the high-order value corresponds to the center position of the gray value image of the same level, and the lower the value of the binary integer string is, the more it corresponds to the same level. The edge position of the grayscale image. Therefore, the comparison of the values corresponding to the center position is more meaningful and the accuracy is correspondingly higher.

B. If the comparisons of the high-order segments of the binary integer strings are the same, it is determined that the corresponding binary integer strings are similar.

C. If the comparisons of m binary integer strings corresponding to the same-level gray value images are respectively similar, it is determined that there is an image that is partially similar to the original image in the image database.

It should be noted that the comparison of the high-order segments of the binary integer strings here can be the same or different, but the same binary integer segment, the entire binary integer string corresponding to it is difficult to be the same, so it can only be determined that they are similar. . However, since the valuables in the pattern are more concentrated in the center position, it can be judged that the two compared binary integer strings are similar if the alignment of the high-order segment of the binary integer string is the same. Based on this, if the comparisons of the high-order segments of the m binary integer strings corresponding to the same-level gray value images are the same, it can be determined that there is an image in the image database that is partially similar to the original image. In some achievable embodiments, if the m is 3 or 4, it can be accurately determined that there is an image that is partially similar to the original image in the image database.

Generally speaking, the purpose of this application is to cut the same-level gray value image according to the connected domain, cut a picture into hundreds of valuable subgraphs, convert each subgraph into a Hash integer, and convert the subgraph search into Integer range comparison for the database. Corresponding to a gray value image of the same level, dozens and hundreds of binary integer strings can be formed, and for an original image, there are hundreds of binary integer strings. Therefore, in the local comparison of the image, refinement more precise.

To sum up, a method, device, device and medium for local image comparison based on gray value proposed in this application binarize image features based on gray value, and divide the image local details through the segmentation of connected domains. The features are converted into binary value strings, which are convenient for storage and comparison, which greatly saves the comparison processing speed and memory required for storage, and also improves the accuracy and efficiency of local image comparison.

As shown in FIG. 3 , it is a schematic block diagram of an apparatus for local image comparison based on gray value in an embodiment of the present application. As shown, the apparatus 300 includes:

The first processing module 301 is used to perform Gaussian blur processing and gray value processing on the obtained original image; the gray value of each pixel is proportionally reduced to within a preset range, and according to the reduced size of each pixel. Gray value, divided into multiple gray value images of the same order corresponding to different integer gray values;

The second processing module 302 is configured to perform binarization processing on each gray value image of the same level, and combine the pixels with the same gray value to form one or more connected domains; according to the screening conditions, select one or more including Connected domains of independent complete patterns, and for each of the connected domains, respectively construct a minimum rectangular frame for covering all the independent complete patterns; binarize the pixels in each of the minimum rectangular frames using the P-Hash algorithm Processing to obtain a binary integer string for determining whether there are at least partially similar images in the image database by comparing the binary integer strings.

It should be noted that the information exchange, execution process, etc. among the modules/units of the above-mentioned device are based on the same concept as the method embodiments described in the present application, and the technical effects brought by them are the same as those of the method embodiments of the present application. For the content, reference may be made to the descriptions in the method embodiments shown in the foregoing application, and details are not repeated here.

It should also be noted that it should be understood that the division of each module of the above apparatus is only a division of logical functions, and may be fully or partially integrated into a physical entity during actual implementation, or may be physically separated. And these units can all be implemented in the form of software calling through processing elements; they can also all be implemented in hardware; some modules can also be implemented in the form of calling software through processing elements, and some modules can be implemented in hardware. For example, the second processing module 302 may be a separately established processing element, or may be integrated into a certain chip of the above-mentioned system to be implemented, in addition, it may also be stored in the memory of the above-mentioned system in the form of program code, and the above-mentioned device is executed by a certain chip. A processing element invokes and executes the functions of the second processing module 302 above. The implementation of other modules is similar. In addition, all or part of these modules can be integrated together, and can also be implemented independently. The processing element described here may be an integrated circuit with signal processing capability. In the implementation process, each step of the above-mentioned method or each of the above-mentioned modules can be completed by an integrated logic circuit of hardware in the processor element or an instruction in the form of software.

For example, the above modules may be one or more integrated circuits configured to implement the above methods, such as: one or more specific integrated circuits (Application Specific Integrated Circuit, ASIC for short), or one or more microprocessors ( digital signal processor, referred to as DSP), or, one or more Field Programmable Gate Array (Field Programmable Gate Array, referred to as FPGA) and the like. For another example, when one of the above modules is implemented in the form of processing element scheduling program code, the processing element may be a general-purpose processor, such as a central processing unit (Central Processing Unit, CPU for short) or other processors that can call program codes. For another example, these modules can be integrated together and implemented in the form of a system-on-a-chip (SOC for short).

As shown in FIG. 4 , a schematic structural diagram of a computer device in an embodiment of the present application is shown. As shown in the figure, the computer device 400 includes: a memory 401 and a processor 402; the memory 401 is used to store computer instructions; the processor 402 executes the computer instructions to implement the method described in FIG. 1 .

In some embodiments, the number of the memory 401 in the computer device 400 may be one or more, the number of the processor 402 may be one or more, and in FIG. example.

In an embodiment of the present application, the processor 402 in the computer device 400 loads one or more instructions corresponding to the process of the application program into the memory 401 according to the steps shown in FIG. The controller 402 is used to run the application program stored in the memory 401, thereby implementing the method described in FIG. 1 .

The memory 401 may include random access memory (Random Access Memory, RAM for short), and may also include non-volatile memory (non-volatile memory), such as at least one disk memory. The memory 401 stores operating systems and operation instructions, executable modules or data structures, or their subsets, or their extended sets, wherein the operation instructions may include various operation instructions for implementing various operations. The operating system may include various system programs for implementing various basic services and handling hardware-based tasks.

The processor 402 may be a general-purpose processor, including a central processing unit (Central Processing Unit, CPU for short), a network processor (Network Processor, NP for short), etc.; it may also be a digital signal processor (Digital Signal Processing, DSP for short) ), Application Specific Integrated Circuit (ASIC for short), Field-Programmable Gate Array (FPGA for short) or other programmable logic devices, discrete gate or transistor logic devices, and discrete hardware components.

In some specific applications, various components of the computer device 400 are coupled together through a bus system, where the bus system may include a power bus, a control bus, a status signal bus, and the like in addition to a data bus. However, for the sake of clarity of illustration, the various buses are referred to as bus systems in FIG. 4 .

In an embodiment of the present application, the present application provides a computer-readable storage medium on which a computer program is stored, and when the program is executed by a processor, implements the method described in FIG. 1 .

At any possible level of incorporation of technical details, the present application may be a system, method and/or computer program product. The computer program product may include a computer-readable storage medium having computer-readable program instructions loaded thereon for causing a processor to implement various aspects of the present application.

A computer-readable storage medium may be a tangible device that can hold and store instructions for use by the instruction execution device. The computer-readable storage medium may be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. More specific examples (non-exhaustive list) of computer readable storage media include: portable computer disks, hard disks, random access memory (RAM), read only memory (ROM), erasable programmable read only memory (EPROM) or flash memory), static random access memory (SRAM), portable compact disk read only memory (CD-ROM), digital versatile disk (DVD), memory sticks, floppy disks, mechanically coded devices, such as printers with instructions stored thereon Hole cards or raised structures in grooves, and any suitable combination of the above. Computer-readable storage media, as used herein, are not to be construed as transient signals per se, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (eg, light pulses through fiber optic cables), or through electrical wires transmitted electrical signals.

The computer-readable programs described herein can be downloaded to various computing/processing devices from computer-readable storage media, or to external computers or external storage devices over a network such as the Internet, a local area network, a wide area network, and/or a wireless network. The network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and/or edge servers. A network adapter card or network interface in each computing/processing device receives computer-readable program instructions from a network and forwards the computer-readable program instructions for storage in a computer-readable storage medium in each computing/processing device .

Computer program instructions for carrying out the operations of the present application may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, integrated circuit configuration data, or in one or more source or object code written in any combination of programming languages, including object-oriented programming languages, such as Smalltalk, C++, etc., and procedural programming languages, such as the "C" language or similar programming languages. The computer readable program instructions may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer, or entirely on the remote computer or server implement. In the case of a remote computer, the remote computer may be connected to the user's computer through any kind of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (eg, using an Internet service provider through the Internet connect). In some embodiments, custom electronic circuits, such as programmable logic circuits, field programmable gate arrays (FPGAs), or programmable logic arrays (PLAs), can be personalized by utilizing state information of computer readable program instructions. Computer readable program instructions are executed to implement various aspects of the present application.

To sum up, a method, device, device and medium for local image comparison based on gray value provided by the present application, by subjecting the acquired original image to Gaussian blurring and gray value processing; The gray value of the pixel is reduced to the preset range in equal proportion, and according to the reduced gray value of each pixel, it is divided into multiple gray value images of the same level corresponding to different integer gray values; The image is binarized, and the pixels with the same gray value are combined to form one or more connected domains; one or more connected domains containing independent complete patterns are selected according to the screening conditions, and for each connected domain, one or more connected domains are selected. Constructing a minimum rectangular frame for covering all the independent complete patterns; binarizing the pixels in each of the minimum rectangular frames using the P-Hash algorithm to obtain a binary integer string for passing in the image database The comparison of binary integer strings determines whether there are at least partially similar images.

The present application effectively overcomes various shortcomings in the prior art and has high industrial application value.

The above-mentioned embodiments merely illustrate the principles and effects of the present application, but are not intended to limit the present invention. Anyone skilled in the art can make modifications or changes to the above embodiments without departing from the spirit and scope of the present application. Therefore, all equivalent modifications or changes made by those skilled in the art without departing from the spirit and technical idea disclosed in the present invention should still be covered by the claims of the present application.

Claims (10)

1. an image local comparison method based on gray value, is characterized in that, described method comprises: Perform Gaussian blur processing and gray value processing on the acquired original image; The gray value of each pixel is reduced proportionally to a preset range, and according to the reduced gray value of each pixel, it is divided into a plurality of gray value images of the same level corresponding to different integer gray values; Binarize each gray value image of the same level, and combine the pixels with the same gray value to form one or more connected domains; Selecting one or more connected domains containing independent complete patterns according to the screening conditions, and constructing a minimum rectangular frame for covering all the independent complete patterns for each of the connected domains; Binarization is performed on the pixels in each of the minimum rectangular frames by using the P-Hash algorithm to obtain a binary integer string, which is used to judge whether there are at least partially similar images in the image database through the comparison of the binary integer strings . 2. The method according to claim 1, wherein the gray value of each pixel is proportionally reduced to within a preset range, and is divided into multiple groups according to the reduced gray value of each pixel. A gray value image of the same order corresponding to different integer gray values, including: Reduce the gray value of each pixel from the range 0-255×255 to the range 0-N in equal proportions; For the reduced grayscale value, round the non-integer grayscale value to an integer grayscale value; Corresponding pixel points are respectively selected according to the gray value of the same integer in 0-N, so as to obtain the same-level gray value image of N+1 pixel points only containing the same integer gray value. 3. The method according to claim 2, wherein the binarization processing is performed on each gray value image of the same level, and the pixel points with the same gray value are combined to form one or more connected domains, include: Let the grayscale value of each of the same-level grayscale value images belonging to the selected same integer be 1, and let the grayscale value of not belonging to the selected same integer value be 0, so that each of the same-level grayscale value images The grayscale value is only 0 or 1; The pixel points with the gray value of 0 are respectively combined to form one or more connected domains; and/or the pixel points with the gray value of 1 are combined to form one or more connected domains. 4. The method according to claim 1, wherein the screening condition comprises any one or more of the following combinations: 1) The number of pixels in the connected domain is greater than 121; 2) the area of the connected domain accounts for at most one third of the entire gray value image of the same order; 3) The ratio of the longest length to the longest width in the connected domain is between 0.2 and 5. 5. The method according to claim 1, wherein the pixel points in each of the minimum rectangular frames are subjected to binarization processing using a P-Hash algorithm to obtain a binary integer string, comprising: Divide the minimum rectangular frame into n×n small blocks on average according to n×n grids; Calculate the sum of the gray values of all pixels in each small block; Average the sum of the non-0 grayscale values in all the small blocks, and compare the sum of the grayscale values of each non-0 small block with the average value; If it is less than the average value, the value corresponding to the non-0 small block is recorded as 0; if it is greater than or equal to the average value, the value corresponding to the non-0 small block is recorded as 1; According to the method of gradually diffusing from the center position of the smallest rectangular frame to the outer circle, count the values corresponding to the small blocks at the center position and the values corresponding to the small blocks on the outer circle of each layer from the inside to the outside in turn, so as to obtain the corresponding A string of binary integers for a grayscale image of the same order. 6. The method according to claim 5, wherein the writing format of the binary integer string comprises: Corresponding to the center position and the sections of the binary integer string counted by the outer circles of each layer from the inside to the outside, corresponding to the order from the inside to the outside, the sections of the binary integer strings are sequentially arranged from the high order to the low order; The above-mentioned high-to-low corresponds to right-to-left or left-to-right; and / or, In the outer ring of each layer, the value corresponding to the small block in the upper left corner starts from the value corresponding to the small block in the upper left corner. The values corresponding to each small block on the outer ring are counted clockwise, and are written in the order from low to high. 7. The method according to claim 1, wherein, determining whether there are locally similar images in the image database through the comparison of binary integer strings, comprising: An alignment line is set for the binary integer string, so that the high-order segment of the binary integer string on the right or left side of the alignment line in the binary integer string corresponds to the binary integer corresponding to each integer in the image database The values of the sections are compared; wherein, the comparison line can be adjusted according to the comparison situation or scene requirements, so as to adjust the comparison accuracy; If the comparisons of the high-order segments of the binary integer strings are the same, it is determined that the corresponding binary integer strings are similar; If the comparisons of the m binary integer strings corresponding to the same-level gray value images are respectively similar, it is determined that there is an image that is locally similar to the original image in the image database. 8. A device for local comparison of images based on gray values, wherein the device comprises: The first processing module is used to perform Gaussian blur processing and gray value processing on the acquired original image; the gray value of each pixel is proportionally reduced to a preset range, and the gray value of each pixel is reduced according to the reduced gray value. The degree value is divided into multiple gray value images of the same order corresponding to different integer gray values; The second processing module is used for binarizing each gray value image of the same level, and combining the pixels with the same gray value to form one or more connected domains; according to the screening conditions, select one or more including independent The connected domain of the complete pattern, and for each of the connected domains, a minimum rectangular frame for covering all the independent complete patterns is respectively constructed; the pixels in each of the minimum rectangular frames are binarized using the P-Hash algorithm. to obtain a binary integer string for judging whether there are at least partially similar images in the image database by comparing the binary integer strings. 9. A computer device, characterized in that, the device comprises: a memory and a processor; the memory is used to store computer instructions; the processor executes the computer instructions to implement any one of claims 1 to 7. method described. 10. A computer-readable storage medium, characterized by storing computer instructions that, when executed, perform the method of any one of claims 1 to 7.

Images