An intelligent baking production line control system

Abstract

The present invention relates to the field of baking technology, and in particular to an intelligent baking production line control system, comprising a control terminal, which is a main control end of the system and is used to issue control commands, a capture module, which is used to capture cake blank image data transmitted on the baking production line, and a combination unit, which is used to receive all cake blank image data collected in the capture module and combine the cake blank image data. The present invention can collect images of cake blanks transmitted on the baking production line, and extract contour images of cake blanks in the image data based on the collected cake blank image data, and then perform similarity comparison with the extracted cake blank contour images, and finally realize quality evaluation of cake blanks, thereby avoiding the artificial visual inspection link of cake blanks, saving cake blank production costs, improving cake blank production efficiency and precision, and bringing quality control to cake blank production.

Description

Technical Field

[0001] The present invention relates to the technical field of baking, and in particular to an intelligent baking production line control system. Background Art

[0002] A baking production line refers to a combination of equipment and processes used to produce a variety of baked goods, which usually includes raw material processing, molding, baking, cooling, and packaging.

[0003] However, in the cake blank forming stage of the baking production line, most of the current methods are to screen out cake blanks with appearance defects by manual visual inspection. This process consumes a lot of labor costs and has poor efficiency. At the same time, due to the use of manual inspection, the accuracy of cake blank defect detection is also limited. Summary of the invention

[0004] In view of the above-mentioned shortcomings of the prior art, the present invention provides an intelligent baking production line control system, which solves the technical problems raised in the above-mentioned background technology.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0006] An intelligent baking production line control system includes: a control terminal, which is the main control end of the system and is used to issue control commands; a capture module, which is used to capture cake blank image data transmitted on the baking production line; a combination unit, which is used to receive all cake blank image data collected in the capture module and combine the cake blank image data; an enhancement unit, which is used to receive cake blank image data subjected to combination processing in the combination unit and perform image enhancement on the cake blank image data obtained by the combination processing; an extraction module, which is used to receive cake blank image data captured in the capture module and perform image contour extraction on the cake blank image data; a comparison module, which is used to obtain the image contour extracted in the extraction module; Performing similarity comparison on image contours; a logic unit for setting a cake blank contour image selection logic applied by a comparison module when performing contour image similarity comparison; a selection unit for receiving the contour image selection logic set in the logic unit, and selecting cake blank contour images based on the contour image selection logic; a determination module for setting a similarity determination threshold, receiving image contour similarity comparison results in the comparison module, and determining whether image contours corresponding to the image contour similarity comparison results are similar based on the similarity determination threshold; an evaluation module for receiving image contour similarity determination results in the determination module, and evaluating whether cake blanks transmitted on the current baking production line are qualified based on the similarity determination results;

[0007] The control terminal is interactively connected to a capture module via a local area network, the capture module is electrically connected to a combination unit and an enhancement unit at a lower level via a medium, the capture module is interactively connected to an extraction module via a local area network, the extraction module is interactively connected to the enhancement unit via a local area network, the extraction module is interactively connected to a comparison module via a local area network, the comparison module is electrically connected to a logic unit and a selection unit at a lower level via a medium, and the comparison module is interactively connected to a determination module and an evaluation module via a local area network.

[0008] Furthermore, the capture module is composed of several groups of cameras, and several groups of cameras are synchronously operated to collect image data of cake blanks transmitted on the baking production line. Several groups of cameras are equidistantly distributed in a straight line in the transmission direction of the cake blanks on the baking production line. When the combination unit combines the cake blank image data, it performs continuous combination based on the transmission direction of the cake blanks on the baking production line. The cake blank image data obtained by the combined processing records the image data of the cake blank group, the background of the cake blank group image data is the baking production line, and the foreground of the cake blank group image data is each cake blank transmitted on the baking production line.

[0009] Furthermore, the enhancement unit performs image enhancement on the cake blank image data obtained by the combination processing, that is, performs enhancement processing on the cake blank group image data, and the enhancement processing logic of the image data is expressed as follows:

[0010]

[0011] Where: x′ i,j is the gray value of the image after image data enhancement; H(x i,j ) is the equalized gray value; V D is the dark adaptation function; x i,j is the grayscale image value; s is the enhancement parameter; m i,j For x i,j The neighborhood mean centered at ▽x i,j is the gradient value of the image at (i, j) after gradient processing; ε is the noise gradient threshold; V L is the bright adaptation function;

[0012] Among them, the image gray value x′ after image data enhancement processing i,j After obtaining, based on x′ i,j Perform image enhancement on the cake dough group image data, the enhancement parameter s ≥ 0, and is proportional to the gradient. λ is the amplification factor;

[0013] The dark adaptation function V D and the light adaptation function V L The manifestation is:

[0014]

[0015] Where: a is the adaptive adjustment parameter, a∈[0,1]; V str is the image brightness component;

[0016] The value of the adaptive adjustment parameter a follows the setting logic that the more cake doughs there are in the foreground of the cake dough group image data, the larger the value of the adaptive adjustment parameter a is, and vice versa.

[0017] Furthermore, the cake blank image data received by the extraction module during the operation phase, i.e., the cake blank group image data after image enhancement, the image contour extraction performed on the cake blank image data in the extraction module, i.e., the extraction of the contour image of the cake blank in the foreground of the cake blank group image data;

[0018] Among them, the overall color of the baking production line corresponding to the background of the cake blank group image data is consistent, and the colors of the cake blanks corresponding to the foreground of the cake blank group image data are consistent with each other. When the extraction module extracts the contour image of the cake blank, it extracts the contour image of the cake blank based on the boundary between the color of the baking production line and the color of the cake blank.

[0019] Furthermore, the operation of performing similarity comparison on the image contours in the comparison module is an operation of performing similarity comparison on the contour images of each cake blank;

[0020] The more cake dough outline images are included in the foreground of the cake dough group image data, the more cake dough outline images are selected;

[0021] There are no more than two groups of cake blank outline images adjacent to each other in the foreground of the selected cake blank outline images;

[0022] The cake blank contour image in the foreground of the selected cake blank contour image is no larger than forty percent of the cake blank contour images included in all the foregrounds.

[0023] Furthermore, the contour image similarity comparison logic used in the comparison module is expressed as:

[0024]

[0025] Where: sim(a,b) is the similarity between contour image a and contour image b; S i is the area of ​​the closed geometric figure after the i-th discrete curve evolution of the contour image; S0 is the area of ​​the closed geometric figure enclosed by the contour image; n0 is the number of sampling points applied in the discrete curve evolution; n is the set of points on the contour image; θ qis the curvature of the points on the qth group of contour images;

[0026] in, Represents the area difference of the contour image a after the discrete curve evolution, represents the area difference of the contour image b after the discrete curve evolution, represents the sum of the curvatures of each point on the contour image a. It represents the sum of the curvatures of each point on the contour image b. The larger the value of sim(a,b), the higher the similarity between the contour images a and b. Conversely, the lower the similarity between the contour images a and b.

[0027] Furthermore, after completing the determination of whether the image contours are similar, the determination module further counts the determination results to obtain the ratio of the number of determination results of similar image contours to the number of determination results of dissimilar image contours.

[0028] Furthermore, during the operation stage of the evaluation module, the ratio of the number of image contour similarity and image contour dissimilarity judgment results in the judgment module is synchronously obtained. When the ratio is greater than or equal to 19:1, the evaluation result is that the cake blanks transmitted on the current baking production line are qualified; otherwise, the evaluation result is that the cake blanks transmitted on the current baking production line are unqualified.

[0029] Compared with the known public technology, the technical solution provided by the present invention has the following beneficial effects:

[0030] The present invention provides an intelligent baking production line control system. During operation, the system can collect images of cake blanks transmitted on the baking production line, and extract contour images of the cake blanks in the image data based on the collected image data of the cake blanks, and then perform similarity comparison with the extracted contour images of the cake blanks, and finally realize quality evaluation of the cake blanks, thereby avoiding the manual visual inspection link of the cake blanks, saving the production cost of the cake blanks, improving the production efficiency and accuracy of the cake blanks, and bringing quality control to the production of the cake blanks. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0032] Figure 1 It is a structural diagram of a control system for an intelligent baking production line;

[0033] Figure 2Schematic diagram of the evolution process of the discrete curve of the cake blank contour image in the present invention;

[0034] The numbers in the figure represent: 1. control terminal; 2. capture module; 21. combination unit; 22. enhancement unit; 3. extraction module; 4. comparison module; 41. logic unit; 42. selection unit; 5. judgment module; 6. evaluation module. DETAILED DESCRIPTION

[0035] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0036] The present invention will be further described below in conjunction with the embodiments.

[0037] Example 1

[0038] An intelligent baking production line control system of this embodiment, such as Figure 1 As shown, it includes a control terminal 1, the control terminal 1 is interactively connected to a capture module 2 through a local area network, the capture module 2 is electrically connected to a combination unit 21 and an enhancement unit 22 through a medium, the capture module 2 is interactively connected to an extraction module 3 through a local area network, the extraction module 3 is interactively connected to the enhancement unit 22 through the local area network, the extraction module 3 is interactively connected to a comparison module 4 through the local area network, the comparison module 4 is electrically connected to a logic unit 41 and a selection unit 42 through a medium, and the comparison module 4 is interactively connected to a determination module 5 and an evaluation module 6 through the local area network;

[0039] The control terminal 1 is the main control terminal of the system and is used to issue control commands. The capture module 2 is used to capture the cake blank image data transmitted on the baking production line. The combination unit 21 is used to receive all the cake blank image data collected in the capture module 2 and combine the cake blank image data. The enhancement unit 22 is used to receive the cake blank image data that has been combined in the combination unit 21 and perform image enhancement on the cake blank image data obtained by the combination. The extraction module 3 is used to receive the cake blank image data captured in the capture module 2 and perform image contour extraction on the cake blank image data. The comparison module 4 is used to obtain the image contour extracted in the extraction module 3 and perform image enhancement on the image contour. Similarity comparison, logic unit 41, used to set the cake blank contour image selection logic applied by comparison module 4 when performing contour image similarity comparison, selection unit 42, used to receive the contour image selection logic set in logic unit 41, and select the cake blank contour image based on the contour image selection logic, judgment module 5, used to set a similarity judgment threshold, receive the image contour similarity comparison result in comparison module 4, and judge whether the image contours corresponding to the image contour similarity comparison result are similar based on the similarity judgment threshold, and evaluation module 6, used to receive the image contour similarity judgment result in judgment module 5, and judge whether the cake blank transmitted on the current baking production line is qualified based on the similarity judgment result;

[0040] The enhancement unit 22 performs image enhancement on the cake blank image data obtained by the combination process, that is, performs enhancement processing on the cake blank group image data. The enhancement processing logic of the image data is expressed as follows:

[0041]

[0042] Where: x′ i,j is the gray value of the image after image data enhancement; H(x i,j ) is the equalized gray value; V D is the dark adaptation function; x i,j is the grayscale image value; s is the enhancement parameter; m i,j For x i,j The neighborhood mean centered at ; is the gradient value of the image at (i, j) after gradient processing; ε is the noise gradient threshold; V L is the bright adaptation function;

[0043] Among them, the image gray value x′ after image data enhancement processing i,j After obtaining, based on x′ i,j Perform image enhancement on the cake dough group image data, the enhancement parameter s ≥ 0, and is proportional to the gradient. λ is the amplification factor;

[0044] Dark adaptation function VD and the light adaptation function V L The manifestation is:

[0045]

[0046] Where: a is the adaptive adjustment parameter, a∈[0,1]; V str is the image brightness component;

[0047] The value of the adaptive adjustment parameter a follows the setting logic that the more cake blanks there are in the foreground of the cake blank group image data, the larger the value of the adaptive adjustment parameter a is, and vice versa, the smaller the value of the adaptive adjustment parameter a is;

[0048] The contour image similarity comparison logic used in the comparison module 4 is expressed as:

[0049]

[0050] Where: sim(a,b) is the similarity between contour image a and contour image b; S i is the area of ​​the closed geometric figure after the i-th discrete curve evolution of the contour image; S0 is the area of ​​the closed geometric figure enclosed by the contour image; n0 is the number of sampling points applied in the discrete curve evolution; n is the set of points on the contour image; θ q is the curvature of the points on the qth group of contour images;

[0051] in, Represents the area difference of the contour image a after the discrete curve evolution, represents the area difference of the contour image b after the discrete curve evolution, represents the sum of the curvatures of each point on the contour image a. It represents the sum of the curvatures of each point on the contour image b. The larger the value of sim(a,b), the higher the similarity between the contour images a and b. Conversely, the lower the similarity between the contour images a and b.

[0052] After the determination module 5 completes the determination of whether the image contours are similar, it further counts the determination results to obtain the ratio of the number of determination results of similar image contours to the number of determination results of dissimilar image contours;

[0053] During the operation phase of the evaluation module 6, the ratio of the number of judgment results of image contour similarity and image contour dissimilarity in the judgment module 5 is synchronously obtained. When the ratio is greater than or equal to 19:1, the evaluation result is that the cake dough transmitted on the current baking production line is qualified; otherwise, the evaluation result is that the cake dough transmitted on the current baking production line is unqualified.

[0054] In this embodiment, the control terminal 1 controls the capture module 2 to run the capture of cake blank image data transmitted on the baking production line, the combination unit 21 receives all the cake blank image data collected in the capture module 2 in real time, and combines the cake blank image data, the enhancement unit 22 synchronously receives the cake blank image data that performs the combination processing in the combination unit 21, and performs image enhancement on the cake blank image data obtained by the combination processing, the extraction module 3 is post-operated to receive the cake blank image data captured in the capture module 2, and performs image contour extraction on the cake blank image data, the comparison module 4 further obtains the image contour extracted in the extraction module 3, and performs similarity comparison on the image contour. , the logic unit 41 synchronously sets the cake blank contour image selection logic applied by the comparison module 4 when performing contour image similarity comparison, the selection unit 42 receives the contour image selection logic set in the logic unit 41 in real time, selects the cake blank contour image based on the contour image selection logic, and then the determination module 5 sets the similarity determination threshold, receives the image contour similarity comparison result in the comparison module 4, determines whether the image contours corresponding to the image contour similarity comparison result are similar based on the similarity determination threshold, and finally receives the image contour similarity determination result in the determination module 5 through the evaluation module 6, and evaluates whether the cake blank transmitted on the current baking production line is qualified based on the similarity determination result;

[0055] Through the above embodiments, intelligent supervision and control are brought to the quality of cake blanks on the baking production line, ensuring that the cake blanks produced on the baking production line have better quality and lower appearance defect rate.

[0056] Example 2

[0057] In terms of specific implementation, based on Example 1, this example refers to Figure 1 The intelligent baking production line control system in Example 1 is further described in detail:

[0058] The capture module 2 is composed of several groups of cameras, which operate synchronously to collect image data of cake blanks transmitted on the baking production line. The several groups of cameras are distributed in a straight line and equidistantly in the transmission direction of the cake blanks on the baking production line. When the combination unit 21 combines the cake blank image data, it performs continuous combination based on the transmission direction of the cake blanks on the baking production line. The cake blank image data obtained by the combined processing records the image data of the cake blank group. The background of the cake blank group image data is the baking production line, and the foreground of the cake blank group image data is each cake blank transmitted on the baking production line.

[0059] Through the above settings, the operation logic of the capture module 2 and its lower-level submodules is further limited, and necessary operation data support is further provided for the operation of subsequent modules in the system.

[0060] like Figure 1 As shown, the cake blank image data received by the extraction module 3 during the operation phase, i.e., the cake blank group image data after image enhancement, the image contour extraction performed on the cake blank image data in the extraction module 3, i.e., the extraction of the contour image of the cake blank in the foreground of the cake blank group image data;

[0061] The overall color of the baking production line corresponding to the background of the cake blank group image data is consistent, and the colors of the cake blanks corresponding to the foreground of the cake blank group image data are consistent with each other. When extracting the contour image of the cake blank, the extraction module 3 extracts the contour image of the cake blank based on the boundary between the color of the baking production line and the color of the cake blank.

[0062] The operation of performing similarity comparison on the image contours in the comparison module 4, i.e., performing similarity comparison on the contour images of each cake blank;

[0063] The more cake dough outline images are included in the foreground of the cake dough group image data, the more cake dough outline images are selected;

[0064] There are no more than two groups of cake blank outline images adjacent to each other in the foreground of the selected cake blank outline images;

[0065] The cake blank contour image in the foreground of the selected cake blank contour image is no larger than forty percent of the cake blank contour images included in all the foregrounds.

[0066] Through the above configuration, an extraction logic for extracting the cake blank contour image from the cake blank image data is provided during the operation phase of the extraction module 3 .

[0067] In summary, during operation, the system in the above embodiment can capture images of cake blanks transmitted on the baking production line, and extract the contour image of the cake blank in the image data based on the captured cake blank image data, and then perform similarity comparison with the extracted cake blank contour image, and finally realize the quality assessment of the cake blank, thereby avoiding the manual visual inspection link of the cake blank, saving the production cost of the cake blank, improving the production efficiency and accuracy of the cake blank, and bringing quality control to the production of the cake blank.

[0068] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An intelligent baking production line control system, characterized in that: include: The control terminal (1) is the main control terminal of the system and is used to issue control commands; A capture module (2), used for capturing cake blank image data transmitted on a baking production line; An extraction module (3) is used to receive the cake blank image data captured by the capture module (2) and extract image contours from the cake blank image data; A comparison module (4) is used to obtain the image contour extracted by the extraction module (3) and perform a similarity comparison on the image contour; A determination module (5) is used to set a similarity determination threshold, receive the image contour similarity comparison result in the comparison module (4), and determine whether the image contours corresponding to the image contour similarity comparison result are similar based on the similarity determination threshold; An evaluation module (6) is used to receive the image contour similarity determination result in the determination module (5), and based on the similarity determination result, evaluate whether the cake blanks currently transmitted on the baking production line are qualified; The capture module (2) is composed of a plurality of groups of cameras, and the plurality of groups of cameras are operated synchronously to collect image data of cake blanks transmitted on the baking production line; The capture module (2) is provided with submodules at the lower level, including: A combining unit (21) is used to receive all cake blank image data collected by the capturing module (2) and combine the cake blank image data; An enhancement unit (22) is used to receive the cake blank image data subjected to the combination processing in the combination unit (21), and perform image enhancement on the cake blank image data obtained by the combination processing; wherein a plurality of groups of cameras are linearly and equidistantly distributed in the direction of cake blank transmission on the baking production line; when the combination unit (21) performs combination processing on the cake blank image data, it continuously combines based on the direction of cake blank transmission on the baking production line; the cake blank image data obtained by the combination processing records cake blank group image data, wherein the background of the cake blank group image data is the baking production line, and the foreground of the cake blank group image data is each cake blank transmitted on the baking production line; The operation of performing similarity comparison on the image contours in the comparison module (4) is an operation of performing similarity comparison on the contour images of each cake blank. The comparison module (4) is provided with a submodule at a lower level, including: A logic unit (41) is used to set the cake blank outline image selection logic used by the comparison module (4) when performing outline image similarity comparison; A selection unit (42) is used to receive the contour image selection logic set in the logic unit (41), and select the cake blank contour image based on the contour image selection logic; The contour image selection logic set in the logic unit (41) obeys: The more cake dough outline images are included in the foreground of the cake dough group image data, the more cake dough outline images are selected; There are no more than two groups of cake blank outline images adjacent to each other in the foreground of the selected cake blank outline images; The cake blank outline image in the foreground of the selected cake blank outline image is no larger than 40 percent of the cake blank outline images included in all the foregrounds; The contour image similarity comparison logic used in the comparison module (4) is expressed as: Where: sim(a,b) is the similarity between contour image a and contour image b; S i is the area of ​​the closed geometric figure after the i-th discrete curve evolution of the contour image; S0 is the area of ​​the closed geometric figure enclosed by the contour image; n0 is the number of sampling points applied in the discrete curve evolution; n is the set of points on the contour image; is the curvature of the points on the qth group of contour images; in, Represents the area difference of the contour image a after the discrete curve evolution, represents the area difference of the contour image b after the discrete curve evolution, represents the sum of the curvatures of each point on the contour image a. It represents the sum of the curvatures of each point on the contour image b. The larger the value of sim(a,b), the higher the similarity between the contour images a and b. Conversely, the lower the similarity between the contour images a and b. The control terminal (1) is interactively connected to a capture module (2) via a local area network; the capture module (2) is electrically connected to a combination unit (21) and an enhancement unit (22) via a medium; the capture module (2) is interactively connected to an extraction module (3) via a local area network; the extraction module (3) is interactively connected to the enhancement unit (22) via a local area network; the extraction module (3) is interactively connected to a comparison module (4) via a local area network; the comparison module (4) is electrically connected to a logic unit (41) and a selection unit (42) via a medium; the comparison module (4) is interactively connected to a determination module (5) and an evaluation module (6) via a local area network.

2. The intelligent baking production line control system according to claim 1, characterized in that: The enhancement unit (22) performs image enhancement on the cake blank image data obtained by the combination process, that is, performs enhancement processing on the cake blank group image data. The enhancement processing logic of the image data is expressed as follows: Where: x′ i,j is the gray value of the image after image data enhancement; H(x i,j ) is the equalized gray value; V D is the dark adaptation function; xi,j is the grayscale image value; s is the enhancement parameter; mi,j is the neighborhood mean centered on xi,j; ▽x i,j is the gradient value of the image at (i, j) after gradient processing; ε is the noise gradient threshold; V L is the bright adaptation function; Among them, the image gray value x′ after image data enhancement processing i,j After obtaining, based on x′ i,j Perform image enhancement on the cake dough group image data, the enhancement parameter s ≥ 0, and is proportional to the gradient. λ is the amplification factor.

3. The intelligent baking production line control system according to claim 2, characterized in that: The dark adaptation function V D and the light adaptation function V L The manifestation is: Where: a is the adaptive adjustment parameter, a∈[0,1]; V str is the image brightness component; The value of the adaptive adjustment parameter a follows the setting logic that the more cake doughs there are in the foreground of the cake dough group image data, the larger the value of the adaptive adjustment parameter a is, and vice versa.

4. The intelligent baking production line control system according to claim 1, characterized in that: The cake blank image data received by the extraction module (3) during the operation phase, i.e., the cake blank group image data after image enhancement, the image contour extraction performed on the cake blank image data in the extraction module (3), i.e., the extraction of the contour image of the cake blank in the foreground of the cake blank group image data; The background of the cake blank group image data corresponds to the overall color of the baking production line, and the foreground of the cake blank group image data corresponds to the colors of the cake blanks. When extracting the contour image of the cake blank, the extraction module (3) extracts the contour image of the cake blank based on the boundary between the color of the baking production line and the color of the cake blank.

5. The intelligent baking production line control system according to claim 1, characterized in that: After completing the determination of whether the image contours are similar, the determination module (5) further performs statistics on the determination results to obtain the ratio of the number of determination results of similar image contours to the number of determination results of dissimilar image contours.

6. The intelligent baking production line control system according to claim 1, characterized in that: During the operation phase of the evaluation module (6), the ratio of the number of image contour similarity judgment results to the number of image contour dissimilarity judgment results in the judgment module (5) is synchronously obtained. When the ratio is greater than or equal to 19:1, the evaluation result is that the cake blanks transmitted on the current baking production line are qualified. Otherwise, the evaluation result is that the cake blanks transmitted on the current baking production line are unqualified.

Images