Artificial Intelligence-based Kelp Quality Sorting Color Selection Management System and Method

Abstract

The present invention relates to the technical field of kelp quality management, and discloses an artificial intelligence-based kelp quality sorting and color sorting management system and method. The system includes a kelp product color sorting information processing module, a kelp product color and belt quality grade evaluation module, and a kelp product quality evaluation and sorting execution module; by presetting standard data of kelp product quality grades corresponding to different appearance colors and belt quality grades, and combining with the quality grade parameters of the appearance color and belt of the kelp product, a comprehensive judgment of the kelp product quality grade is carried out, and the kelp product quality grade parameters are efficiently and accurately constructed to realize the intelligent evaluation of the kelp product quality grade; based on the kelp product quality grade parameters, the kelp product quality classification and sorting operation is efficiently and intelligently executed, improving the efficiency and accuracy of the kelp product quality sorting.

Description

Technical Field

[0001] The present invention relates to the technical field of kelp quality management, and specifically to an artificial intelligence-based kelp quality sorting and color sorting management system and method. Background Art

[0002] During the harvesting process of kelp products, manual sorting and color sorting of kelp products according to the appearance color and belt body defect characteristics of kelp products are required. However, manual sorting and color sorting of kelp products not only have low efficiency but also have a high error rate in the quality sorting and color sorting of kelp products.

[0003] The Chinese patent application with the publication number CN117726278A discloses an agricultural product flow management method and system. By obtaining the agricultural product information and supplier information of bulk agricultural products, determining the inspection plan, controlling the inspection device based on the inspection plan to obtain the inspection data of the storage unit, determining the quality evaluation of the storage unit based on the inspection data, and sorting the bulk agricultural products based on the quality evaluation; in response to the number of non-conforming products being less than the first threshold, increasing the conveyor belt transmission speed; in response to the number of agricultural products meeting the third preset condition being greater than the third threshold, sorting the agricultural products separately; the third preset condition is that the quality grade of the agricultural product is higher than the overall quality grade; and managing the flow of the sorted agricultural products according to different sorting grades to improve the quality management efficiency and accuracy of agricultural products; however, the above technical solutions cannot conduct refined evaluation and analysis on the product quality of each agricultural product, reducing the accuracy of agricultural product quality management. Summary of the Invention

[0004] (I) Technical Problems to be Solved

[0005] To solve the problem that manual sorting and color sorting of kelp products not only have low efficiency but also have a high error rate in the quality sorting and color sorting of kelp products, and achieve the purposes of accurately collecting the appearance feature images of kelp products, scientifically establishing the appearance sample images of kelp products, intelligently evaluating the appearance color quality grade of kelp products, precisely analyzing the belt body quality grade of kelp products, comprehensively evaluating the quality grade of kelp products, and efficiently performing the sorting and color sorting operations of kelp products.

[0006] (II) Technical Solutions

[0007] The present invention is realized through the following technical solutions: a kelp quality sorting and color sorting management method, the method comprising the following steps:

[0008] S1. Collect the appearance feature image data of kelp products and perform image preprocessing to generate standard kelp product appearance feature image data;

[0009] S2. Perform generation processing on the appearance image samples of kelp products for the standard kelp product appearance feature image data to generate kelp product sample appearance image data;

[0010] S3. Analyze and process the appearance color characteristics of the kelp product sample based on the appearance image data of the kelp product sample and the appearance color image data of the qualified kelp product to generate the appearance color analysis data of the kelp product sample;

[0011] S4. Perform a weighted value measurement process on the qualified appearance color samples of the kelp product based on the appearance color analysis data of the kelp product sample to generate the weighted data of the qualified appearance color samples of the kelp product, and conduct an appearance color quality grade analysis process with the weighted threshold of the qualified appearance color samples of the high-quality kelp product to construct the appearance color quality grade data of the kelp product;

[0012] S5. Identify the belt body defect characteristics of the kelp product sample based on the appearance image data of the kelp product sample and the belt body defect characteristic image data of the kelp product to generate the belt body defect identification data of the kelp product sample;

[0013] S6. Perform a weighted value measurement process on the samples without belt body defects of the kelp product based on the belt body defect identification data of the kelp product sample to generate the weighted data of the samples without belt body defects of the kelp product, and conduct a belt body quality grade analysis process with the weighted threshold of the samples without belt body defects of the high-quality kelp product to construct the belt body quality grade data of the kelp product;

[0014] S7. Based on the appearance color of the kelp product, the quality grade parameters of the belt body, and the quality grade data of the kelp product corresponding to different appearance colors and belt body quality grades, conduct a comprehensive judgment process on the quality grade of the kelp product to construct the quality grade data of the kelp product and perform the quality sorting operation of the kelp product.

[0015] Preferably, the operation steps of collecting the appearance feature image data of the kelp product and performing image preprocessing to generate the standard appearance feature image data of the kelp product are as follows:

[0016] S11. Online collect the appearance feature image of the kelp product passing through the monitoring port through the cloud lens installed on the material conveyor belt, and generate the appearance feature image data of the kelp product ;

[0017] S12. After performing image noise reduction preprocessing on the appearance feature image data of the kelp product using the adaptive filtering method, generate the standard appearance feature image data of the kelp product .

[0018] Preferably, the operation steps of performing the appearance image sample generation process on the standard appearance feature image data of the kelp product to generate the appearance image data of the kelp product sample are as follows:

[0019] S21. For the image data of the appearance characteristics of the standard kelp product Perform appearance characteristic image grid division processing using a square with side length L, and generate a set of appearance image data of kelp product samples , ; where represents the th appearance image data of a kelp product sample, represents the total number of kelp product sample appearances.

[0020] Preferably, the operation steps for generating the appearance color analysis data of the kelp product sample by performing appearance color characteristic analysis processing on the appearance image data of the kelp product sample and the appearance color image data of the qualified kelp product are as follows:

[0021] S31. Establish a set of appearance color image data of qualified kelp products , ; where represents the th appearance color image data of a qualified kelp product, represents the maximum value of the number of appearance color types of qualified kelp products. The appearance color types of qualified kelp products include brownish green, earthy yellow, dark green, and dark brown;

[0022] S32. Match the appearance image data in the set of appearance image data of the kelp product sample with the appearance color image data in the set of appearance color image data of the qualified kelp product in an orderly manner according to the numbering of the appearance image quantity of the kelp product sample, and generate a set of appearance color analysis data of the kelp product sample . The specific operation steps for generating the set of appearance color analysis data of the kelp product sample are as follows:

[0023] S321. Initialize the algorithm iteration times data and execute the ocean current simulation process; set the maximum algorithm iteration times data as T, and the ocean current simulation process is to search for ocean currents containing qualified kelp product appearance color image data matching the appearance image data of the kelp product sample in the search space of the set of appearance color image data of the qualified kelp product nutrients, and the appearance color search jellyfish is the qualified kelp product appearance color image data matching the appearance image data of the kelp product sample Nutrient attraction, the direction of the ocean current is determined by the average value of all vectors from each jellyfish searching for the appearance color in the ocean to the jellyfish searching for the appearance color at the optimal position;

[0024] S322. During the execution of the algorithm, the jellyfish searching for the appearance color in the jellyfish swarm searching for the appearance color move around their own positions in the search space to search for the qualified kelp product appearance color image data matching the appearance image data of the kelp product sample and update their positions. The corresponding position update formula for each jellyfish searching for the appearance color is as follows: where represents the position of the individual jellyfish searching for the appearance color after the th iteration in the search space of the qualified kelp product appearance color image data set , represents the position of the individual jellyfish searching for the appearance color after the th iteration in the search space of the qualified kelp product appearance color image data set represents a random function with a value in the interval; and represent the upper and lower bounds of the search space, that is, the upper and lower bounds of the data of the qualified kelp product appearance color image data that matches the appearance image data of the kelp product sample searched in the search space of the qualified kelp product appearance color image data set matching the appearance image data of the kelp product sample , =0.1 represents the motion system data;

[0025] S323. During the execution of the algorithm, the jellyfish searching for the appearance color in the jellyfish swarm searching for the appearance color search for the qualified kelp product appearance color image data matching the appearance image data of the kelp product sample in the search space of the qualified kelp product appearance color image data set The control time of the behavior is as follows: where represents the control time required for the individual jellyfish searching for the appearance color after the th iteration,

[0026] S324. The algorithm execution satisfies the boundary conditions. When the jellyfish searching for the appearance color moves out of the bounded qualified kelp product appearance color image data set When searching the search space, the appearance color search jellyfish returns to the set of appearance color image data of qualified kelp products Continue to search in the opposite search area in the search space to find the appearance image data of the kelp product sample The qualified kelp product appearance color image data that matches The moving boundary condition of the appearance color search jellyfish satisfies the following formula:

[0027] , where represents the individual of the appearance color search jellyfish in the updated position in the dimensional space, that is, in the qualified kelp product appearance color image data set with the spatial position dimension of The updated position in the search space in the set ; represents the individual of the appearance color search jellyfish in the position before the update in the dimensional space, that is, in the qualified kelp product appearance color image data set with the spatial position dimension of The position before the update in the search space in the set ;

[0028] S325. When the algorithm meets the maximum iteration data, output the appearance image data of the kelp product sample and the image feature matching result with the qualified kelp product appearance color image data to generate a set of appearance color analysis data of the kelp product sample , where represents the appearance color analysis data of the kelp product sample corresponding to the appearance image data of the kelp product sample ;

[0029] When and are successfully matched in image features, it means that the appearance color of the kelp product sample is qualified, and then output the appearance color analysis data of the kelp product sample as qualified;

[0030] When and are not successfully matched in image features, it means that the appearance color of the kelp product sample is unqualified, and then output the appearance color analysis data of the kelp product sample as unqualified.

[0031] Preferably, based on the analyzed data of the appearance color of the kelp product samples, perform a weighting value measurement process on the qualified appearance color samples of the kelp product to generate the weight data of the qualified appearance color samples of the kelp product, and perform an analysis process on the appearance color quality grade of the kelp product with the weight threshold of the qualified appearance color samples of the high-quality kelp product. The operation steps for constructing the appearance color quality grade data of the kelp product are as follows:

[0032] S41. Establish the weight threshold of the qualified appearance color samples of the high-quality kelp product , where the weight threshold of the qualified appearance color samples of the high-quality kelp product represents the minimum value of the weight of the qualified appearance color samples of the high-quality kelp product among all appearance samples;

[0033] S42. Use the breadth-first search algorithm to search in the set of analyzed data of the appearance color of the kelp product samples to find the number of kelp product samples whose analyzed data of the appearance color of the kelp product samples are qualified, and generate the number of qualified appearance color samples of the kelp product ;

[0034] S43. Perform a numerical division measurement process on the number of qualified appearance color samples of the kelp product and the total number of the kelp product samples to generate the weight data of the qualified appearance color samples of the kelp product , where ;

[0035] S44. Compare the weight data of the qualified appearance color samples of the kelp product with the weight threshold of the qualified appearance color samples of the high-quality kelp product to construct the appearance color quality grade data of the kelp product based on the numerical comparison result ;

[0036] When ≥ , it means that the appearance color of the kelp product meets the weight requirement of the qualified appearance color samples in the appearance color of the high-quality kelp product, and the appearance color quality grade data of the kelp product is output as grade one;

[0037] When < , it means that the appearance color of the kelp product does not meet the weight requirement of the qualified appearance color samples in the appearance color of the high-quality kelp product, and the appearance color quality grade data of the kelp product is output as grade two.

[0038] Preferably, the operation steps for identifying the belt body defect characteristics of the kelp product sample based on the appearance image data of the kelp product sample and the belt body defect characteristic image data of the kelp product are as follows:

[0039] S51. Establish a set of belt body defect characteristic image data of the kelp product , wherein represents the th type of belt body defect characteristic image data of the kelp product, represents the maximum value of the number of belt body defect types of the kelp product, and the belt body defect types of the kelp product include belt body scratches, belt body black spots, belt body holes, and belt body breaks;

[0040] S52. When the appearance color quality grade data of the kelp product is constructed, use the uniform cost search algorithm to match the appearance image data of the kelp product sample in the appearance image data set of the kelp product sample with the belt body defect characteristic image data in the belt body defect characteristic image data set of the kelp product in an orderly manner according to the numbering of the appearance image of the kelp product sample, and generate a set of belt body defect identification data of the kelp product sample based on the image feature matching result wherein represents the belt body defect identification data corresponding to the appearance image data of the kelp product sample ; When successfully matches the image features with

[0041] , it indicates that the kelp product sample has belt body defect characteristics, and the belt body defect identification data of the kelp product sample is output as existent; When

[0042] does not successfully match the image features with , it indicates that the kelp product sample does not have belt body defect characteristics, and the belt body defect identification data of the kelp product sample is output as non - existent.

[0043] Preferably, based on the belt body defect identification data of the kelp product sample, perform a weight value measurement process on the kelp product samples without belt body defects, generate weight data of the kelp product samples without belt body defects, and perform a belt body quality grade analysis process on the kelp product with the weight threshold of the high - quality kelp product samples without belt body defects, and the operation steps for constructing the belt body quality grade data of the kelp product are as follows:

[0044] S61. Establish a weight threshold for samples of high-quality kelp products without belt body defects , where the weight threshold for samples of high-quality kelp products without belt body defects represents the minimum value of the weight of samples without belt body defects among all appearance samples of high-quality kelp products;

[0045] S62. Use the breadth-first search algorithm to search in the data set for identifying belt body defects of kelp product samples to find the number of kelp product samples that do not exist in the data for identifying belt body defects of kelp product samples, and generate the number of kelp product samples without belt body defects ; ;

[0046] S63. Perform a numerical division operation on the number of kelp product samples without belt body defects and the total number of kelp product samples to generate weight data for kelp product samples without belt body defects , where ;

[0047] S64. Compare the weight data for kelp product samples without belt body defects with the weight threshold for samples of high-quality kelp products without belt body defects to construct quality grade data for the belt body of kelp products based on the numerical comparison result ;

[0048] When ≥ , it indicates that the belt body defects of the kelp product meet the weight requirements of samples without belt body defects in the belt body defects of high-quality kelp products, and then output the quality grade data for the belt body of the kelp product as grade one;

[0049] When < , it indicates that the belt body defects of the kelp product do not meet the weight requirements of samples without belt body defects in the belt body defects of high-quality kelp products, and then output the quality grade data for the belt body of the kelp product as grade two.

[0050] Preferably, the operation steps for comprehensively judging the quality grade of kelp products based on the appearance color of kelp products, the quality grade parameters of the belt body, and the quality grade data of kelp products corresponding to different appearance colors and belt body qualities, constructing the quality grade data of kelp products, and performing the quality sorting operation of kelp products are as follows:

[0051] S71. Establish a data set of quality grade data for kelp products corresponding to different appearance colors and belt body qualities , where Indicates that when the appearance color quality grade data of the kelp product and the belt body quality grade data of the kelp product are both of the first grade, the corresponding kelp product quality grade data for different appearance colors and belt body quality grades is , indicating that the kelp product quality grade is of the first grade; Indicates that when the appearance color quality grade data of the kelp product is of the first grade and the belt body quality grade data of the kelp product is of the second grade, or when the appearance color quality grade data of the kelp product is of the second grade and the belt body quality grade data of the kelp product is of the first grade, the corresponding kelp product quality grade data for different appearance colors and belt body quality grades is , indicating that the kelp product quality grade is of the second grade; Indicates that when the appearance color quality grade data of the kelp product and the belt body quality grade data of the kelp product are both of the second grade, the corresponding kelp product quality grade data for different appearance colors and belt body quality grades is , indicating that the kelp product quality grade is of the third grade;

[0052] S72. Match the appearance color quality grade data of the kelp product constructed in steps S44 and S64 and the belt body quality grade data of the kelp product with the kelp product quality grade data corresponding to different appearance colors and belt body quality grades in the set of kelp product quality grade data corresponding to different appearance colors and belt body quality grades, search for the kelp product quality grade data corresponding to the appearance color quality grade data of the kelp product and the belt body quality grade data of the kelp product , and construct the kelp product quality grade data through data identification ;

[0053] S73. Perform the quality sorting operation of the kelp product on the kelp product through the material conveyor belt according to the kelp product quality grade data .

[0054] An artificial intelligence-based kelp quality sorting and color sorting management system for implementing the kelp quality sorting and color sorting management method. The system includes a kelp product color sorting information processing module, a kelp product color and ribbon quality grade evaluation module, and a kelp product quality evaluation and sorting execution module;

[0055] The kelp product color sorting information processing module includes a kelp product appearance feature image acquisition unit, a kelp product appearance feature image preprocessing unit, and a kelp product sample appearance image generation unit;

[0056] The kelp product appearance feature image acquisition unit acquires kelp product appearance feature image data through a cloud lens; the kelp product appearance feature image preprocessing unit performs image preprocessing on the kelp product appearance feature image data to generate standard kelp product appearance feature image data; the kelp product sample appearance image generation unit performs kelp product appearance image sample generation processing on the standard kelp product appearance feature image data to generate kelp product sample appearance image data;

[0057] The kelp product color and ribbon quality grade evaluation module includes a qualified kelp product appearance color image storage unit, a kelp product sample appearance color analysis unit, a kelp product qualified appearance color sample weight measurement unit, a high-quality kelp product qualified appearance color sample weight threshold storage unit, and a kelp product appearance color quality grade analysis unit; a kelp product ribbon defect feature image storage unit, a kelp product sample ribbon defect identification unit, a kelp product non-existent ribbon defect sample weight measurement unit, a high-quality kelp product non-existent ribbon defect sample weight threshold storage unit, and a kelp product ribbon quality grade analysis unit;

[0058] The qualified kelp product appearance color image storage unit is used to store the qualified kelp product appearance color image data; the kelp product sample appearance color analysis unit analyzes and processes the appearance color characteristics of the kelp product sample based on artificial intelligence by comparing the kelp product sample appearance image data with the qualified kelp product appearance color image data, and generates the kelp product sample appearance color analysis data; the kelp product qualified appearance color sample weight measurement unit performs the weight value measurement process of the qualified kelp product appearance color sample based on the kelp product sample appearance color analysis data, and generates the kelp product qualified appearance color sample weight data; the high-quality kelp product qualified appearance color sample weight threshold storage unit is used to store the high-quality kelp product qualified appearance color sample weight threshold; the kelp product appearance color quality grade analysis unit performs the appearance color quality grade analysis process of the kelp product according to the kelp product qualified appearance color sample weight data and the high-quality kelp product qualified appearance color sample weight threshold, and constructs the kelp product appearance color quality grade data; the kelp product belt body defect feature image storage unit is used to store the kelp product belt body defect feature image data; the kelp product sample belt body defect recognition unit performs the belt body defect feature recognition process of the kelp product sample according to the kelp product sample appearance image data and the kelp product belt body defect feature image data, and generates the kelp product sample belt body defect recognition data; the kelp product non-belt-body-defect sample weight measurement unit performs the weight value measurement process of the kelp product non-belt-body-defect sample based on the kelp product sample belt body defect recognition data, and generates the kelp product non-belt-body-defect sample weight data; the high-quality kelp product non-belt-body-defect sample weight threshold storage unit is used to store the high-quality kelp product non-belt-body-defect sample weight threshold; the kelp product belt body quality grade analysis unit performs the belt body quality grade analysis process of the kelp product by comparing the kelp product non-belt-body-defect sample weight data with the high-quality kelp product non-belt-body-defect sample weight threshold, and constructs the kelp product belt body quality grade data;

[0059] The kelp product quality evaluation and sorting execution module includes a kelp product quality grade storage unit corresponding to different appearance colors and belt body quality grades, a kelp product quality grade comprehensive judgment unit, and a kelp product quality sorting operation execution unit;

[0060] The kelp product quality grade storage unit corresponding to different appearance colors and belt body mass grades is used to store the kelp product quality grade data corresponding to different appearance colors and belt body mass grades; the comprehensive judgment unit for the kelp product quality grade makes a comprehensive judgment and processing of the kelp product quality grade based on the appearance color of the kelp product, the quality grade parameters of the belt body, and the kelp product quality grade data corresponding to different appearance colors and belt body mass grades, and constructs the kelp product quality grade data; the kelp product quality sorting operation execution unit executes the kelp product quality sorting operation according to the kelp product quality grade data.

[0061] (III) Beneficial effects

[0062] The present invention provides an artificial intelligence-based kelp quality sorting and color selection management system and method. It has the following beneficial effects:

[0063] First, through the cooperation of the kelp product appearance feature image acquisition unit and the kelp product appearance feature image preprocessing unit, the cloud lens is used to dynamically collect the kelp product appearance feature image parameters and perform image noise reduction preprocessing, accurately generating the standard kelp product appearance feature image parameters, improving the authenticity of the kelp product appearance feature image acquisition; the kelp product sample appearance image generation unit performs image grid division processing on the kelp product appearance feature image parameters, scientifically constructs the image sample of the kelp product to be detected, and realizes the refined operation of the kelp product quality sorting and color selection.

[0064] Second, through the cooperation of the kelp product sample appearance color analysis unit and the kelp product sample belt body defect recognition unit, the artificial intelligence algorithm is used to perform intelligent and standardized analysis on the appearance color and belt body defects of the kelp product image sample, improving the accuracy and efficiency of the kelp product quality sorting and color selection; the kelp product qualified appearance color sample weight measurement unit and the kelp product non-belt body defect sample weight measurement unit cooperate with each other to digitally count the weights of the qualified appearance color samples and the samples without belt body defects in the kelp product image sample, realizing the precise quantification processing of the appearance color and belt body defect characteristics in the kelp product; the kelp product appearance color quality grade analysis unit and the kelp product belt body quality grade analysis unit cooperate with each other to compare the kelp product appearance color weight and belt body defect weight with the appearance color and belt body quality threshold values, scientifically analyzing the appearance color and belt body quality grades of the kelp product, and improving the scientific nature of the kelp quality sorting and color selection.

[0065] III. Through the cooperation of the storage unit for the quality grades of kelp products corresponding to different appearance colors and belt weights and the comprehensive judgment unit for the quality grades of kelp products, the data of the quality grades of kelp products corresponding to different appearance colors and belt weights preset by the standard are comprehensively judged with the appearance color and the quality grade parameters of the belt of the kelp products, and the quality grade parameters of the kelp products are efficiently and accurately constructed, realizing the intelligent evaluation of the quality grades of kelp products; the quality sorting operation execution unit for kelp products efficiently and intelligently executes the quality classification and sorting operation of kelp products based on the quality grade parameters of kelp products, improving the efficiency and accuracy of the quality sorting of kelp products. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] Figure 1 FIG. is a schematic diagram of the modules of the kelp quality sorting and color selection management system based on artificial intelligence provided by the present invention;

[0067] Figure 2 FIG. is a flowchart of the kelp quality sorting and color selection management method provided by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0068] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0069] Embodiments of the kelp quality sorting and color selection management system and method based on artificial intelligence are as follows:

[0070] Embodiment 1:

[0071] Please refer to Figure 1 - Figure 2 , the kelp quality sorting and color selection management method, the method includes the following steps:

[0072] S1. Collect the image data of the appearance characteristics of kelp products and perform image preprocessing to generate standard image data of the appearance characteristics of kelp products;

[0073] S2. Perform the generation processing of the appearance image samples of kelp products on the standard image data of the appearance characteristics of kelp products to generate the sample appearance image data of kelp products;

[0074] S3. Analyze the appearance color characteristics of the kelp product samples based on the sample appearance image data of the kelp products and the image data of the appearance colors of qualified kelp products to generate the analysis data of the appearance colors of the kelp product samples;

[0075] S4. Perform weighted numerical measurement processing on the qualified appearance color sample of kelp products based on the kelp product sample appearance color analysis data, generate the kelp product qualified appearance color sample weight data, and perform appearance color quality grade analysis processing on the kelp product with the weight threshold of the qualified appearance color sample of high-quality kelp products to construct the kelp product appearance color quality grade data;

[0076] S5. Perform belt defect feature recognition processing on the kelp product sample according to the kelp product sample appearance image data and the kelp product belt defect feature image data to generate the kelp product sample belt defect recognition data;

[0077] S6. Perform weighted numerical measurement processing on the kelp product sample without belt defects based on the kelp product sample belt defect recognition data, generate the kelp product sample weight data without belt defects, and perform belt quality grade analysis processing on the kelp product with the weight threshold of the kelp product sample without belt defects of high-quality kelp products to construct the kelp product belt quality grade data;

[0078] S7. Perform comprehensive quality grade judgment processing on the kelp product based on the quality grade parameters of the kelp product appearance color and belt, and the kelp product quality grade data corresponding to different appearance colors and belt quality grades, construct the kelp product quality grade data, and perform the kelp product quality sorting operation.

[0079] Further, please refer to Figure 1 - Figure 2 , the operation steps for collecting the kelp product appearance feature image data and performing image preprocessing to generate the standard kelp product appearance feature image data are as follows:

[0080] S11. Online collect the appearance feature image of the kelp product passing through the monitoring port through the cloud lens installed on the material conveyor belt, and generate the kelp product appearance feature image data ;

[0081] S12. After performing image noise reduction preprocessing on the kelp product appearance feature image data using the adaptive filtering method, generate the standard kelp product appearance feature image data .

[0082] The operation steps for generating the kelp product sample appearance image data by performing kelp product appearance image sample generation processing on the standard kelp product appearance feature image data are as follows:

[0083] S21. Perform appearance feature image grid division processing on the standard kelp product appearance feature image data using a square with side length L, and generate the kelp product sample appearance image data set , ; where Indicates the appearance image data of the th kelp product sample,

[0084] Through the cooperation of the kelp product appearance feature image acquisition unit and the kelp product appearance feature image preprocessing unit, the appearance feature image parameters of the kelp product are dynamically collected by the cloud lens and the image noise reduction preprocessing is carried out, and the standard kelp product appearance feature image parameters are accurately generated, improving the authenticity of the collection of the kelp product appearance feature images; the kelp product sample appearance image generation unit conducts image grid division processing on the kelp product appearance feature image parameters, scientifically constructs the image samples of the kelp products to be detected, and realizes the refined operation of the quality sorting and color sorting of the kelp products.

[0085] Further, please refer to Figure 1 - Figure 2 , and the operation steps for generating the appearance color analysis data of the kelp product sample based on the appearance image data of the kelp product sample and the appearance color image data of the qualified kelp product are as follows:

[0086] S31. Establish a collection of the appearance color image data of the qualified kelp product , ; where indicates the th kind of appearance color image data of the qualified kelp product, represents the maximum value of the number of appearance color types of the qualified kelp product, and the appearance color types of the qualified kelp product include brownish green, earthy yellow, dark green and dark brown;

[0087] S32. Match the appearance image data of the kelp product sample in the collection of the appearance image data of the kelp product sample with the appearance color image data of the qualified kelp product in the collection of the appearance color image data of the qualified kelp product in an orderly manner according to the numbering of the appearance image quantity of the kelp product sample, and generate a collection of the appearance color analysis data of the kelp product sample based on the image feature matching result. The specific operation steps for executing the generation of the collection of the appearance color analysis data of the kelp product sample are as follows: S321. Initialize the algorithm iteration times data and execute the ocean current simulation process; set the maximum algorithm iteration times data as T, and the ocean current simulation process is that there is an ocean current in the search space of the collection

[0088] of the appearance color image data of the qualified kelp product that contains the appearance image data of the kelp product sample in the collection Matched qualified kelp product appearance color image data Nutrients, the appearance color search jellyfish is captured by the appearance image data of the kelp product sample Matched qualified kelp product appearance color image data Nutrient attraction, the direction of the ocean current is determined by the average value of all vectors from each appearance color search jellyfish in the ocean to the appearance color search jellyfish located at the optimal position;

[0089] S322. During the execution of the algorithm, the appearance color search jellyfish in the appearance color search jellyfish group move around their own positions in the search space of the qualified kelp product appearance color image data set to search for the qualified kelp product appearance color image data that matches the appearance image data of the kelp product sample Matched qualified kelp product appearance color image data and update their positions. The corresponding position update formula for each appearance color search jellyfish is as follows: , where represents the individual appearance color search jellyfish the position in the search space of the qualified kelp product appearance color image data set after the th iteration, represents the individual appearance color search jellyfish the position in the search space of the qualified kelp product appearance color image data set after the th iteration, represents a random function with a value range of ; and represent the upper and lower bounds of the search space, that is, the upper and lower bounds of the data for searching the qualified kelp product appearance color image data that matches the appearance image data of the kelp product sample in the search space of the qualified kelp product appearance color image data set to search for the qualified kelp product appearance color image data that matches the appearance image data of the kelp product sample Matched qualified kelp product appearance color image data , =0.1 represents the motion system data;

[0090] S323. During the execution of the algorithm, the appearance color search jellyfish in the appearance color search jellyfish group search for the qualified kelp product appearance color image data that matches the appearance image data of the kelp product sample in the search space of the qualified kelp product appearance color image data set The control time of the behavior is as follows: Matched qualified kelp product appearance color image data , where represents the control time required for the individual appearance color search jellyfish after the th iteration, ​ is the maximum number of iteration data;

[0091] S324. When the algorithm execution meets the boundary condition and the appearance color search jellyfish moves out of the qualified kelp product appearance color image data set with boundaries search space, the appearance color search jellyfish returns to the opposite search area in the qualified kelp product appearance color image data set search space and continues to search for the qualified kelp product appearance color image data that matches the appearance image data of the kelp product sample ; the movement boundary condition of the appearance color search jellyfish satisfies the following formula: where

[0092] ; among them represents the updated position of the appearance color search jellyfish individual in the dimensional space, that is, the updated position in the search space of the qualified kelp product appearance color image data set with the spatial position dimension of ; represents the position before the update of the appearance color search jellyfish individual in the dimensional space, that is, the position before the update in the search space of the qualified kelp product appearance color image data set with the spatial position dimension of ; ; ;

[0093] S325. When the algorithm meets the maximum iteration data, output the image feature matching result of the appearance image data of the kelp product sample and the qualified kelp product appearance color image data and generate a kelp product sample appearance color analysis data set where represents the kelp product sample appearance color analysis data corresponding to the appearance image data of the kelp product sample ;

[0094] When and are successfully matched in image features, it indicates that the appearance color of the kelp product sample is qualified, and then output the kelp product sample appearance color analysis data as qualified;

[0095] When and fail to match in image features, it indicates that the appearance color of the kelp product sample is unqualified, and then output the kelp product sample appearance color analysis data as unqualified.

[0096] Based on the analysis data of the appearance color of kelp product samples, perform weighted numerical measurement processing on the qualified appearance color samples of kelp products to generate the weighted data of the qualified appearance color samples of kelp products, and conduct analysis processing on the appearance color quality grade of kelp products with the weighted threshold of the qualified appearance color samples of high-quality kelp products. The operation steps for constructing the appearance color quality grade data of kelp products are as follows:

[0097] S41. Establish the weighted threshold of the qualified appearance color samples of high-quality kelp products , where the weighted threshold of the qualified appearance color samples of high-quality kelp products represents the minimum value of the weight of the qualified appearance color samples of high-quality kelp products among all appearance samples;

[0098] S42. Use the breadth-first search algorithm to search in the set of analysis data of the appearance color of kelp product samples to find the number of kelp product samples whose analysis data of the appearance color is qualified, and generate the number of qualified appearance color samples of kelp products ; ;

[0099] S43. Perform numerical division measurement processing on the number of qualified appearance color samples of kelp products and the total number of kelp product samples to generate the weighted data of the qualified appearance color samples of kelp products , where ;

[0100] S44. Compare the weighted data of the qualified appearance color samples of kelp products with the weighted threshold of the qualified appearance color samples of high-quality kelp products to construct the appearance color quality grade data of kelp products based on the numerical comparison result ;

[0101] When ≥ , it means that the appearance color of the kelp product meets the weight requirement of the qualified appearance color samples in the appearance color of high-quality kelp products, and the output appearance color quality grade data is grade one;

[0102] When < , it means that the appearance color of the kelp product does not meet the weight requirement of the qualified appearance color samples in the appearance color of high-quality kelp products, and the output appearance color quality grade data is grade two.

[0103] Perform the identification process of the belt body defect characteristics of the kelp product samples based on the appearance image data of the kelp product samples and the image data of the belt body defect characteristics of the kelp product, and the operation steps for generating the identification data of the belt body defects of the kelp product samples are as follows:

[0104] S51. Establish a set of image data of the belt body defect characteristics of the kelp product , ; where represents the th type of image data of the belt body defect characteristics of the kelp product, represents the maximum value of the number of belt body defect types of the kelp product. The belt body defect types of the kelp product include belt body scratches, belt body black spots, belt body holes, and belt body breaks;

[0105] S52. When the data of the appearance color quality grade of the kelp product is constructed, use the uniform cost search algorithm to match the appearance image data of the kelp product samples in the set of appearance image data of the kelp product samples with the image data of the belt body defect characteristics of the kelp product in the set of image data of the belt body defect characteristics of the kelp product in an orderly manner according to the numbering of the appearance images of the kelp product samples, and generate a set of identification data of the belt body defects of the kelp product samples based on the image feature matching results, where represents the identification data of the belt body defects of the kelp product samples corresponding to the appearance image data of the kelp product samples ; where represents the identification data of the belt body defects of the kelp product samples corresponding to the appearance image data of the kelp product samples ;

[0106] When and are successfully matched in terms of image features, it indicates that the kelp product sample has belt body defect characteristics, and then output the identification data of the belt body defects of the kelp product sample as existing;

[0107] When and are not successfully matched in terms of image features, it indicates that the kelp product sample does not have belt body defect characteristics, and then output the identification data of the belt body defects of the kelp product sample as non - existent.

[0108] Perform the weight value measurement process for the kelp product samples without belt body defects based on the identification data of the belt body defects of the kelp product samples, generate the weight data of the kelp product samples without belt body defects, and perform the belt body quality grade analysis process of the kelp product with the weight threshold of the high - quality kelp product samples without belt body defects, and the operation steps for constructing the belt body quality grade data of the kelp product are as follows:

[0109] S61. Establish the weight threshold of the sample without belt body defects for high-quality kelp products , where the weight threshold of the sample without belt body defects for high-quality kelp products represents the minimum value of the weight of the samples without belt body defects among all appearance samples of high-quality kelp products;

[0110] S62. Use the breadth-first search algorithm to search for the belt body defect recognition data of kelp product samples in the data set to obtain the number of kelp product samples without belt body defects, and generate the number of kelp product samples without belt body defects ; ;

[0111] S63. Divide the number of kelp product samples without belt body defects by the total number of kelp product samples to generate the weight data of kelp product samples without belt body defects , where ;

[0112] S64. Compare the weight data of kelp product samples without belt body defects with the weight threshold of the sample without belt body defects for high-quality kelp products to construct the belt body quality grade data of kelp products ;

[0113] When ≥ , it means that the belt body defects of kelp products meet the weight requirements of the samples without belt body defects in the belt body defects of high-quality kelp products, and the belt body quality grade data is output as first grade;

[0114] When < , it means that the belt body defects of kelp products do not meet the weight requirements of the samples without belt body defects in the belt body defects of high-quality kelp products, and the belt body quality grade data is output as second grade.

[0115] Through the mutual cooperation of the kelp product sample appearance color analysis unit and the kelp product sample belt defect identification unit, an artificial intelligence algorithm is used to perform intelligent and standardized analysis on the appearance color and belt defects of the kelp product image samples, improving the accuracy and efficiency of the quality sorting and color selection of kelp products; the qualified appearance color sample weight measurement unit of the kelp product and the non-belt defect sample weight measurement unit of the kelp product cooperate with each other to digitally count the weights of the qualified appearance color samples and the samples without belt defects in the kelp product image samples, realizing the precise quantification of the appearance color and belt defect characteristics in the kelp products; the kelp product appearance color quality grade analysis unit and the kelp product belt quality grade analysis unit cooperate with each other to compare the kelp product appearance color weight and the belt defect weight with the appearance color and belt quality threshold values, scientifically analyzing the appearance color and belt quality grades of the kelp products, and improving the scientific nature of the kelp quality sorting and color selection.

[0116] Further, please refer to Figure 1 - Figure 2 Based on the quality grade parameters of the appearance color and belt of the kelp product and the kelp product quality grade data corresponding to different appearance colors and belt quality grades, the following operation steps for comprehensive judgment of the quality grade of the kelp product, constructing the kelp product quality grade data, and performing the kelp product quality sorting operation are as follows:

[0117] S71. Establish a set of kelp product quality grade data corresponding to different appearance colors and belt quality grades where represents that when the kelp product appearance color quality grade data and the kelp product belt quality grade data are both at the first level, the kelp product quality grade data corresponding to different appearance colors and belt quality grades is , indicating that the kelp product quality grade is at the first level; represents that when the kelp product appearance color quality grade data is at the first level and the kelp product belt quality grade data is at the second level, or when the kelp product appearance color quality grade data is at the second level and the kelp product belt quality grade data is at the first level, the kelp product quality grade data corresponding to different appearance colors and belt quality grades is , indicating that the kelp product quality grade is at the second level; represents that when the kelp product appearance color quality grade data and the kelp product belt quality grade data The data of different appearance colors corresponding to the second level and the kelp product quality level corresponding to the belt body mass level are , indicating that the quality level of the kelp product is the third level;

[0118] S72. The kelp product appearance color quality level data constructed in the S44 step and the S64 step and the kelp product belt body mass level data are matched with the kelp product quality level data set corresponding to different appearance colors and belt body mass levels according to the keywords of appearance color and belt body mass level in the kelp product quality level data corresponding to different appearance colors and belt body mass levels, and the kelp product appearance color quality level data and the kelp product belt body mass level data corresponding to the kelp product quality level data corresponding to different appearance colors and belt body mass levels are searched out, and the kelp product quality level data is constructed through data identification ;

[0119] S73. According to the kelp product quality level data the kelp product is sorted by quality through the material conveyor belt.

[0120] Through the cooperation of the kelp product quality level storage unit corresponding to different appearance colors and belt body mass levels and the kelp product quality level comprehensive judgment unit, the preset standard kelp product quality level data corresponding to different appearance colors and belt body mass levels and the quality level parameters of the appearance color and belt body of the kelp product are comprehensively judged for the kelp product quality level, and the kelp product quality level parameters are constructed efficiently and accurately, realizing the intelligent evaluation of the kelp product quality level; the kelp product quality sorting operation execution unit efficiently and intelligently executes the kelp product quality classification and sorting operation based on the kelp product quality level parameters, improving the efficiency and accuracy of the kelp product quality sorting.

[0121] Example 2:

[0122] Please refer to Figure 1 - Figure 2 , a kelp quality sorting color selection management system based on artificial intelligence, used to implement the kelp quality sorting color selection management method. The system includes a kelp product color selection information processing module, a kelp product color and belt body mass level evaluation module, and a kelp product quality evaluation and sorting execution module;

[0123] The kelp product color selection information processing module includes a kelp product appearance feature image acquisition unit, a kelp product appearance feature image preprocessing unit, and a kelp product sample appearance image generation unit;

[0124] The kelp product appearance feature image acquisition unit acquires kelp product appearance feature image data through a cloud lens; the kelp product appearance feature image preprocessing unit preprocesses the kelp product appearance feature image data to generate standard kelp product appearance feature image data; the kelp product sample appearance image generation unit performs kelp product appearance image sample generation processing on the standard kelp product appearance feature image data to generate kelp product sample appearance image data;

[0125] The kelp product color and ribbon quality grade evaluation module includes a qualified kelp product appearance color image storage unit, a kelp product sample appearance color analysis unit, a qualified kelp product appearance color sample weight measurement unit, a high-quality kelp product qualified appearance color sample weight threshold storage unit, and a kelp product appearance color quality grade analysis unit; a kelp product ribbon defect feature image storage unit, a kelp product sample ribbon defect identification unit, a kelp product ribbon defect-free sample weight measurement unit, a high-quality kelp product ribbon defect-free sample weight threshold storage unit, and a kelp product ribbon quality grade analysis unit;

[0126] A qualified kelp product appearance color image storage unit for storing qualified kelp product appearance color image data; a kelp product sample appearance color analysis unit that, based on artificial intelligence, performs appearance color feature analysis processing on the kelp product sample appearance image data and the qualified kelp product appearance color image data to generate kelp product sample appearance color analysis data; a kelp product qualified appearance color sample weight measurement unit that performs weight value measurement processing on the kelp product qualified appearance color samples based on the kelp product sample appearance color analysis data to generate kelp product qualified appearance color sample weight data; a high-quality kelp product qualified appearance color sample weight threshold storage unit for storing the high-quality kelp product qualified appearance color sample weight threshold; a kelp product appearance color quality grade analysis unit that performs kelp product appearance color quality grade analysis processing based on the kelp product qualified appearance color sample weight data and the high-quality kelp product qualified appearance color sample weight threshold to construct kelp product appearance color quality grade data; a kelp product belt body defect feature image storage unit for storing kelp product belt body defect feature image data; a kelp product sample belt body defect recognition unit that performs belt body defect feature recognition processing on the kelp product sample based on the kelp product sample appearance image data and the kelp product belt body defect feature image data to generate kelp product sample belt body defect recognition data; a kelp product non-belt-body-defect sample weight measurement unit that performs weight value measurement processing on the kelp product non-belt-body-defect samples based on the kelp product sample belt body defect recognition data to generate kelp product non-belt-body-defect sample weight data; a high-quality kelp product non-belt-body-defect sample weight threshold storage unit for storing the high-quality kelp product non-belt-body-defect sample weight threshold; a kelp product belt body quality grade analysis unit that performs kelp product belt body quality grade analysis processing by comparing the kelp product non-belt-body-defect sample weight data with the high-quality kelp product non-belt-body-defect sample weight threshold to construct kelp product belt body quality grade data;

[0127] The kelp product quality evaluation and sorting execution module includes a kelp product quality grade storage unit corresponding to different appearance colors and belt body quality grades, a kelp product quality grade comprehensive judgment unit, and a kelp product quality sorting operation execution unit;

[0128] The kelp product quality grade storage unit corresponding to different appearance colors and belt body quality grades is used to store kelp product quality grade data corresponding to different appearance colors and belt body quality grades; the kelp product quality grade comprehensive judgment unit performs kelp product quality grade comprehensive judgment processing based on the kelp product appearance color and belt body quality grade parameters and the kelp product quality grade data corresponding to different appearance colors and belt body quality grades to construct kelp product quality grade data; the kelp product quality sorting operation execution unit executes the kelp product quality sorting operation according to the kelp product quality grade data.

[0129] Although embodiments of the present invention have been shown and described, it will be understood by those of ordinary skill in the art that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention, and the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for managing kelp quality sorting and color selection, characterized in that: The method comprises the following steps: S1, collecting kelp product appearance feature image data and performing image preprocessing to generate standard kelp product appearance feature image data, said S1 comprises the following steps: S11. Collect the appearance feature images of the kelp products passing through the monitoring port online through the cloud camera installed on the material conveyor belt, and generate the appearance feature image data of the kelp products ; S12, using an adaptive filtering method to filter the kelp product appearance feature image data After image noise reduction preprocessing, the standard kelp product appearance feature image data is generated ; S2, performing a process of generating a kelp product appearance image sample on the standard kelp product appearance feature image data to generate kelp product sample appearance image data, wherein S2 comprises the following steps: S21, the standard kelp product appearance feature image data A square with a side length of L is used to perform grid division processing on the appearance feature image, and a data set of kelp product sample appearance images is generated. , ;in Indicates Appearance image data of kelp product samples, represents the total number of kelp product samples; S3, performing appearance color feature analysis processing on the kelp product sample according to the appearance image data of the kelp product sample and the appearance color image data of the qualified kelp product to generate appearance color analysis data of the kelp product sample, wherein S3 comprises the following steps: S31. Establishing a data set of appearance color images of qualified kelp products , ;in Indicates Appearance color image data of qualified kelp products, Indicates the maximum number of appearance color types of qualified kelp products; S32, collecting the kelp product sample appearance image data set Appearance image data of kelp product samples described in The qualified kelp product appearance color image data set is ordered according to the number of kelp product sample appearance images. Appearance color image data of qualified kelp products described in Perform image feature matching and generate a data set of kelp product sample appearance color analysis based on the image feature matching results , execute to generate kelp product sample appearance color analysis data set The specific steps are as follows: S321, initializing the algorithm iteration data and executing the ocean current simulation process; setting the maximum algorithm iteration data to T; S322, during the algorithm execution process, the appearance color search jellyfish group appearance color search jellyfish in the qualified kelp product appearance color image data set Searching for image data related to the appearance of kelp product samples by moving around their own positions in the search space Matching qualified kelp product appearance color image data and update the location; S323, algorithm execution, appearance color search, appearance color search, jellyfish group, appearance color search, jellyfish in qualified kelp product appearance color image data set Search the search space for the image data related to the appearance of kelp product samples Matching qualified kelp product appearance color image data The duration of the behavior; S324, the algorithm executes to meet the boundary conditions, when the appearance color search jellyfish moves out of the qualified kelp product appearance color image data set with boundaries When searching the space, the appearance color search for jellyfish returns to the qualified kelp product appearance color image data set. The opposite search area in the search space continues to search for the image data of the kelp product sample appearance Matching qualified kelp product appearance color image data ; S325: When the algorithm meets the maximum iteration number data, output the appearance image data of the kelp product sample Image data of the appearance and color of the qualified kelp products The image feature matching results are used to generate a data set for the appearance color analysis of kelp product samples ,in Indicates the appearance image data of kelp product samples Corresponding kelp product sample appearance color analysis data; when and If the image feature matching is successful, the appearance color analysis data of the kelp product sample will be output. To be qualified; when and If the image features are not matched successfully, the appearance color analysis data of the kelp product sample will be output. for failure to meet the standards; S4, based on the kelp product sample appearance color analysis data, weight numerical measurement processing of qualified kelp product appearance color samples is performed to generate weight data of qualified kelp product appearance color samples, and the appearance color quality grade analysis processing of kelp products is performed with the weight threshold of qualified appearance color samples of high-quality kelp products to construct kelp product appearance color quality grade data, wherein S4 comprises the following steps: S41. Establish the weight threshold of qualified appearance color samples of high-quality kelp products ; S42, using a breadth-first search algorithm to analyze the appearance color data set of the kelp product sample Search for the appearance color analysis data of the kelp product sample The number of qualified kelp product samples and the number of qualified appearance color samples of kelp products are generated ; S43, the number of qualified appearance color samples of the kelp product Total number of kelp product samples Perform numerical quotient measurement processing to generate weight data of qualified appearance color samples of kelp products ,in ; S44, the weight data of qualified appearance color samples of the kelp product Weight threshold of qualified appearance color samples of high-quality kelp products Perform numerical comparison and construct the kelp product appearance color quality grade data based on the numerical comparison results ; when , then output the kelp product appearance color quality grade data is level one; when , then output the kelp product appearance color quality grade data is level 2; S5, performing body defect feature recognition processing of the kelp product sample according to the appearance image data of the kelp product sample and the body defect feature image data of the kelp product, and generating body defect recognition data of the kelp product sample, wherein S5 comprises the following steps: S51. Establishing a dataset of defect feature images of kelp products , ;in Indicates The defect feature image data of kelp products. It indicates the maximum value of the number of defect types of kelp products; S52, when the kelp product appearance color quality grade data When the construction is completed, the kelp product sample appearance image data set is collected using a unified cost search algorithm Appearance image data of kelp product samples described in The image data set of the kelp product body defect feature is ordered according to the number of the kelp product sample appearance images. The kelp product body defect feature image data described in Perform image feature matching and generate a kelp product sample body defect recognition data set based on the image feature matching results ,in Indicates the appearance image data of kelp product samples Corresponding kelp product sample body defect identification data; when and If the image feature matching is successful, the kelp product sample body defect recognition data will be output. for existence; when and If the image features are not matched successfully, the kelp product sample body defect recognition data is output for not existing; S6, based on the kelp product sample strip defect identification data, weighted numerical measurement processing of kelp product strip defect-free samples is performed to generate weighted data of kelp product strip defect-free samples and analyze the strip quality grade of kelp products with the weight threshold of high-quality kelp product strip defect-free samples to construct kelp product strip quality grade data, wherein S6 includes the following steps: S61. Establish a sample weight threshold for high-quality kelp products without body defects ; S62, using a breadth-first search algorithm to identify the kelp product sample body defect data set Search for the kelp product sample's kelp body defect identification data The number of kelp product samples that do not exist, and the number of kelp product samples without body defects ; S63, the number of samples of the kelp product without body defects Total number of kelp product samples Perform numerical quotient measurement processing to generate sample weight data for kelp products without kelp body defects ,in ; S64, the kelp product does not have the weight data of the sample with body defects The sample weight threshold for the high-quality kelp products without body defects Carry out numerical comparison and construct the kelp product body quality grade data based on the numerical comparison results ; when ≥ , then output the kelp product body quality grade data is level one; when < , then output the kelp product body quality grade data is level 2; S7, based on the appearance color of the kelp product, the quality grade parameters of the strip body and the kelp product quality grade data corresponding to different appearance colors and strip body quality grades, the quality grade data of the kelp product is comprehensively judged and processed, the kelp product quality grade data is constructed and the kelp product quality sorting operation is performed, and the S7 includes the following steps: S71. Establish a data set of kelp product quality grades corresponding to different appearance colors and kelp body quality grades ,in Indicates the appearance color quality grade data of the kelp product And the kelp product body quality grade data The corresponding data of kelp product quality grades for different appearance colors and kelp body quality grades are as follows: , Indicates that the quality grade of kelp products is Grade One; Indicates the appearance color quality grade data of the kelp product The kelp product body quality grade data is Grade II or when the kelp product appearance color quality grade data The kelp product body quality grade data is The corresponding data of different appearance colors and kelp quality grades when it is level 1 are as follows: , Indicates that the quality grade of kelp products is Level 2; Indicates the appearance color quality grade data of the kelp product And the kelp product body quality grade data The corresponding data of kelp product quality grades for different appearance colors and kelp body quality grades are as follows: , It means that the quality grade of kelp products is Grade III; S72, the kelp product appearance color quality grade data constructed in step S44 and step S64 And the kelp product body quality grade data A data set of kelp product quality grades corresponding to the different appearance colors and kelp body quality grades The kelp product quality grade data corresponding to the different appearance colors and belt quality grades described in the above are matched according to the appearance color and belt quality grade keywords to search for the kelp product appearance color quality grade data And the kelp product body quality grade data The corresponding different appearance colors and belt body quality grades correspond to the kelp product quality grade data and the kelp product quality grade data is constructed through data identification. ; S73, according to the kelp product quality grade data The kelp products are sorted by quality through the material conveyor belt.

2. An artificial intelligence-based kelp quality sorting and color sorting management system, used to implement the kelp quality sorting and color sorting management method described in claim 1, characterized in that: The system comprises a kelp product color sorting information processing module, a kelp product color and kelp body quality grade evaluation module, and a kelp product quality evaluation and sorting execution module.

Images