CN114359023A - Method, equipment and system for dispatching picture stream to center based on complexity - Google Patents
Method, equipment and system for dispatching picture stream to center based on complexity Download PDFInfo
- Publication number
- CN114359023A CN114359023A CN202210019824.8A CN202210019824A CN114359023A CN 114359023 A CN114359023 A CN 114359023A CN 202210019824 A CN202210019824 A CN 202210019824A CN 114359023 A CN114359023 A CN 114359023A
- Authority
- CN
- China
- Prior art keywords
- picture
- complexity
- matrix
- recognized
- identified
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Landscapes
- Image Analysis (AREA)
Abstract
The invention discloses a method, electronic equipment and a medium for dispatching picture shunt to a center based on complexity, which comprises the following steps: s1, receiving a picture to be identified sent by an X-ray machine; s2, analyzing the picture to be recognized to obtain the complexity of the picture to be recognized; s3, determining the picture to be recognized corresponding to the complexity of the picture to be recognized which is greater than the complexity threshold value as a complex picture based on the complexity of the picture to be recognized and a preset complexity threshold value; and S4, sending the picture to be identified which is judged to be the complex picture to the central server. According to the method, the sharpening matrix and the negation matrix are introduced through the Laplace transformation and the inverse transformation, so that the complexity of the picture to be identified of the X-ray security inspection machine is more objective and accurate, and the operation efficiency is improved.
Description
Technical Field
The invention relates to the field of security check intelligent image judgment, in particular to a method for dispatching image sub-streams to a center based on complexity, an edge image recognition box and a system.
Background
The intelligent security inspection image judging system based on the edge image identifying box is characterized in that the edge image identifying box is arranged in each station entering security inspection point, an X-ray machine is used for detecting security inspection packages, passengers can place the packages on a conveying crawler belt to receive X-ray inspection when entering the station, and the X-ray machine outputs X-ray imaging videos in the process; the edge recognition box is responsible for discerning contraband in the X-ray imaging video, acquire contraband information, the parcel information after the edge recognition box is handled is sent to equipment such as unpacking platform, local recognition picture, however at this in-process, if discernment object is sheltered from or partly shelters from, discernment object material is too complicated, article warp reasons such as twist, often produce a series of complicated parcel images, because the intelligent recognition picture box of deployment on the marginal website because the restriction on reasons such as hardware performance, article on these complicated pictures have the difficulty in discerning.
Disclosure of Invention
The invention aims to provide a method, an edge image recognition box and a system for dispatching image streams to a center based on complexity.
The method for dispatching the picture shunt to the center based on the complexity is applied to an edge map recognition box, and specifically comprises the following steps:
s1, receiving a picture to be identified sent by an X-ray machine;
s2, analyzing the picture to be recognized to obtain the complexity of the picture to be recognized;
s3, determining the picture to be recognized corresponding to the complexity of the picture to be recognized which is greater than the complexity threshold value as a complex picture based on the complexity of the picture to be recognized and a preset complexity threshold value;
and S4, sending the picture to be identified which is judged to be the complex picture to the central server.
Further, in the step S2, a complexity model is adopted to obtain the complexity of the picture to be recognized, where the complexity model includes a transformation process and a labeling process, the transformation process includes at least one of a laplace transform and an inverse transform, and the complexity model specifically includes the following steps:
s201, converting the picture to be identified to obtain a conversion matrix D;
s202, marking the picture to be identified to obtain a marking matrix C;
s203, obtaining the complexity K of the picture to be identified according to the transformation matrix D and the mark matrix C.
Further, in the step S2, a complexity model is used to obtain the complexity of the picture to be recognized, where the complexity model includes a transformation process and a labeling process, the transformation process includes a laplace transformation and an inverse transformation, and the complexity model specifically includes the following steps:
s200, converting the picture to be identified into a gray picture;
s201, performing inverse transformation on the gray-scale picture to obtain an inverse matrix M, and performing Laplace operator transformation on the gray-scale picture to obtain a sharpening matrix L;
s202, marking the picture to be identified to obtain a marking matrix C;
s203, multiplying the negation matrix M and the mark matrix C in a contraposition mode to obtain a matrix B, summing all elements of the matrix B to obtain SUM (B),
multiplying the sharpening matrix L and the marking matrix C in a contraposition mode to obtain a matrix A, summing all elements of the matrix A to obtain SUM (A),
and performing weighted summation on the sum (b) and sum (a) to obtain the complexity K of the picture to be recognized, where K is α × sum (b) + β × sum (a), where α and β are corresponding weighting coefficients.
Further, the marking process is as follows: and carrying out binarization processing on the acquired picture to be recognized to obtain a corresponding mark matrix C, wherein each element in the mark matrix C represents the marking degree of a corresponding pixel point in the picture to be recognized.
Further, each element in the marking matrix C is obtained according to the following manner:
for each element in the type matrix C, judging whether a corresponding pixel point of the element in the picture to be identified is in a marked article area, if not, the pixel point is 0, and if so, the pixel point is 1; and determining the marking degree of the element based on the 0 or 1.
Further, in the step S2, a consistency model is used to obtain the complexity of the picture to be recognized, where the consistency model specifically includes the following steps:
SA, converting the picture to be identified into a gray picture, and obtaining a gray matrix M' of the gray picture;
SB, substituting all elements in the gray matrix M' into a formula
Obtaining the complexity U of the picture to be identified, wherein m and n are the number of rows and columns of the picture to be identified respectively, f (i, j) is the gray value of the pixel (i, j) of the gray picture,
is a 3 × 3 neighborhood of pixels centered on pixel (i, j)The gray level average of (1).
Further, in step S2, an entropy model is used to obtain the complexity of the picture to be recognized, where the entropy model specifically includes the following steps:
sa, converting the picture to be identified into a gray picture, and obtaining a gray matrix M' of the gray picture;
sb, obtaining a gray level co-occurrence matrix D according to the gray level matrix M' of the gray level picture;
sc, substituting all elements in the gray level co-occurrence matrix D into a formula
And obtaining the complexity S of the picture to be recognized, wherein P (i, j) is the ith row and the jth column in the gray level co-occurrence matrix D, and N is the gray level of the gray level picture.
Further, the method for obtaining the complexity of the picture to be recognized by using the complexity model, the consistency model and the entropy model in the step S2 specifically includes the following steps:
s001, inputting the picture to be recognized into a complexity model to obtain the complexity K of the picture to be recognized;
s002, inputting the picture to be recognized into a consistency model to obtain the complexity U of the picture to be recognized;
s003, inputting the picture to be recognized into an entropy model to obtain the complexity S of the picture to be recognized;
and S004, carrying out weighted summation on the complexity K, U, S of the picture to be recognized to obtain the complexity Q of the picture to be recognized, which is a multiplied by K + b multiplied by U + c multiplied by S, wherein a, b and c are corresponding weighting coefficients.
An apparatus for complexity-based scheduling of picture streamlets to a hub, comprising:
the receiving module is used for receiving the picture to be identified sent by the X-ray machine;
the complexity module is used for analyzing the picture to be identified to obtain the complexity of the picture to be identified;
the judging module is used for judging the picture to be identified corresponding to the complexity of the picture to be identified which is greater than the complexity threshold value as a complex picture based on the complexity of the picture to be identified and a preset complexity threshold value;
and the scheduling module is used for sending the picture to be identified which is judged to be the complex picture to the central server.
A system for complexity-based hierarchical scheduling of pictures to a hub, comprising:
the system comprises a central server, an edge computing node and a plurality of X-ray machines of security check points, wherein the edge computing node is composed of a plurality of edge image recognition boxes; the central server is connected with each edge attempt box and each edge recognition box is connected with the X-ray machine of the security check point where the edge recognition box is located;
the central server is used for receiving the pictures to be identified which are judged to be complex pictures and sent by each edge image identifying box and carrying out image identifying processing;
each edge map box is used for:
receiving a picture to be identified sent by an X-ray machine;
analyzing the picture to be identified to obtain the complexity of the picture to be identified;
determining the picture to be recognized corresponding to the complexity of the picture to be recognized which is greater than the complexity threshold value as a complex picture based on the complexity of the picture to be recognized and a preset complexity threshold value;
sending the picture to be identified which is judged to be the complex picture to a central server;
the X-ray machine of the security check point is used for scanning the passenger's packages, obtaining X-ray scanning pictures and sending the X-ray scanning pictures to the edge image identifying box of the security check point.
The preset complexity threshold value is an average value of the complexity of the picture to be identified according to the history.
According to the method, Laplace convolution operation and negation operation are introduced aiming at the situations that the pixel color of an X-ray imaging picture is complex and objects in the picture are overlapped, and complexity calculation of a background part is eliminated through marking processing, so that the complexity of the picture to be identified aiming at the X-ray security inspection machine is more objective and accurate.
The invention has the following beneficial effects:
the method comprises the steps of respectively obtaining a sharpening matrix L of a sharpened picture, an inverse matrix M of an inverse gray picture and a marking matrix C by carrying out Laplace convolution operation, inverse operation and marking processing on a gray picture of a picture to be identified, carrying out counterpoint multiplication on the marking matrix C of the picture to be identified and the sharpening matrix L of the sharpened picture and the inverse matrix M of the inverse gray picture to obtain A, B, calculating sum values SUM (A) and SUM (B) of all elements in a A, B matrix, carrying out weighted summation of the SUM (A) and SUM (B) to obtain the complexity of the picture to be identified, converting the color complexity into the intensity of color change in the picture by the Laplace convolution operation, adding the factor of a deeper area generated by overlapping objects in the picture into the complexity calculation by the inverse operation, carrying out marking processing, the method reduces the unmarked object area (background part) in the picture to be recognized, so that the complexity of obtaining the picture to be recognized is more objective and accurate, and the operation efficiency is improved.
Drawings
FIG. 1 is a schematic flow chart of a method for complexity-based scheduling of picture streaming to a hub according to the present invention;
FIG. 2 is a schematic diagram of an apparatus for complexity-based scheduling of picture segments to a hub according to the present invention;
FIG. 3 is a schematic diagram of a complexity calculation flow in embodiment 1 of the present invention;
FIG. 4 is a schematic diagram of a system for complexity-based scheduling of picture segments to a hub in accordance with the present invention;
FIG. 5 is a schematic diagram of complexity calculation flow according to embodiment 4 of the present invention;
Detailed Description
The present invention will be described in further detail with reference to examples and drawings, but the present invention is not limited to these examples.
In the description of the present invention, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "longitudinal", "lateral", "horizontal", "inner", "outer", "front", "rear", "top", "bottom", and the like indicate orientations or positional relationships that are based on the orientations or positional relationships shown in the drawings, or that are conventionally placed when the product of the present invention is used, and are used only for convenience in describing and simplifying the description, but do not indicate or imply that the device or element referred to must have a particular orientation, be constructed in a particular orientation, and be operated, and thus should not be construed as limiting the invention.
In the description of the present invention, it should also be noted that, unless otherwise explicitly specified or limited, the terms "disposed," "open," "mounted," "connected," and "connected" are to be construed broadly, e.g., as meaning either a fixed connection, a removable connection, or an integral connection; can be mechanically or electrically connected; they may be connected directly or indirectly through intervening media, or they may be interconnected between two elements. The specific meanings of the above terms in the present invention can be understood in specific cases to those skilled in the art.
Example 1
The present embodiment aims to provide a method for scheduling picture streams to a center based on complexity.
For the intelligent security check field, the complexity of the picture to be recognized refers to the inherent difficulty level of finding or extracting a real parcel in a given image, which is related to the passenger flow situation of subway security check, including the following situations:
in the first situation, the identified object in the picture is blocked or partially blocked, namely the object is overlapped in the X-ray imaging picture;
in the second situation, the material of the article in the picture is too complex, so that the pixel color of the X-ray imaged picture is complex;
and in the third case, the object in the picture is deformed and twisted, so that the pixel colors of the picture imaged by the X-ray are complex.
The method comprises the steps of converting a picture to be identified into a gray picture, obtaining a gray matrix M ', wherein the gray consistency can reflect the uniformity of the image, if the value of the gray matrix M' is smaller, the gray matrix M 'corresponds to a simple image, and otherwise, the gray matrix M' corresponds to a complex image.
And for the cases II and III, performing convolution operation on the gray level picture through a Laplace operator to obtain a sharpening matrix L reflecting the color change degree.
For the case one, the grayscale image is subjected to negation operation to obtain a negation grayscale image, and the obtained negation matrix M is added to the complexity calculation by using the factor of a darker region generated by overlapping objects in the image.
Specifically, assume that the pixel function of the picture to be recognized is f (x, y), and x and y are horizontal and vertical coordinates respectively.
Then, the laplacian operator is:
since picture pixels are discrete data, the above equation can be approximated as:
the above equation can be transformed into a matrix form:
after optimization, the method comprises the following steps:
the matrix is used as a single convolution kernel, the convolution operation is carried out on the matrix M formed by the picture pixels, and the obtained matrix L is a Laplace transform matrix and indicates the intensity of the change of the picture color.
Specifically, a mark matrix C is established, wherein each element of C can only be 0 or 1, where 0 represents that the area where the pixel is located is not marked as an article by the edge marking box detection, and 1, on the contrary, belongs to a part marked as an article by the edge marking box detection.
In one embodiment, that is, in the implementation of g, all elements of a gray matrix M' are grayed to be changed into two bytes which are represented by 0 to 255, then the two bytes are subjected to inverse coding, and the obtained matrix M and matrix C are multiplied by each other in a bit-to-bit manner to form a matrix B:
B＝h(M，C)
then add each element of B to get sum (B).
In the same way, the matrix L and the matrix C are multiplied in the counterpoint mode to obtain a new matrix A:
A＝u(L，C)
thus, in this example, we obtain a specific formula for the picture to be recognized:
k＝α×SUM(h(M，C))+β×SUM(u(M，C))
alpha and beta are positive real numbers adjusted according to the actual safety inspection condition of the subway.
According to practical experience, a specific threshold v can be obtained for judging whether the picture is complex, when k is smaller than v, the picture is a simple picture, and when k is larger than or equal to v, the picture is a complex picture.
Example 2
The method for dispatching the picture shunt to the center based on the complexity is applied to an edge map recognition box, and specifically comprises the following steps:
s1, receiving a picture to be identified sent by an X-ray machine;
s2, analyzing the picture to be recognized to obtain the complexity of the picture to be recognized;
s3, determining the picture to be recognized corresponding to the complexity of the picture to be recognized which is greater than the complexity threshold value as a complex picture based on the complexity of the picture to be recognized and a preset complexity threshold value;
and S4, sending the picture to be identified which is judged to be the complex picture to the central server.
In the step S2, a complexity model is used to obtain the complexity of the picture to be recognized, where the complexity model includes transformation processing and labeling processing, the transformation processing is inverse transformation, and the complexity model specifically includes the following steps:
s201, converting the picture to be identified to obtain a conversion matrix D;
s202, marking the picture to be identified to obtain a marking matrix C;
s203, obtaining the complexity K of the picture to be identified according to the transformation matrix D and the mark matrix C.
The method specifically comprises the following steps:
s200, converting the picture to be identified into a gray picture;
s201, performing inverse transformation on the gray-scale picture to obtain an inverse matrix M;
s202, marking the picture to be identified to obtain a marking matrix C;
s203, multiplying the negation matrix M and the mark matrix C in a contraposition mode to obtain a matrix B, summing all elements of the matrix B to obtain SUM (B),
the sum (b) is the complexity K of the picture to be identified.
Specifically, the label matrix C is: and carrying out binarization processing on the acquired picture to be recognized to obtain a corresponding mark matrix C, wherein each element in the mark matrix C represents the marking degree of a corresponding pixel point in the picture to be recognized.
Specifically, each element in the marking matrix C is obtained according to the following manner:
for each element in the type matrix C, judging whether a corresponding pixel point of the element in the picture to be identified is in a marked article area, if not, the pixel point is 0, and if so, the pixel point is 1; and determining the marking degree of the element based on the 0 or 1.
Example 3
The method for dispatching the picture shunt to the center based on the complexity is applied to an edge map recognition box, and specifically comprises the following steps:
s1, receiving a picture to be identified sent by an X-ray machine;
s2, analyzing the picture to be recognized to obtain the complexity of the picture to be recognized;
s3, determining the picture to be recognized corresponding to the complexity of the picture to be recognized which is greater than the complexity threshold value as a complex picture based on the complexity of the picture to be recognized and a preset complexity threshold value;
and S4, sending the picture to be identified which is judged to be the complex picture to the central server.
In the step S2, a complexity model is used to obtain the complexity of the picture to be recognized, where the complexity model includes transformation processing and labeling processing, the transformation processing is laplace transformation, and the complexity model specifically includes the following steps:
s201, converting the picture to be identified to obtain a conversion matrix D;
s202, marking the picture to be identified to obtain a marking matrix C;
s203, obtaining the complexity K of the picture to be identified according to the transformation matrix D and the mark matrix C.
The method specifically comprises the following steps:
s200, converting the picture to be identified into a gray picture;
s201, performing Laplace operator transformation on the gray level picture to obtain a sharpening matrix L;
s202, marking the picture to be identified to obtain a marking matrix C;
s203, multiplying the sharpening matrix L and the marking matrix C in a contraposition mode to obtain a matrix A, summing all elements of the matrix A to obtain SUM (A),
the sum (a) is the complexity K of the picture to be identified.
Specifically, the label matrix C is: and carrying out binarization processing on the acquired picture to be recognized to obtain a corresponding mark matrix C, wherein each element in the mark matrix C represents the marking degree of a corresponding pixel point in the picture to be recognized.
Specifically, each element in the marking matrix C is obtained according to the following manner:
for each element in the type matrix C, judging whether a corresponding pixel point of the element in the picture to be identified is in a marked article area, if not, the pixel point is 0, and if so, the pixel point is 1; and determining the marking degree of the element based on the 0 or 1.
Specifically, the Laplace operator is
Example 4
The method for dispatching the picture shunt to the center based on the complexity is applied to an edge map recognition box, and specifically comprises the following steps:
s1, receiving a picture to be identified sent by an X-ray machine;
s2, analyzing the picture to be recognized to obtain the complexity of the picture to be recognized;
s3, determining the picture to be recognized corresponding to the complexity of the picture to be recognized which is greater than the complexity threshold value as a complex picture based on the complexity of the picture to be recognized and a preset complexity threshold value;
and S4, sending the picture to be identified which is judged to be the complex picture to the central server.
Specifically, the step S2 of obtaining the complexity of the picture to be recognized by using a complexity model, a consistency model, and an entropy model specifically includes the following steps:
s001, inputting the picture to be recognized into a complexity model to obtain the complexity K of the picture to be recognized;
s002, inputting the picture to be recognized into a consistency model to obtain the complexity U of the picture to be recognized;
s003, inputting the picture to be recognized into an entropy model to obtain the complexity S of the picture to be recognized;
and S004, carrying out weighted summation on the complexity K, U, S of the picture to be recognized to obtain the complexity Q of the picture to be recognized, which is a multiplied by K + b multiplied by U + c multiplied by S, wherein a, b and c are corresponding weighting coefficients.
And introducing a consistency model for reflecting the uniformity of the picture to be recognized, wherein if the U value is smaller, the image corresponds to a simple image, and otherwise, the image corresponds to a complex image.
An entropy model is introduced for measuring the randomness of the image texture, and the S value is larger when the P (i, j) values in the co-occurrence matrix are equal.
Specifically, in the step S2, a complexity model is used to obtain the complexity of the picture to be recognized, where the complexity model includes a transformation process and a labeling process, the transformation process includes at least one of a laplace transform and an inverse transform, and the complexity model specifically includes the following steps:
s201, converting the picture to be identified to obtain a conversion matrix D;
s202, marking the picture to be identified to obtain a marking matrix C;
s203, obtaining the complexity K of the picture to be identified according to the transformation matrix D and the mark matrix C.
Specifically, in the step S2, a complexity model is used to obtain the complexity of the picture to be recognized, where the complexity model includes transform processing and labeling processing, the transform processing includes laplace transform and inverse transform, and the complexity model specifically includes the following steps:
s200, converting the picture to be identified into a gray picture;
s201, performing inverse transformation on the gray-scale picture to obtain an inverse matrix M, and performing Laplace operator transformation on the gray-scale picture to obtain a sharpening matrix L;
s202, marking the picture to be identified to obtain a marking matrix C;
s203, multiplying the negation matrix M and the mark matrix C in a contraposition mode to obtain a matrix B, summing all elements of the matrix B to obtain SUM (B),
multiplying the sharpening matrix L and the marking matrix C in a contraposition mode to obtain a matrix A, summing all elements of the matrix A to obtain SUM (A),
and performing weighted summation on the sum (b) and sum (a) to obtain the complexity K of the picture to be recognized, where K is α × sum (b) + β × sum (a), where α and β are corresponding weighting coefficients.
The mark matrix C is: and carrying out binarization processing on the acquired picture to be recognized to obtain a corresponding mark matrix C, wherein each element in the mark matrix C represents the marking degree of a corresponding pixel point in the picture to be recognized.
Each element in the marking matrix C is obtained according to the following mode:
for each element in the type matrix C, judging whether a corresponding pixel point of the element in the picture to be identified is in a marked article area, if not, the pixel point is 0, and if so, the pixel point is 1; and determining the marking degree of the element based on the 0 or 1.
The laplacian operator is
Specifically, the consistency model specifically includes the following steps:
SA, converting the picture to be identified into a gray picture, and obtaining a gray matrix M' of the gray picture;
SB, substituting all elements in the gray matrix M' into a formula
Obtaining a second complexity U of the picture to be recognized, wherein m and n are respectively the number of rows and the number of columns of the picture to be recognized, and f (i, j) is the gray value of the pixel (i, j) of the gray picture,
is the mean of the gray levels of the 3 × 3 neighborhood pixels centered on pixel (i, j).
Specifically, the entropy model specifically includes the following steps:
sa, converting the picture to be identified into a gray picture, and obtaining a gray matrix M' of the gray picture;
sb, obtaining a gray level co-occurrence matrix D according to the gray level matrix M' of the gray level picture;
sc, substituting all elements in the gray level co-occurrence matrix D into a formula
And obtaining a third complexity S of the picture to be recognized, wherein P (i, j) is the ith row in the gray level co-occurrence matrix D, the jth column element value, and N is the gray level of the gray level picture.
Example 5
An object of this embodiment is to provide an apparatus for scheduling picture streams to a hub based on complexity, including:
the receiving module is used for receiving the picture to be identified sent by the X-ray machine;
the complexity module is used for analyzing the picture to be identified to obtain the complexity of the picture to be identified;
the judging module is used for judging the picture to be identified corresponding to the complexity of the picture to be identified which is greater than the complexity threshold value as a complex picture based on the complexity of the picture to be identified and a preset complexity threshold value;
and the scheduling module is used for sending the picture to be identified which is judged to be the complex picture to the central server.
Example 6
A system for complexity-based hierarchical scheduling of pictures to a hub, comprising:
the system comprises a central server, an edge computing node and a plurality of X-ray machines of security check points, wherein the edge computing node is composed of a plurality of edge image recognition boxes; the central server is connected with each edge attempt box and each edge recognition box is connected with the X-ray machine of the security check point where the edge recognition box is located;
the central server is used for receiving the pictures to be identified which are judged to be complex pictures and sent by each edge image identifying box and carrying out image identifying processing;
each edge map box is used for:
receiving a picture to be identified sent by an X-ray machine;
analyzing the picture to be identified to obtain the complexity of the picture to be identified;
determining the picture to be recognized corresponding to the complexity of the picture to be recognized which is greater than the complexity threshold value as a complex picture based on the complexity of the picture to be recognized and a preset complexity threshold value;
sending the picture to be identified which is judged to be the complex picture to a central server;
the X-ray machine of the security check point is used for scanning the passenger's packages, obtaining X-ray scanning pictures and sending the X-ray scanning pictures to the edge image identifying box of the security check point.
Embodiment 6, a computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, is able to carry out the method for complexity-based scheduling of a picture stream to a hub.
The foregoing is only a preferred embodiment of the present invention, and the present invention is not limited thereto in any way, and any simple modification, equivalent replacement and improvement made to the above embodiment within the spirit and principle of the present invention still fall within the protection scope of the present invention.
Claims (10)
1. The method for dispatching the picture shunt to the center based on the complexity is characterized by being applied to an edge map recognition box and specifically comprising the following steps of:
s1, receiving a picture to be identified sent by an X-ray machine;
s2, analyzing the picture to be recognized to obtain the complexity of the picture to be recognized;
s3, determining the picture to be recognized corresponding to the complexity of the picture to be recognized which is greater than the complexity threshold value as a complex picture based on the complexity of the picture to be recognized and a preset complexity threshold value;
and S4, sending the picture to be identified which is judged to be the complex picture to the central server.
2. The method according to claim 1, wherein the complexity model is used in step S2 to obtain the complexity of the picture to be recognized, the complexity model includes a transformation process and a labeling process, the transformation process includes at least one of a laplacian transformation and an inverse transformation, and the complexity model specifically includes the following steps:
s201, converting the picture to be identified to obtain a conversion matrix D;
s202, marking the picture to be identified to obtain a marking matrix C;
s203, obtaining the complexity K of the picture to be identified according to the transformation matrix D and the mark matrix C.
3. The method according to claim 1, wherein the complexity model is used in step S2 to obtain the complexity of the picture to be recognized, the complexity model includes a transformation process and a labeling process, the transformation process includes a laplacian transformation and an inverse transformation, and the complexity model specifically includes the following steps:
s200, converting the picture to be identified into a gray picture;
s201, performing inverse transformation on the gray-scale picture to obtain an inverse matrix M, and performing Laplace operator transformation on the gray-scale picture to obtain a sharpening matrix L;
s202, marking the picture to be identified to obtain a marking matrix C;
s203, multiplying the negation matrix M and the mark matrix C in a contraposition mode to obtain a matrix B, summing all elements of the matrix B to obtain SUM (B),
multiplying the sharpening matrix L and the marking matrix C in a contraposition mode to obtain a matrix A, summing all elements of the matrix A to obtain SUM (A),
and performing weighted summation on the sum (b) and sum (a) to obtain the complexity K of the picture to be recognized, where K is α × sum (b) + β × sum (a), where α and β are corresponding weighting coefficients.
4. A method for complexity-based picture streaming scheduling according to claim 2 or 3, wherein the marking is performed by: and carrying out binarization processing on the acquired picture to be recognized to obtain a corresponding mark matrix C, wherein each element in the mark matrix C represents the marking degree of a corresponding pixel point in the picture to be recognized.
5. The method for complexity-based picture streaming scheduling according to claim 4, wherein each element in the marking matrix C is obtained according to the following way:
for each element in the type matrix C, judging whether a corresponding pixel point of the element in the picture to be identified is in a marked article area, if not, the pixel point is 0, and if so, the pixel point is 1; and determining the marking degree of the element based on the 0 or 1.
6. The method for complexity-based picture streaming scheduling to the center according to claim 1, wherein the complexity of the picture to be recognized is obtained in the step S2 by using a consistency model, and the consistency model specifically includes the following steps:
SA, converting the picture to be identified into a gray picture, and obtaining a gray matrix M' of the gray picture;
SB, substituting all elements in the gray matrix M' into a formula
Obtaining the complexity U of the picture to be identified, wherein m and n are the number of rows and columns of the picture to be identified respectively, f (i, j) is the gray value of the pixel (i, j) of the gray picture,
is 3 centered on the pixel (i, j)Grayscale mean of x 3 neighborhood pixels.
7. The method for complexity-based picture shunting to center according to claim 1, wherein the complexity of the picture to be recognized is obtained in step S2 by using an entropy model, and the entropy model specifically includes the following steps:
sa, converting the picture to be identified into a gray picture, and obtaining a gray matrix M' of the gray picture;
sb, obtaining a gray level co-occurrence matrix D according to the gray level matrix M' of the gray level picture;
sc, substituting all elements in the gray level co-occurrence matrix D into a formula
And obtaining the complexity S of the picture to be recognized, wherein P (i, j) is the ith row and the jth column in the gray level co-occurrence matrix D, and N is the gray level of the gray level picture.
8. The method for complexity-based picture streaming scheduling to the center according to any one of claims 2, 6, and 7, wherein the step S2 of obtaining the complexity of the picture to be recognized by using a complexity model, a consistency model, and an entropy model specifically includes the following steps:
s001, inputting the picture to be recognized into a complexity model to obtain the complexity K of the picture to be recognized;
s002, inputting the picture to be recognized into a consistency model to obtain the complexity U of the picture to be recognized;
s003, inputting the picture to be recognized into an entropy model to obtain the complexity S of the picture to be recognized;
and S004, carrying out weighted summation on the complexity K, U, S of the picture to be recognized to obtain the complexity Q of the picture to be recognized, which is a multiplied by K + b multiplied by U + c multiplied by S, wherein a, b and c are corresponding weighting coefficients.
9. Apparatus for complexity-based hierarchical scheduling of pictures to a hub, comprising:
the receiving module is used for receiving the picture to be identified sent by the X-ray machine;
the complexity module is used for analyzing the picture to be identified to obtain the complexity of the picture to be identified;
the judging module is used for judging the picture to be identified corresponding to the complexity of the picture to be identified which is greater than the complexity threshold value as a complex picture based on the complexity of the picture to be identified and a preset complexity threshold value;
and the scheduling module is used for sending the picture to be identified which is judged to be the complex picture to the central server.
10. A system for complexity-based hierarchical scheduling of pictures to a hub, comprising:
the system comprises a central server, an edge computing node and a plurality of X-ray machines of security check points, wherein the edge computing node is composed of a plurality of edge image recognition boxes; the central server is connected with each edge attempt box and each edge recognition box is connected with the X-ray machine of the security check point where the edge recognition box is located;
the central server is used for receiving the pictures to be identified which are judged to be complex pictures and sent by each edge image identifying box and carrying out image identifying processing;
each edge map box is used for:
receiving a picture to be identified sent by an X-ray machine;
analyzing the picture to be identified to obtain the complexity of the picture to be identified;
determining the picture to be recognized corresponding to the complexity of the picture to be recognized which is greater than the complexity threshold value as a complex picture based on the complexity of the picture to be recognized and a preset complexity threshold value;
sending the picture to be identified which is judged to be the complex picture to a central server;
the X-ray machine of the security check point is used for scanning the passenger's packages, obtaining X-ray scanning pictures and sending the X-ray scanning pictures to the edge image identifying box of the security check point.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210019824.8A CN114359023B (en) | 2022-01-10 | 2022-01-10 | Method, equipment and system for dispatching picture shunt to center based on complexity |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210019824.8A CN114359023B (en) | 2022-01-10 | 2022-01-10 | Method, equipment and system for dispatching picture shunt to center based on complexity |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114359023A true CN114359023A (en) | 2022-04-15 |
| CN114359023B CN114359023B (en) | 2022-11-18 |
Family
ID=81107995
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210019824.8A Active CN114359023B (en) | 2022-01-10 | 2022-01-10 | Method, equipment and system for dispatching picture shunt to center based on complexity |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114359023B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104732545A (en) * | 2015-04-02 | 2015-06-24 | 西安电子科技大学 | Texture image segmentation method combined with sparse neighbor propagation and rapid spectral clustering |
| CN107895139A (en) * | 2017-10-19 | 2018-04-10 | 金陵科技学院 | A kind of SAR image target recognition method based on multi-feature fusion |
| CN108389216A (en) * | 2018-02-06 | 2018-08-10 | 西安交通大学 | Local auto-adaptive threshold segmentation method towards on-line ferrograph image wear Particles Recognition |
| CN109255815A (en) * | 2018-08-30 | 2019-01-22 | 天津理工大学 | A kind of object detection and recognition methods based on order spherical harmonic |
| WO2019015418A1 (en) * | 2017-07-18 | 2019-01-24 | Oppo广东移动通信有限公司 | Unlocking control method and related product |
| CN109711284A (en) * | 2018-12-11 | 2019-05-03 | 江苏博墨教育科技有限公司 | A kind of test answer sheet system intelligent recognition analysis method |
| CN111242923A (en) * | 2020-01-13 | 2020-06-05 | 国网湖北省电力有限公司宜昌供电公司 | Canny operator-based insulator crack detection method based on maximum entropy threshold optimization |
| CN112926568A (en) * | 2021-03-05 | 2021-06-08 | 北京全路通信信号研究设计院集团有限公司 | Interlocking table blueprint identification method and device, electronic equipment and storage medium |
-
2022
- 2022-01-10 CN CN202210019824.8A patent/CN114359023B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104732545A (en) * | 2015-04-02 | 2015-06-24 | 西安电子科技大学 | Texture image segmentation method combined with sparse neighbor propagation and rapid spectral clustering |
| WO2019015418A1 (en) * | 2017-07-18 | 2019-01-24 | Oppo广东移动通信有限公司 | Unlocking control method and related product |
| CN107895139A (en) * | 2017-10-19 | 2018-04-10 | 金陵科技学院 | A kind of SAR image target recognition method based on multi-feature fusion |
| CN108389216A (en) * | 2018-02-06 | 2018-08-10 | 西安交通大学 | Local auto-adaptive threshold segmentation method towards on-line ferrograph image wear Particles Recognition |
| CN109255815A (en) * | 2018-08-30 | 2019-01-22 | 天津理工大学 | A kind of object detection and recognition methods based on order spherical harmonic |
| CN109711284A (en) * | 2018-12-11 | 2019-05-03 | 江苏博墨教育科技有限公司 | A kind of test answer sheet system intelligent recognition analysis method |
| CN111242923A (en) * | 2020-01-13 | 2020-06-05 | 国网湖北省电力有限公司宜昌供电公司 | Canny operator-based insulator crack detection method based on maximum entropy threshold optimization |
| CN112926568A (en) * | 2021-03-05 | 2021-06-08 | 北京全路通信信号研究设计院集团有限公司 | Interlocking table blueprint identification method and device, electronic equipment and storage medium |
Non-Patent Citations (4)
| Title |
|---|
| WU,Y 等: "n-Dimensional Discrete Cat Map Generation Using Laplace Expansions", 《IEEE TRANSACTIONS ON CYBERNETICS》 * |
| 余海燕: "基于多目立体视觉的体育舞蹈错误动作校正方法", 《赤峰学院学报(自然科学版)》 * |
| 陆春君等: "基于纹理频谱子集的纹理识别", 《江南大学学报(自然科学版)》 * |
| 高振宇等: "图像复杂度描述方法研究", 《中国图象图形学报》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114359023B (en) | 2022-11-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110060237B (en) | Fault detection method, device, equipment and system | |
| US9558403B2 (en) | Chemical structure recognition tool | |
| CN109741551B (en) | Commodity identification settlement method, device and system | |
| CN111695609A (en) | Target damage degree determination method, target damage degree determination device, electronic device, and storage medium | |
| CN111597857B (en) | Logistics package detection method, device, equipment and readable storage medium | |
| CN112149533A (en) | Target detection method based on improved SSD model | |
| CN112001362A (en) | Image analysis method, image analysis device and image analysis system | |
| CN107133647A (en) | A kind of quick Manuscripted Characters Identification Method | |
| CN112070079B (en) | X-ray contraband package detection method and device based on feature map weighting | |
| CN114972922A (en) | Coal and gangue sorting and identifying method, device and equipment based on machine learning | |
| CN111767915A (en) | License plate detection method, device, equipment and storage medium | |
| CN110766007A (en) | Certificate shielding detection method, device and equipment and readable storage medium | |
| CN115302963A (en) | Bar code printing control method, system and medium based on machine vision | |
| CN111597875A (en) | Traffic sign identification method, device, equipment and storage medium | |
| CN111985269A (en) | Detection model construction method, detection device, server and medium | |
| CN114219402A (en) | Logistics tray stacking identification method, device, equipment and storage medium | |
| CN111079575B (en) | Material identification method and system based on package image characteristics | |
| CN114359023B (en) | Method, equipment and system for dispatching picture shunt to center based on complexity | |
| CN111507119B (en) | Identification code recognition method, identification code recognition device, electronic equipment and computer readable storage medium | |
| CN110458004B (en) | Target object identification method, device, equipment and storage medium | |
| CN113239738B (en) | Image blurring detection method and blurring detection device | |
| CN112288372B (en) | Express bill identification method capable of simultaneously identifying one-dimensional bar code and three-segment code characters | |
| CN114663803A (en) | Logistics center hanging clothing classification method and device based on video streaming | |
| CN113191227A (en) | Cabinet door state detection method, device, equipment and storage medium | |
| CN111797922A (en) | Text image classification method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |