AI-based Intelligent Monitoring and Quality Improvement Method for the Processing of Fish Maw

Abstract

The present invention discloses an intelligent monitoring and quality improvement method for the processing of fish maw based on AI, which relates to the field of fish maw processing and includes: obtaining the optimal soaking time range corresponding to fish maws of different types, thicknesses, sizes and weights, and establishing an optimal soaking comparison table for fish maws; setting temperature, weight, thickness and image acquisition devices to obtain the data of the temperature, weight, thickness and image of each fish maw after cleaning; based on the edge detection algorithm, obtaining the surface contour of each fish maw after cleaning and obtaining the surface area of each fish maw; establishing a theoretical fish maw weight model to obtain the theoretical weight value of each fish maw after cleaning; establishing a weight evaluation stability coefficient to screen out the optimal soaking time for fish maws; setting a fish maw transfer and shunt device to shunt the fish maws. The advantages of the present invention are: effectively controlling the soaking time of fish maws, optimizing the soaking effect of fish maws, and further improving the quality of fish maws.

Description

Technical Field

[0001] The present invention relates to the field of fish maw processing, and specifically to an intelligent monitoring and quality improvement method for the fish maw processing process based on AI. Background Art

[0002] The main processes of large-scale fish maw processing include: raw material detection, cleaning, soaking, shaping, drying, grading, and packaging. The soaking process of fish maw is an important link affecting the quality of fish maw. On the one hand, if the soaking time of fish maw is too short, it will lead to insufficient water absorption, softening of the outer layer, while the inner layer may still be relatively hard, affecting the taste of fish maw. Moreover, a short soaking time is difficult to remove the fishy smell of fish maw itself, seriously affecting the quality of fish maw. On the other hand, if the soaking time of fish maw is too long, it is easy to make the fish maw overly soft, resulting in the loss of water-soluble nutrients in the fish maw or local deterioration of the fish maw, seriously affecting the quality of fish maw. Therefore, ensuring that fish maw is soaked within the optimal soaking time and effectively improving the quality of fish maw is of great significance.

[0003] Traditional fish maw soaking does not distinguish the thickness, size, and weight of fish maw, but soaks them uniformly, or soaks and presses a large number of fish maw by using a fish maw soaking device. However, whether it is the traditional technology or the fish maw soaking device, it is difficult to avoid the influence of soaking time on fish maw of different qualities, resulting in the reduction or deterioration of the quality of some fish maw, thus seriously affecting the overall quality of fish maw and the corporate image. Summary of the Invention

[0004] To solve the above technical problems, an intelligent monitoring and quality improvement method for the fish maw processing process based on AI is provided. This technical solution solves the problem in the above background art that it is difficult to avoid the influence of soaking time on fish maw of different qualities, resulting in the reduction or deterioration of the quality of some fish maw, thus seriously affecting the overall quality of fish maw and the corporate image.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0006] An intelligent monitoring and quality improvement method for the fish maw processing process based on AI, including:

[0007] Obtain the optimal soaking time range corresponding to fish maw of different types, thicknesses, sizes, and weights, and establish an optimal soaking comparison table for fish maw;

[0008] Set up acquisition devices for temperature, weight, thickness, and image, and obtain the data of temperature, weight, thickness, and image of each fish maw after cleaning;

[0009] Based on the full-view picture of fish maw and the edge detection algorithm, obtain the surface contour of each fish maw after cleaning, and obtain the surface area of each fish maw;

[0010] According to the weight calculation formula, establish a theoretical fish maw weight model to obtain the theoretical weight value of each fish maw after temperature correction after cleaning;

[0011] Based on the theoretical weight value of the fish maw after temperature correction, the actual measurement value and the target value, establish a weight evaluation stability coefficient to screen out the optimal soaking time of the fish maw;

[0012] According to the optimal soaking comparison table of the fish maw, set up a fish maw transfer and diversion device, and divert the fish maw according to the selected soaking time of the fish maw.

[0013] Preferably, the specific steps of obtaining the optimal soaking time range corresponding to fish maws of different types, thicknesses, sizes and weights and establishing an optimal soaking comparison table for fish maws include:

[0014] Classify the fish maw raw materials according to the type, thickness, size, weight of the fish maw and the historical data of the fish maw processing raw materials;

[0015] Obtain the optimal soaking time range corresponding to fish maws of different types, thicknesses, sizes and weights, and set the highest threshold and the lowest threshold for each time range according to the optimal soaking time range;

[0016] Establish an optimal soaking comparison table for fish maws according to the type, thickness, size and weight of the fish maw and the corresponding optimal soaking time range.

[0017] Preferably, the specific steps of setting the acquisition devices for temperature, weight, thickness and image and obtaining the data of the temperature, weight, thickness and image of each fish maw after cleaning include:

[0018] Set up a transport conveyor belt, and at the conveyor belt diversion port position, set the acquisition devices for temperature, weight, thickness and image;

[0019] The acquisition of temperature, weight and image is carried out through a temperature sensor, a weighing instrument and a high-definition camera;

[0020] The specific thickness acquisition method includes:

[0021] At a fixed position above the conveyor belt, set a precision vernier caliper that moves vertically;

[0022] Install a flat plate of a fixed size under the precision vernier caliper, and integrate a pressure sensor on the lower surface of the flat plate;

[0023] Set a pressure threshold. When the pressure sensor touches the surface of the fish maw, as the pressure of the extrusion between the two increases, when the pressure of the extrusion between the two reaches the pressure threshold, it means that the current is the best point for measuring the thickness of the fish maw, and obtain the fish maw thickness according to the relative displacement of the precision vernier caliper.

[0024] Preferably, the steps of obtaining the surface contour of each fish maw after cleaning and the surface area of each fish maw based on the overall picture of the fish maw and the edge detection algorithm specifically include:

[0025] Collect the overall picture of the fish maw at a fixed position on the conveyor belt through an image acquisition device, and save and process it according to the acquisition sequence;

[0026] After performing grayscale processing on the collected overall picture of the fish maw, convert it into a grayscale picture of the overall fish maw;

[0027] According to the filtering algorithm, filter out the noise values in the grayscale picture of the overall fish maw to optimize the quality of the grayscale picture of the overall fish maw;

[0028] Based on the overall picture of the fish maw, the edge detection algorithm, and the grayscale picture of the overall fish maw with noise values filtered out, obtain the surface contour of each fish maw after cleaning;

[0029] According to the number of pixels in the surface contour of each fish maw after cleaning, convert to the actual surface area of the fish maw surface.

[0030] Preferably, the steps of establishing a theoretical fish maw weight model and obtaining the temperature-corrected theoretical weight value of each fish maw after cleaning according to the weight calculation formula specifically include:

[0031] Determine the volume of the fish maw according to the actually measured surface area and thickness of the fish maw surface;

[0032] Obtain the density value of the fish maw according to the type of fish maw;

[0033] According to the weight calculation formula of volume and density, combine the volume and density value of the fish maw to obtain the initial theoretical weight value of the fish maw;

[0034] According to the temperature value detected by the temperature sensor, correct the initial theoretical weight value of the fish maw to obtain the temperature-corrected theoretical weight value of the fish maw;

[0035] Based on the above acquisition and optimization of the temperature-corrected theoretical weight value of the fish maw, establish a theoretical fish maw weight model and obtain the temperature-corrected theoretical weight value of each fish maw after cleaning.

[0036] Preferably, the steps of establishing a weight evaluation stability coefficient based on the temperature-corrected theoretical weight value, actual measurement value, and target value of the fish maw and screening out the optimal fish maw soaking time specifically include:

[0037] Based on the test experiment, obtain several groups of temperature-corrected theoretical weight values and actual measurement values of the fish maw;

[0038] By means of manual comprehensive detection of these groups of fish maw, obtain the target values of the weights of these groups of fish maw;

[0039] Based on the theoretically corrected weight value, actual measured value, and target value of the fish maw according to temperature, establish an objective function for the weight value of the fish maw;

[0040] According to the gradient descent algorithm, through iterative optimization, screen out the optimal objective function value, that is, the weight evaluation stability coefficient;

[0041] According to the optimal soaking comparison table of the fish maw, combined with the type, thickness, size, and weight of the collected fish maw, based on the weight evaluation stability coefficient, determine the optimal soaking time of the fish maw.

[0042] Preferably, the fish maw conveying and shunting device is set according to the optimal soaking comparison table of the fish maw, and the fish maw is shunted according to the screened soaking time of the fish maw, which specifically includes:

[0043] According to the optimal soaking comparison table of the fish maw, divide the area according to the optimal soaking time range of the fish maw, and divide several fish maw soaking treatment points;

[0044] According to the number of fish maw soaking treatment points, set the same number of fish maw conveying and shunting devices, and convey the corresponding fish maw to the fish maw soaking treatment points through the conveying device;

[0045] According to the screened optimal soaking time of the fish maw, shunt the fish maw through the conveying and shunting device.

[0046] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0047] By obtaining the optimal soaking time range corresponding to fish maws of different types, thicknesses, sizes, and weights, determining the corresponding relationship between them, providing a reference standard for subsequent fish maw shunting, and by setting temperature, weight, thickness, and image acquisition devices, obtaining the temperature, weight, thickness, and image data of each fish maw after cleaning. Secondly, based on the full-view picture of the fish maw, the edge detection algorithm, and the grayscale picture of the full-view of the fish maw with noise values filtered out, obtain the surface contour of each fish maw after cleaning, and according to the number of pixels in the surface contour, convert the actual surface area of the fish maw surface. Furthermore, according to the weight calculation formula of volume and density, obtain the initial theoretical weight value of the fish maw, and correct the initial theoretical weight value of the fish maw by detecting the change value of temperature, thereby establishing a theoretical fish maw weight model, obtaining the theoretically corrected weight value of each fish maw after cleaning. Finally, based on the theoretically corrected weight value, actual measured value, and target value of the fish maw according to temperature, establish a weight evaluation stability coefficient, screen out the optimal soaking time of the fish maw, and shunt the fish maw by setting a fish maw conveying and shunting device, thereby effectively controlling the soaking time of the fish maw, making the soaking effect of the fish maw reach the optimal, and further improving the quality of the fish maw. Description of the Drawings

[0048] Figure 1 Flowchart of an AI-based intelligent monitoring and quality improvement method for the fish maw processing process of the present invention;

[0049] Figure 2 Flowchart of obtaining the surface contour of each fish maw after cleaning and the surface area of each fish maw based on the overall picture of the fish maw and the edge detection algorithm of the present invention;

[0050] Figure 3 Flowchart of establishing a theoretical fish maw weight model according to the weight calculation formula and obtaining the theoretical weight value of each temperature-corrected fish maw after cleaning of the present invention;

[0051] Figure 4 Flowchart of establishing a weight evaluation stability coefficient based on the theoretical weight value, actual measurement value and target value of the temperature-corrected fish maw, and screening out the optimal fish maw soaking time of the present invention;

[0052] Figure 5 Structural diagram of the architecture of the electronic device proposed by the present invention;

[0053] Figure 6 Structural schematic diagram of the computer-readable storage medium proposed by the present invention. Detailed implementation manners

[0054] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments in the following description are only examples, and those skilled in the art can think of other obvious variations.

[0055] Referring to Figure 1 As shown, an AI-based intelligent monitoring and quality improvement method for the fish maw processing process includes:

[0056] Obtain the optimal soaking time range corresponding to fish maws of different types, thicknesses, sizes and weights, and establish an optimal soaking comparison table for fish maws;

[0057] Set the acquisition devices for temperature, weight, thickness and image, and obtain the data of temperature, weight, thickness and image of each fish maw after cleaning;

[0058] Based on the overall picture of the fish maw and the edge detection algorithm, obtain the surface contour of each fish maw after cleaning, and obtain the surface area of each fish maw;

[0059] According to the weight calculation formula, establish a theoretical fish maw weight model, and obtain the theoretical weight value of each temperature-corrected fish maw after cleaning;

[0060] According to the theoretical weight value, actual measurement value and target value of the temperature-corrected fish maw, establish a weight evaluation stability coefficient, and screen out the optimal fish maw soaking time;

[0061] According to the optimal soaking comparison table of fish maw, set up a fish maw conveying and shunting device, and shunt the fish maw according to the screened soaking time of the fish maw.

[0062] It can be explained that in this solution, by obtaining the optimal soaking time ranges corresponding to fish maws of different types, thicknesses, sizes and weights, the corresponding relationship is determined, providing a reference standard for subsequent fish maw shunting. And by setting up acquisition devices for temperature, weight, thickness and images, the data of temperature, weight, thickness and images of each fish maw after cleaning are obtained. Secondly, based on the full-view picture of the fish maw, the edge detection algorithm and the grayscale picture of the full-view of the fish maw with noise values filtered out, the surface contour of each fish maw after cleaning is obtained, and according to the number of pixels in the surface contour, the actual surface area of the fish maw surface is converted. Furthermore, according to the weight calculation formula of volume and density, the initial theoretical weight value of the fish maw is obtained, and the initial theoretical weight value of the fish maw is corrected by temperature and conveying speed, so as to establish a theoretical fish maw weight model and obtain the corrected theoretical weight value of each fish maw after temperature correction after cleaning. Finally, according to the corrected theoretical weight value of the fish maw after temperature correction, the actual measurement value and the target value, a weight evaluation stability coefficient is established, the optimal soaking time of the fish maw is screened out, and by setting up a fish maw conveying and shunting device, the fish maw is shunted, so as to effectively control the soaking time of the fish maw, optimize the soaking effect of the fish maw, and further improve the quality of the fish maw.

[0063] Refer to Figure 2 As shown, the method for obtaining the surface contour of each fish maw after cleaning and the surface area of each fish maw based on the full-view picture of the fish maw and the edge detection algorithm specifically includes:

[0064] Through the image acquisition device, acquire the full-view picture of the fish maw at a fixed position on the conveyor belt, and save and process it according to the acquisition sequence;

[0065] After performing grayscale processing on the acquired full-view picture of the fish maw, convert it into a grayscale picture of the full-view of the fish maw;

[0066] According to the filtering algorithm, filter out the noise values in the grayscale picture of the full-view of the fish maw to optimize the quality of the grayscale picture of the full-view of the fish maw;

[0067] Based on the full-view picture of the fish maw, the edge detection algorithm and the grayscale picture of the full-view of the fish maw with noise values filtered out, obtain the surface contour of each fish maw after cleaning;

[0068] According to the number of pixels in the surface contour of each fish maw after cleaning, convert the actual surface area of the fish maw surface.

[0069] It can be explained that the edge detection algorithm can effectively identify the external contour of an object, especially the shape of the cleaned fish maw placed separately. Since the fish maw has a certain width, it has an obvious external contour against the background color of the conveyor belt. Among them,

[0070] The edge detection algorithm uses the Canny operator's edge detection algorithm. The specific process is as follows:

[0071] Based on the filtered grayscale full picture of the fish maw, calculate the gradient value and gradient direction of the first derivative of the full picture of the fish maw;

[0072] Based on the gradient value and gradient direction, use non-maximum suppression to eliminate some non-edge points;

[0073] According to the double-threshold method, set a high threshold and a low threshold to reduce false edge points in the full picture of the fish maw and optimize edge closure;

[0074] Finally, through hysteresis threshold processing, connect the weak edges and strong edges to form a complete edge;

[0075] The expression for converting the actual surface area of the fish maw based on the number of pixels in the surface contour is:

[0076]

[0077] In the formula, is the actual surface area of the fish maw surface, is the conversion ratio between the pixel area and the actual surface area of the fish maw, is the pixel area of the th pixel point,

[0078] is the number of pixel points in the pixel area of the fish maw. Figure 3 As shown in

[0079] Based on the measured actual surface area and thickness of the fish maw, determine the volume of the fish maw;

[0080] According to the type of fish maw, obtain the density value of the fish maw;

[0081] According to the weight calculation formula of volume and density, combined with the volume and density value of the fish maw, obtain the initial theoretical weight value of the fish maw;

[0082] According to the temperature value detected by the temperature sensor, correct the initial theoretical weight value of the fish maw to obtain the temperature-corrected theoretical weight value of the fish maw;

[0083] Obtain and optimize the theoretical weight value of fish maw after comprehensive temperature correction, establish a theoretical fish maw weight model, and obtain the theoretical weight value of each fish maw after temperature correction after cleaning;

[0084] The expression of the weight calculation formula for volume and density is:

[0085]

[0086] In the formula, is the initial theoretical weight value of the fish maw, is the density value of the fish maw, is the actual surface area of the fish maw, is the thickness of the fish maw, is the acceleration due to gravity;

[0087] The correction of the obtained initial theoretical weight value of the fish maw is:

[0088]

[0089] In the formula, is the theoretical weight value of the fish maw after temperature correction, is the thermal expansion coefficient of the fish maw, is the temperature change.

[0090] It can be explained that since the theoretical weight value of the fish maw after temperature correction is obtained by calculating the weight formula of volume and density, different temperatures will affect the density and volume of the fish maw. Therefore, when obtaining the theoretical weight value of the fish maw after temperature correction, the influence brought by temperature needs to be considered.

[0091] Refer to Figure 4 As shown, the establishment of a weight evaluation stability coefficient based on the theoretical weight value of the fish maw after temperature correction, the actual measurement value and the target value, and the selection of the optimal fish maw soaking time specifically include:

[0092] Based on the test experiment, obtain several groups of theoretical weight values and actual measurement values of the fish maw after temperature correction;

[0093] By adopting the method of manually comprehensively detecting these groups of fish maw, obtain the target values of the weights of these groups of fish maw;

[0094] According to the theoretical weight value of the fish maw after temperature correction, the actual measurement value and the target value, establish an objective function for the fish maw weight value;

[0095] According to the gradient descent algorithm, through iterative optimization, screen out the optimal objective function value, that is, the weight evaluation stability coefficient;

[0096] According to the optimal soaking comparison table of fish maw, combined with the type, thickness, size and weight of the collected fish maw, and based on the weight evaluation stability coefficient, determine the optimal soaking time of the fish maw;

[0097] The objective function of the fish maw weight value is:

[0098]

[0099] In the formula, is the deviation degree value among the theoretical weight value, actual measurement value and target value of the fish maw after temperature correction, is the th theoretical weight value of the fish maw after temperature correction, is the th actual measurement value of the fish maw, is the th target value of the fish maw, , are the weights of the theoretical weight value and actual measurement value of the fish maw after temperature correction respectively, used to balance the importance of the two, is the number of groups of the theoretical weight value, actual measurement value and target value of the fish maw after temperature correction;

[0100] The formula for determining the weight evaluation stability coefficient according to the gradient descent algorithm is:

[0101]

[0102] In the formula, is the weight value of the th weight at the th iteration, is the weight value of the th weight at the th iteration, where, is the weight evaluation stability coefficient value.

[0103] It can be explained that since there are certain errors in the gravity values collected by the gravity collection device and the theoretical weight values of the fish maw after temperature correction, therefore, it is necessary to obtain the target values of the weights of these groups of fish maw by means of comprehensive detection of these groups of fish maw manually, and find the stability coefficient value among the theoretical weight value, actual measurement value and target value of the fish maw after temperature correction, so as to obtain the accurate and optimal soaking time of the fish maw.

[0104] Furthermore, the method according to the embodiment of the present application can also be implemented by means of the architecture of the electronic device shown in Figure 5 As shown in Figure 5As shown, the electronic device 500 may include a bus 501, one or more CPUs 502, a read-only memory (ROM) 503, a random access memory (RAM) 504, a communication port 505 connected to a network, an input / output component 506, a hard disk 507, etc. The storage device in the electronic device 500, such as the ROM 503 or the hard disk 507, may store an AI-based intelligent monitoring and quality improvement method for the fish maw processing process provided in this application. The electronic device 500 may also include a user interface 508. Of course, Figure 5 The architecture shown is only exemplary. When implementing different devices, one or more components in the electronic device shown may be omitted according to actual needs. Figure 5 shown in the electronic device.

[0105] Figure 6 is a schematic diagram of the structure of a computer-readable storage medium provided by an embodiment of the present application. As Figure 6 shown, it is a computer-readable storage medium 600 according to an embodiment of the present application. Computer-readable instructions are stored on the computer-readable storage medium 600. When the computer-readable instructions are run by a processor, an AI-based intelligent monitoring and quality improvement method for the fish maw processing process according to the embodiment of the present application described with reference to the above drawings can be executed. The storage medium 600 includes, but is not limited to, for example, volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and cache memory, etc. Non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc.

[0106] In summary, the advantages of the present invention are as follows: effectively controlling the soaking time of the fish maw, optimizing the soaking effect of the fish maw, and further improving the quality of the fish maw.

[0107] The basic principles, main features and advantages of the present invention have been shown and described above. Those skilled in the art of this industry should understand that the present invention is not limited by the above embodiments. What is described in the above embodiments and the specification is only the principle of the present invention. Without departing from the spirit and scope of the present invention, the present invention will have various changes and improvements, and these changes and improvements all fall within the scope of the present invention claimed. The scope of protection claimed by the present invention is defined by the appended claims and their equivalents.

Claims

1. An AI-based intelligent monitoring and quality improvement method for the processing of fish maw, characterized in that, including: Obtain the optimal soaking time range corresponding to fish maws of different types, thicknesses, sizes and weights, and establish an optimal soaking comparison table for fish maws; Set up acquisition devices for temperature, weight, thickness and images, and obtain the data of temperature, weight, thickness and images of each fish maw after cleaning; Based on the overall picture of the fish maw and the edge detection algorithm, obtain the surface contour of each fish maw after cleaning, and obtain the surface area of each fish maw; According to the weight calculation formula, establish a theoretical fish maw weight model, and obtain the theoretical weight value of each temperature-corrected fish maw after cleaning; According to the theoretical weight value of the temperature-corrected fish maw, the actual measurement value and the target value, establish a weight evaluation stability coefficient, and screen out the optimal soaking time of the fish maw; According to the optimal soaking comparison table of fish maws, set up a fish maw transmission and diversion device, and divert the fish maws according to the selected soaking time of the fish maws; The specific steps of establishing a theoretical fish maw weight model according to the weight calculation formula and obtaining the theoretical weight value of each temperature-corrected fish maw after cleaning include: Correct the obtained initial theoretical weight value of the fish maw to obtain the theoretical weight value of the temperature-corrected fish maw; The correction of the obtained initial theoretical weight value of the fish maw is: ； In the formula, is the theoretical weight value of fish maw after temperature correction, is the coefficient of thermal expansion of fish maw, is the initial theoretical weight value of fish maw, is the temperature change; The specific steps of establishing a weight evaluation stability coefficient according to the theoretical weight value of the temperature-corrected fish maw, the actual measurement value and the target value, and screening out the optimal soaking time of the fish maw include: According to the theoretical weight value of the temperature-corrected fish maw, the actual measurement value and the target value, establish an objective function of the fish maw weight value; According to the gradient descent algorithm, through iterative optimization, screen out the optimal objective function value, that is, the weight evaluation stability coefficient; According to the optimal soaking comparison table of fish maws, combined with the type, thickness, size and weight of the collected fish maws, based on the weight evaluation stability coefficient, determine the optimal soaking time of the fish maws; The objective function of the fish maw weight value is: ； In the formula, is the deviation degree value among the theoretical weight value, actual measurement value, and target value of the fish maw after temperature correction, is the theoretical weight value of the th fish maw after temperature correction, is the actual measurement value of the th fish maw, is the target value of the th fish maw, , are the weights of the theoretical weight value and actual measurement value of the fish maw after temperature correction respectively, used to balance the importance of the two, is the number of groups of the theoretical weight value, actual measurement value, and target value of the fish maw after temperature correction.

2. The intelligent monitoring and quality improvement method for the processing of fish maw based on AI according to claim 1, characterized in that, The specific steps of obtaining the optimal soaking time range corresponding to fish maws of different types, thicknesses, sizes and weights, and establishing an optimal soaking comparison table for fish maws include: Classify the fish maw raw materials according to the type, thickness, size, weight of the fish maw and the historical data of the fish maw processing raw materials; Obtain the optimal soaking time range corresponding to fish maws of different types, thicknesses, sizes and weights, and set the maximum threshold and minimum threshold for each time range according to the optimal soaking time range; According to the type, thickness, size and weight of the fish maw and the corresponding optimal soaking time range, establish an optimal soaking comparison table for fish maws.

3. The intelligent monitoring and quality improvement method for the processing of fish maw based on AI according to claim 2, characterized in that, The specific steps of setting up acquisition devices for temperature, weight, thickness and images, and obtaining the data of temperature, weight, thickness and images of each fish maw after cleaning include: Set up a transport conveyor belt, and at the conveyor belt diversion port position, set up acquisition devices for temperature, weight, thickness and images; The acquisition of temperature, weight and images is carried out through a temperature sensor, a weighing instrument and a high-definition camera; The specific thickness acquisition method includes: At a fixed position above the conveyor belt, set up a precisely calibrated vernier caliper that moves vertically; Install a fixed-size flat plate under the precisely calibrated vernier caliper, and integrate a pressure sensor on the lower surface of the flat plate; Set a pressure threshold. When the pressure sensor touches the surface of the fish maw, as the pressure of the extrusion between the two increases, when the pressure of the extrusion between the two reaches the pressure threshold, it indicates that the current is the best point for measuring the thickness of the fish maw, and the thickness of the fish maw is obtained according to the relative displacement of the precision scale caliper.

4. An intelligent monitoring and quality improvement method for the processing of fish maw based on AI according to claim 3, characterized in that, The method for obtaining the surface contour of each fish maw after cleaning and the surface area of each fish maw based on the overall picture of the fish maw and the edge detection algorithm specifically includes: Collect the overall picture of the fish maw at a fixed position on the conveyor belt through an image acquisition device, and save and process it according to the acquisition sequence; After grayscale processing the collected overall picture of the fish maw, it is converted into a grayscale picture of the overall fish maw; According to the filtering algorithm, filter out the noise values in the grayscale picture of the overall fish maw to optimize the quality of the grayscale picture of the overall fish maw; Based on the overall picture of the fish maw, the edge detection algorithm, and the grayscale picture of the overall fish maw with noise values filtered out, obtain the surface contour of each fish maw after cleaning; According to the number of pixels in the surface contour of each fish maw after cleaning, convert to the actual surface area of the fish maw surface.

5. An intelligent monitoring and quality improvement method for the processing of fish maw based on AI according to claim 4, characterized in that, The method for establishing a theoretical fish maw weight model according to the weight calculation formula and obtaining the theoretical weight value of each temperature-corrected fish maw after cleaning specifically includes: Determine the volume of the fish maw according to the actual surface area and thickness of the fish maw surface measured; Obtain the density value of the fish maw according to the type of fish maw; According to the weight calculation formula of volume and density, combine the volume and density value of the fish maw to obtain the initial theoretical weight value of the fish maw; According to the temperature value detected by the temperature sensor, correct the initial theoretical weight value of the fish maw to obtain the theoretical weight value of the temperature-corrected fish maw; Based on the acquisition and optimization of the theoretical weight value of the temperature-corrected fish maw above, establish a theoretical fish maw weight model, and obtain the theoretical weight value of each temperature-corrected fish maw after cleaning; The expression of the weight calculation formula of volume and density is: ； In the formula, is the initial theoretical weight value of fish maw, is the density value of fish maw, is the actual surface area of the fish maw, is the thickness of the fish maw, is the acceleration due to gravity.

6. The intelligent monitoring and quality improvement method for the processing of fish maw based on AI according to claim 5, wherein, The method for establishing a weight evaluation stability coefficient according to the theoretical weight value of the temperature-corrected fish maw, the actual measurement value, and the target value, and screening out the optimal fish maw soaking time further includes: Based on the test experiment, obtain several groups of theoretical weight values and actual measurement values of the temperature-corrected fish maw; By means of manual comprehensive detection of these groups of fish maw, obtain the target values of the weights of these groups of fish maw.

7. An intelligent monitoring and quality improvement method for the processing of fish maw based on AI according to claim 6, characterized in that, The method for setting a fish maw transfer and diversion device according to the optimal fish maw soaking comparison table and diverting the fish maw according to the selected fish maw soaking time specifically includes: According to the optimal fish maw soaking comparison table, divide into zones according to the optimal fish maw soaking time range, and divide into several fish maw soaking treatment points; According to the number of fish maw soaking treatment points, set the same number of fish maw transfer and diversion devices, and transfer the corresponding fish maw to the fish maw soaking treatment points through the transfer device; According to the selected optimal fish maw soaking time, divert the fish maw through the transfer and diversion device.

8. An electronic device, characterized in that, Including: At least one processor; And, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute an AI-based intelligent monitoring and quality improvement method for fish maw processing as described in any one of claims 1-7.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements an AI-based intelligent monitoring and quality improvement method for fish maw processing as described in any one of claims 1-7.

Images