Food production process monitoring method based on visual detection

Abstract

The present invention relates to the technical field of food production, and in particular to a food production process monitoring method based on visual detection, comprising: deploying a vibration device on the outer wall of a baking station of a baking device, deploying a camera module inside the baking station of the baking device, the baking device continuously running based on a set running logic, and during the continuous running of the baking device, the vibration device and the camera module are linked to run; based on the camera module, baking target image data is collected, and the present invention uses the baking target image data as a baking progress monitoring and judgment support in real time to enable the baking dynamic changes of the baking target in the baking device to be monitored in real time, thereby realizing intelligent control of baking the baking target to an optimal state, ensuring that after the baking target is baked, a semi-finished product with better baking quality is obtained, so as to improve the quality of the final product.

Description

Technical Field

[0001] The present invention relates to the technical field of food production, and specifically relates to a method for monitoring food production processes based on visual inspection. Background Art

[0002] Food safety is of utmost importance as it is related to everyone's physical health and life safety. To ensure food safety, relevant departments have introduced comprehensive food production safety management regulations to prevent unqualified food from entering the market.

[0003] A food production management system disclosed in the invention patent with the application number 202410432080.1 includes a raw material procurement unit, a production planning unit, a production control unit, a quality inspection unit, a warehousing unit, a sales and distribution unit, and a coordination unit. Among them, the raw material procurement unit, the production planning unit, the production control unit, the quality inspection unit, the warehousing unit, and the sales and distribution unit are all connected to the coordination unit through wireless signal data; and the raw material procurement unit, the production planning unit, the production control unit, the quality inspection unit, the warehousing unit, and the sales and distribution unit form a top-down management process in sequence, and the coordination unit is used to conduct overall coordination and control of each unit of the raw material procurement unit, the production planning unit, the production control unit, the quality inspection unit, the warehousing unit, and the sales and distribution unit.

[0004] This application aims to solve the problem that: "In an existing food production management system, during actual use, the implementation of centralized control and data sharing for each production unit is not ideal, affecting management efficiency and response speed. Moreover, in terms of sales management, a perfect sales network and distribution channels have not been established, making it inconvenient to quickly respond to market demands. At the same time, there is a lack of cross-departmental information communication mechanisms and collaboration processes, resulting in poor overall collaboration efficiency and problem-solving capabilities."

[0005] In the food industry, during the process of producing coffee, the roasting of coffee beans directly affects the subsequent grinding process of coffee beans. However, it is difficult to control the roasting state of coffee beans during the roasting process; therefore, a method for monitoring food production processes based on visual inspection is proposed. Summary of the Invention

[0006] In view of the above-mentioned drawbacks of the prior art, the present invention provides a method for monitoring food production processes based on visual inspection, which solves the technical problems proposed in the above background art.

[0007] To achieve the above objectives, the present invention is realized through the following technical solutions:

[0008] A method for monitoring food production processes based on visual inspection includes:

[0009] A vibration device is deployed on the outer wall of the baking station of the baking equipment, and a camera module is deployed inside the baking station of the baking equipment. The baking equipment runs continuously based on the set operation logic. During the continuous operation of the baking equipment, the vibration device and the camera module run in conjunction.

[0010] Collecting baking target image data based on the camera module, storing the collected baking target image data in the camera module, monitoring the number of baking target image data stored in the camera module in real time, and performing an evaluation operation of baking target baking progress when the number of baking target image data is not less than three groups;

[0011] Obtain the baking progress evaluation result of the baking target in real time, and when the baking progress of the baking target is not less than 85%, continuously execute the determination of whether the baking target has been completed;

[0012] When the determination result is yes, the baking target of the baking station in the baking equipment is transferred to the cooling station. When the determination result is no, the baking equipment is controlled to continue to operate until the determination result is yes.

[0013] Furthermore, at least two groups of vibration devices are arranged on the outer wall of the baking device, and the vibration devices are distributed on the outer wall of the baking device in an equidistant and surrounding shape;

[0014] The operation logic used by the baking equipment is customized by the equipment management end user, and the setting logic is subject to: baking cycle, baking temperature change curve, baking mixing cycle, and the vibration device and camera module are operated once after each baking mixing cycle.

[0015] Among them, the vibration equipment is deployed in the middle of the outer wall of the baking equipment. The vibration equipment is provided with an operation cycle. The operation cycle of the vibration equipment is after the baking and stirring cycle. Two adjacent baking cycles are separated by a group of baking and stirring cycles and a vibration equipment operation cycle. The camera module runs once before the start of each baking cycle to collect baking target image data.

[0016] Furthermore, after the baking target image data is acquired and before being stored in the camera module, the baking target image data is optimized and then stored.

[0017] The optimization processing logic of the baking target image data is expressed as:

[0018]

[0019] Where: P′ is the pixel value of the optimized image; P is the pixel value of the original image; α and β are contrast adjustment parameters; C is the actual contrast value; C 0 is the expected contrast value; a, η, ξ are constants; γ, ω, is the parameter for the periodic change of color; D is the color-related variable; δ and λ are the saturation adjustment parameters; S is the actual saturation value; S 0 is the desired saturation value;

[0020] wherein, the constant in the above formula is greater than zero, and the value of the color-related variable D follows: R, G, and B are the three color channel values of the image. Based on the above formula, each pixel in the baked target image data is used as the processing target to complete the output of the optimized image pixel value, and then the original baked target image data is optimized based on the optimized image pixel value.

[0021] Furthermore, before performing the baking progress evaluation operation on the baked target image data, the contour extraction is synchronously performed on each baked target image data based on the Canny edge detection algorithm to obtain the contour image of the baked target image data, and further, the complete contour image is extracted from the contour image of the baked target image data;

[0022] Among them, the contour image of the baked target image data contains the contour images of several baked targets. When extracting the complete contour image from the contour image of the baked target image data, it is identified whether each contour in the contour image is a closed contour, and all non-closed contours in the contour image are discarded based on the recognition result, and the remaining obtained is the complete contour image.

[0023] Furthermore, the evaluation operation of the baking progress of the baked target is:

[0024]

[0025] In the formula: Q is the baking progress of the baked target; h(max) last 、h(max) first are the areas enclosed by the largest contours in the set of contour images with the latest acquisition timestamp and the areas enclosed by the largest contours in the set of contour images with the earliest acquisition timestamp; h(min) last 、h(min) first are the areas enclosed by the smallest contours in the set of contour images with the latest acquisition timestamp and the areas enclosed by the smallest contours in the set of contour images with the earliest acquisition timestamp; χ is the influence factor;

[0026] Among them, the influence factor χ ∈ (0, 1], and the value of the influence factor χ follows that the higher the size uniformity of the baked target, the larger the value of the influence factor, and vice versa, the smaller the value of the influence factor. The size uniformity of the baked target is determined based on the size and shape of the baked target.

[0027] Furthermore, the period of the baking target baking progress evaluation operation is user-defined based on the management of the baking equipment. When the baking progress of the baking target is not less than 85%, the determination period of whether the baking target is completed is continuously executed, which is 0.0001% - 0.1% of the operation period of the baking target baking progress evaluation;

[0028] Among them, the continuous determination period of whether the baking target is completed is always not more than one second.

[0029] Furthermore, the determination logic of whether the baking target is completed is expressed as:

[0030]

[0031] In the formula: k is the baking progress of the baking target; n is the total number of pixels in the image; τ(last) i is the color value of the i-th pixel point in the last collected baking target image data; τ(first) i is the color value of the i-th pixel point in the earliest collected baking target image data; q is the adjustment factor; θ is the normalization factor;

[0032] Among them, the adjustment factor q takes a value of 1 or -1. When the numerator of is greater than the denominator, the adjustment factor q takes a value of -1. When the numerator of is less than the denominator, the adjustment factor q takes a value of 1.

[0033] represents the average of

[0034]

[0035]

[0036] When the above formula holds, it means that the baking target is completed, otherwise it means that the baking target is not completed.

[0036] Furthermore, when the baking equipment continues to run based on the determination result, the baking stirring period of the baking station of the baking equipment is controlled to be shortened, and the operation period of the baking target baking progress evaluation operation of the baking equipment is shortened;

[0036] Among them, the shortening amounts of the baking stirring period of the baking station of the baking equipment and the operation period of the baking target baking progress evaluation operation are proportional to the number of times of determination as "no" in the determination result of whether the baking is completed.

[0037] Furthermore, when the baking target in the baking station of the baking equipment is transferred to the cooling station, a baking station operation message is generated synchronously. The baking station operation message is transmitted to the control target of the baking equipment, and the baking station operation message is displayed on the control panel of the baking station for the users of the baking equipment management terminal to read;

[0038] Among them, the content of the baking station operation message includes: operation time, the continuous determination times of whether the baking target in the baking station has completed baking, and the baking target image data collected by the camera module for the last time before the baking target in the baking station is transferred to the cooling station.

[0039] Adopting the technical solution provided by the present invention, compared with the known public technology, it has the following beneficial effects:

[0040] The present invention provides a method for monitoring the food production process by visual inspection. During the execution of this method, by collecting the baking target image data in real time and using the baking target image data as the support for monitoring and judging the baking progress, the baking dynamic changes of the baking target in the baking equipment can be monitored in real time. Furthermore, based on this, the intelligent control of baking the baking target to the best state is realized, ensuring that after the baking target is baked, a semi-finished product with better baking quality is obtained, so as to improve the quality of the final produced product. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0042] Figure 1 It is a schematic flow chart of a method for monitoring the food production process based on visual inspection. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0043] In order to make the objectives, technical solutions and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are some but not all of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0044] The following further describes the present invention with reference to the embodiments.

[0045] Embodiment 1:

[0046] The food production process monitoring method based on visual detection in this embodiment is as follows: Figure 1 As shown, including:

[0047] A vibration device is deployed on the outer wall of the baking station of the baking equipment, and a camera module is deployed inside the baking station of the baking equipment. The baking equipment runs continuously based on the set operation logic. During the continuous operation of the baking equipment, the vibration device and the camera module run in conjunction.

[0048] Collecting baking target image data based on the camera module, storing the collected baking target image data in the camera module, monitoring the number of baking target image data stored in the camera module in real time, and performing an evaluation operation of baking target baking progress when the number of baking target image data is not less than three groups;

[0049] Obtain the baking progress evaluation result of the baking target in real time, and when the baking progress of the baking target is not less than 85%, continuously execute the determination of whether the baking target has been completed;

[0050] When the determination result is yes, the baking target of the baking station in the baking equipment is transferred to the cooling station, and when the determination result is no, the baking equipment is controlled to continue to operate until the determination result is;

[0051] There are no less than two groups of vibration devices deployed on the outer wall of the baking equipment, and the vibration devices are distributed on the outer wall of the baking equipment in an equidistant and surrounding manner;

[0052] The operation logic used by the baking equipment is customized by the user of the equipment management end. The setting logic is subject to: baking cycle, baking temperature change curve, baking mixing cycle, and the vibration equipment and camera module are operated once after each baking mixing cycle.

[0053] Among them, the vibration device is deployed in the middle of the outer wall of the baking device. The vibration device is set with an operation cycle. The operation cycle of the vibration device is after the baking and stirring cycle. There is a baking and stirring cycle and a vibration device operation cycle between two adjacent baking cycles. The camera module runs once before each baking cycle to collect baking target image data;

[0054] The decision logic of whether the baking target has completed baking is expressed as:

[0055]

[0056] Where: k is the baking target baking progress; n is the total number of pixels in the image; τ(last) i The color value of the i-th pixel in the last collected baking target image data; τ(first) iis the color value of the i-th pixel point in the earliest collected and baked target image data; q is an adjustment factor; θ is a normalization factor;

[0057] Among them, the adjustment factor q takes the value of 1 or -1. When the numerator of is greater than the denominator, the adjustment factor q takes the value of -1. When the numerator of is less than the denominator, the adjustment factor q takes the value of 1.

[0058] denotes the averaging of The above equation holds indicating that the baking target is baked, otherwise it indicates that the baking target is not baked.

[0059] The normalization factor θ > 1 and is user-defined by the baking equipment management terminal. The normalization factor θ controls

[0060] the value of

[0061] to be within the range of 0 to 1. There are several groups of normalization factors, and several groups of normalization factors are bound to the baking target volume and moisture content of the specified baking equipment. Each normalization factor is selected and applied based on the initial volume and moisture content of the baking equipment for baking the baking target.

[0062] Among them, the initial volume and moisture content of the baking target are prior data, which are manually detected and obtained by the user of the baking equipment management terminal. Figure 1 In this embodiment, based on the above method, the method for monitoring the food production process based on visual detection in Embodiment 1 is further specifically described with reference to

[0063] After the baking target image data is collected and before it is stored inside the camera module, the baking target image data optimization process is synchronously executed, and then the storage of the baking target image data is executed.

[0064] The optimization processing logic of the baking target image data is expressed as:

[0065]

[0066] In the formula: P′ is the optimized image pixel value; P is the original image pixel value; α, β are contrast adjustment parameters; C is the actual contrast value; C 0 is the desired contrast value; a, η, ξ are constants; γ, ω, is the color periodic change parameter; D is the color-related variable; δ, λ are the saturation adjustment parameters; S is the actual saturation value; S 0 is the desired saturation value;

[0067] Among them, the constant in the above formula is greater than zero, and the value of the color-related variable D follows: R, G, B are the three color channel values of the image. Based on the above formula, each pixel in the baked target image data is used as the processing target to complete the output of the optimized image pixel value, and then the original baked target image data is optimized based on the optimized image pixel value.

[0068] Through the above formula, further optimization processing is performed on the baked target image data, providing more accurate data support for the further operation of the method in Embodiment 1, thereby ensuring that the output result of the method in Embodiment 1 is more accurate.

[0069] As Figure 1 shown, before performing the baking progress evaluation operation on the baked target image data, contour extraction is synchronously performed on each baked target image data based on the Canny edge detection algorithm to obtain the contour image of the baked target image data, and further the complete contour image is extracted from the contour image of the baked target image data;

[0070] Among them, the contour image of the baked target image data contains the contour images of several baked targets. When extracting the complete contour image from the contour image of the baked target image data, each contour in the contour image is identified whether it is a closed contour, and all non-closed contours in the contour image are discarded based on the recognition result, and the remaining obtained is the complete contour image;

[0071] The evaluation operation of the baking progress of the baked target is:

[0072]

[0073] In the formula: Q is the baking progress of the baked target; h(max) last 、h(max) first are the areas enclosed by the largest contours in the set of contour images with the latest acquisition timestamp and the areas enclosed by the largest contours in the set of contour images with the earliest acquisition timestamp; h(min) last 、h(min) first are the areas enclosed by the smallest contours in the set of contour images with the latest acquisition timestamp and the areas enclosed by the smallest contours in the set of contour images with the earliest acquisition timestamp; χ is the influencing factor;

[0074] Among them, the influence factor χ ∈ (0, 1]. The value of the influence factor χ follows the rule that the higher the size uniformity of the baking target, the larger the value of the influence factor; conversely, the smaller the value of the influence factor. The size uniformity of the baking target is determined based on the size and shape of the baking target.

[0075] The above settings provide further data support for the steps in Example 1 and further provide a specified evaluation logic for the baking progress evaluation of the baking target.

[0076] Example 3:

[0077] At the specific implementation level, based on Example 1, this example refers to Figure 1 to further specifically describe the method for monitoring the food production process based on visual detection in Example 1:

[0078] The period of the baking progress evaluation operation of the baking target is user-defined based on the management of the baking equipment. When the baking progress of the baking target is not less than 85%, the determination period of whether the baking target is completed is continuously executed, which is 0.0001% - 0.1% of the operation period of the baking progress evaluation of the baking target;

[0079] Among them, the continuous determination period of whether the baking target is completed is always no more than one second;

[0080] When the baking equipment continues to run based on the determination result, the baking stirring period control of the baking station of the baking equipment is shortened, and the operation period of the baking progress evaluation of the baking target of the baking equipment is shortened;

[0081] Among them, the reduction amounts of the baking stirring period of the baking station of the baking equipment and the operation period of the baking progress evaluation of the baking target are proportional to the number of times determined as no in the determination result of whether the baking is completed;

[0082] When the baking target in the baking station of the baking equipment is transferred to the cooling station, a baking station operation message is generated synchronously. The baking station operation message is transmitted to the control target of the baking equipment, and the baking station operation message is displayed on the control panel of the baking station for the user of the baking equipment management terminal to read;

[0083] Among them, the content of the baking station operation message includes: operation time, the continuous determination times of whether the baking target in the baking station is completed, and the baking target image data collected by the camera module for the last time before the baking target in the baking station is transferred to the cooling station.

[0084] In this example, through the above settings, further data support for the method in Example 1 is provided.

[0085] In summary, during the execution of the method in the above embodiments, by collecting baking target image data in real time and using the baking target image data as the support for baking progress monitoring and determination, the dynamic changes of the baking target in the baking equipment can be monitored in real time. Furthermore, based on this, the intelligent control of baking the baking target to the optimal state is achieved, ensuring that after the baking target is baked, a semi-finished product with better baking quality is obtained, so as to improve the quality of the final produced product.

[0086] The above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments or equivalently replace some of the technical features. However, 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. A food production process monitoring method based on visual inspection, characterized in that: include: A vibration device is deployed on the outer wall of the baking station of the baking equipment, and a camera module is deployed inside the baking station of the baking equipment. The baking equipment runs continuously based on the set operation logic. During the continuous operation of the baking equipment, the vibration device and the camera module run in conjunction. Collecting baking target image data based on the camera module, storing the collected baking target image data in the camera module, monitoring the number of baking target image data stored in the camera module in real time, and performing an evaluation operation of baking target baking progress when the number of baking target image data is not less than three groups; Obtain the baking progress evaluation result of the baking target in real time, and when the baking progress of the baking target is not less than 85%, continuously execute the determination of whether the baking target has been completed; When the determination result is yes, the baking target of the baking station in the baking equipment is transferred to the cooling station, and when the determination result is no, the baking equipment is controlled to continue to operate until the determination result is yes; There are at least two groups of vibration devices deployed on the outer wall of the baking equipment, and the vibration devices are distributed on the outer wall of the baking equipment in an equidistant and surrounding shape; The operation logic used by the baking equipment is customized by the equipment management end user, and the setting logic is subject to: baking cycle, baking temperature change curve, baking mixing cycle, and the vibration device and camera module are operated once after each baking mixing cycle. The vibration device is deployed in the middle of the outer wall of the baking device. The vibration device is provided with an operation cycle. The operation cycle of the vibration device is after the baking and stirring cycle. Two groups of adjacent baking cycles are separated by a group of baking and stirring cycles and a vibration device operation cycle. The camera module operates once before each baking cycle starts to collect baking target image data. After the baking target image data is collected, before being stored in the camera module, the baking target image data is optimized and then stored. The optimization processing logic of the baking target image data is expressed as: Where: P′ is the pixel value of the optimized image; P is the pixel value of the original image; α and β are contrast adjustment parameters; C is the actual contrast value; C0 is the expected contrast value; a, η, ξ are constants; γ, ω, is the color periodic change parameter; D is the color related variable; δ and λ are the saturation adjustment parameters; S is the actual saturation value; S0 is the expected saturation value; Among them, the constant in the above formula is greater than zero, and the value of the color-related variable D obeys: R, G, and B are the three color channel values ​​of the image. Based on the above formula, each pixel in the baking target image data is used as the processing target, the output of the optimized image pixel value is completed, and then the original baking target image data is optimized based on the optimized image pixel value; Before performing the baking progress evaluation operation, the baking target image data synchronously performs contour extraction on each baking target image data based on the Canny edge detection algorithm to obtain a contour image of the baking target image data, and further extracts a complete contour image from the contour image of the baking target image data; The contour image of the baking target image data includes contour images of several baking targets. When extracting the complete contour image from the contour image of the baking target image data, each contour in the contour image is identified as a closed contour. All non-closed contours in the contour image are discarded based on the identification result, and the remaining one is the complete contour image. The evaluation operation of the baking progress of the baking target is: Where: Q is the baking target baking progress; h(max) last 、h(max) first h(min) is the size of the area encompassed by the largest contour in the group of contour images with the latest acquisition time stamp, and the size of the area encompassed by the largest contour in the group of contour images with the earliest acquisition time stamp; last 、h(min) first is the size of the area encompassed by the smallest contour in the group of contour images with the latest acquisition time stamp, and the size of the area encompassed by the smallest contour in the group of contour images with the earliest acquisition time stamp; χ is the influencing factor; Among them, the influence factor χ∈(0,1], the value of the influence factor χ follows, the higher the size symmetry of the baking target, the larger the value of the influence factor, and vice versa, the smaller the value of the influence factor, the size symmetry of the baking target is determined based on the size and shape of the baking target.

2. The food production process monitoring method based on visual detection according to claim 1 is characterized in that: The baking target baking progress evaluation operation cycle is based on the user definition of the baking equipment management. When the baking target baking progress is not less than 85%, the baking target baking progress evaluation operation cycle is continuously executed, which is 0.0001% to 0.1% of the baking target baking progress evaluation operation cycle. The continuous determination period of whether the baking target has been baked is always no longer than one second.

3. The food production process monitoring method based on visual detection according to claim 1 is characterized in that: The decision logic of whether the baking target has completed baking is expressed as: Where: k is the baking target baking progress; n is the total number of pixels in the image; τ(last) i The color value of the i-th pixel in the last collected baking target image data; τ(first) i is the color value of the i-th pixel in the earliest collected baking target image data; q is the adjustment factor; θ is the normalization factor; The adjustment factor q takes the value of 1 or -1. When the numerator is greater than the denominator, the adjustment factor q takes the value of -1. When the numerator is smaller than the denominator, the adjustment factor q takes the value of 1. Express If the above formula is true, it means that the baking target has been completed, otherwise it means that the baking target has not been completed.

4. The food production process monitoring method based on visual detection according to claim 3 is characterized in that: The normalization factor θ>1, and is defined by the user of the baking equipment management end, and the normalization factor θ controls The value of is in the range of 0 to 1, and the normalization factor θ is set in several groups, and the several groups of normalization factors are bound to the baking target volume and moisture content baked by the specified baking equipment, and each normalization factor is selected and applied based on the initial volume and moisture content of the baking target baked by the baking equipment; Among them, the initial volume and moisture content of the baking target are prior data, which are manually detected and obtained by the user of the baking equipment management end.

5. The food production process monitoring method based on visual detection according to claim 1 or 3, characterized in that: When the baking equipment continues to operate based on the determination result, the baking stirring cycle control of the baking station of the baking equipment is shortened, and the baking target baking progress evaluation operation cycle of the baking equipment is shortened; Among them, the shortening amount of the baking stirring cycle of the baking station of the baking equipment and the baking target baking progress evaluation operation cycle is proportional to the number of times the baking is judged as no in the judgment result of whether the baking is completed.

6. The method for monitoring food production process based on visual detection according to claim 1, characterized in that: When a baking target in a baking station of the baking equipment is transferred to a cooling station, a baking station operation message is synchronously generated, the baking station operation message is transmitted to a control target of the baking equipment, and the baking station operation message is displayed on a control panel of the baking station for a user of a baking equipment management end to read; The content of the baking station operation message includes: the operation time, the number of consecutive determinations of whether the baking station has completed baking of the baking target, and the baking target image data collected by the camera module for the last time before the baking target in the baking station is transferred to the cooling station.

Images