Glutinous rice chicken package lotus leaf treatment management method based on artificial intelligence

Abstract

The invention discloses a glutinous rice chicken packaging lotus leaf processing management method based on artificial intelligence, and relates to the technical field of food making, and the method comprises the following steps: decomposing a lotus leaf processing process into at least one processing step; decomposing the appearance requirements to obtain associated requirements of the processing steps; the machining process of the machining steps is monitored, and the machining steps are corrected; forming a lotus leaf processing scale of the processing steps; and forming an abnormal feature set for the processing steps, identifying abnormal conditions occurring in the processing steps based on the abnormal feature set, forming a scheduling scheme, and performing abnormal repair according to the scheduling scheme. According to the method, the identification accuracy is ensured by obtaining the association requirements of the processing steps, correcting the processing steps, forming the lotus leaf processing scale of the processing steps, forming the abnormal feature set for the processing steps and forming the scheduling scheme, so that the processing precision is improved, and the use consumption requirements of the lotus leaves can be met under the condition that excessive equipment is not added.

Description

Technical Field

[0001] The present invention relates to the technical field of food production, and specifically relates to a management method for processing lotus leaves for packaging glutinous rice chicken based on artificial intelligence. Background Art

[0002] The method of making glutinous rice chicken is to put fillings such as chicken, barbecued pork, ribs, salted duck eggs, and black mushrooms into glutinous rice, and then wrap it tightly with lotus leaves and steam it in a steamer; when eating glutinous rice chicken, it is full of the fragrance of lotus leaves, sticky when chewing and has the meaty aroma of chicken. Lotus leaves have a relatively important impact on the storage and appearance of glutinous rice chicken. Therefore, it is necessary to ensure that the lotus leaves meet the packaging requirements.

[0003] The processing of lotus leaves involves multiple processes. In order to ensure the processing quality and quantity of lotus leaves, it is necessary to overall control the production line. However, it is difficult for the existing technology to accurately respond to various situations that may occur in the processing of lotus leaves. Summary of the Invention

[0004] To solve the above technical problems, a management method for processing lotus leaves for packaging glutinous rice chicken based on artificial intelligence is provided, and this technical solution solves the problems raised in the above background art.

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

[0006] A management method for processing lotus leaves for packaging glutinous rice chicken based on artificial intelligence, including:

[0007] Decompose the lotus leaf processing process into at least one processing step, and process according to the order of at least one processing step to obtain lotus leaf products that meet the packaging standards of glutinous rice chicken. The processing steps are lotus leaf screening, lotus leaf cleaning, lotus leaf drying, and lotus leaf sorting;

[0008] Obtain the consumption speed of lotus leaf products for packaging glutinous rice chicken, and obtain the appearance requirements of lotus leaf products for packaging glutinous rice chicken;

[0009] Decompose the appearance requirements to obtain the associated requirements of the processing steps;

[0010] Based on the associated requirements of the processing steps, monitor the processing process of the processing steps, and based on the monitoring results, correct the processing steps;

[0011] Based on the consumption speed of lotus leaf products, form the lotus leaf processing scale of the processing steps;

[0012] Form an abnormal feature set for the processing steps. Based on the abnormal feature set, identify abnormal situations that occur in the processing steps. Based on the identification results, form a scheduling plan and perform abnormal repair according to the scheduling plan.

[0013] Preferably, the appearance requirements for the finished lotus leaf product used for wrapping the sticky rice chicken are as follows:

[0014] The appearance requirements for the finished lotus leaf product include the target shape of the lotus leaf, the target color of the lotus leaf, the allowable proportion of impurities on the lotus leaf surface, the target dryness of the lotus leaf, and the allowable proportion of defects on the lotus leaf.

[0015] Preferably, the decomposition of the appearance requirements to obtain the associated requirements for the processing steps includes the following steps:

[0016] Regarding the target shape of the lotus leaf, the target color of the lotus leaf, and the allowable proportion of defects on the lotus leaf as the associated requirements for the processing steps corresponding to the lotus leaf screening;

[0017] Regarding the target color of the lotus leaf, the allowable proportion of impurities on the lotus leaf surface, and the allowable proportion of defects on the lotus leaf as the associated requirements for the processing steps corresponding to the lotus leaf cleaning;

[0018] Regarding the target dryness of the lotus leaf as the associated requirement for the processing step corresponding to the lotus leaf air-drying;

[0019] Regarding the target shape of the lotus leaf as the associated requirement for the processing step corresponding to the lotus leaf arranging.

[0020] Preferably, the monitoring of the processing process of the processing steps based on the associated requirements of the processing steps includes the following steps:

[0021] Obtain a sample image of the finished lotus leaf product that meets the associated requirements. According to the presentation characteristics of the associated requirements, perform feature extraction on the sample image to obtain the sample image features of the associated requirements, where the dryness is identified using infrared imaging;

[0022] Perform image recognition on the to-be-tested lotus leaf obtained through the processing steps to obtain the actual image features of the associated requirements;

[0023] Based on the human eye recognition data, form a preset proportion range;

[0024] When the difference proportion between the actual image features of the associated requirements and the sample image features belongs to the preset proportion range, then regard the to-be-tested lotus leaf as a non-conforming lotus leaf;

[0025] Calculate the proportion of non-conforming lotus leaves in the to-be-tested lotus leaves as the feature proportion;

[0026] When the feature proportion exceeds the preset value, regard the processing step corresponding to the associated requirements as the processing step to be corrected, otherwise, do not perform any processing, where the preset value is the allowable error value for lotus leaf processing obtained based on experience.

[0027] Preferably, the formation of the preset proportion range based on the human eye recognition data includes the following steps:

[0028] Uniformly divide the value range of the difference ratio between the actual image features and the sample image features to obtain at least one recognition point;

[0029] Obtain at least one image to be verified whose difference ratio from the sample image is equal to the value at the recognition point;

[0030] When the human eye cannot recognize the sample image from the mixed image of the sample image and at least one image to be verified, then take the recognition point as the target recognition point;

[0031] Take the interval formed by the maximum value and the minimum value of the target recognition point as the preset ratio range.

[0032] Preferably, the correction of the processing steps based on the monitoring results includes the following steps:

[0033] Obtain the actual processing parameters of the processing steps to be corrected, and take the average of the difference ratios between the actual image features and the sample image features required for the associated requirements of the processing steps to be corrected to obtain the average difference ratio;

[0034] Obtain at least one processing device in the processing steps to be corrected, and each processing device performs automated processing on one lotus leaf;

[0035] Statistically calculate the proportion of unqualified lotus leaves processed by the processing device as the target proportion, and take the processing device with the target proportion exceeding the preset value as the processing device to be corrected;

[0036] Accumulate the target proportions of the processing devices to be corrected to obtain the comprehensive proportion;

[0037] Multiply the average difference ratio by the target proportion of the processing device to be corrected and divide by the comprehensive proportion to obtain the adjusted allocation ratio;

[0038] Subtract the product of the actual processing parameters and the adjusted allocation ratio from the actual processing parameters to obtain the corrected processing parameters;

[0039] Adjust the actual processing parameters of the processing device to be corrected to the corrected processing parameters.

[0040] Preferably, the formation of the lotus leaf processing scale of the processing steps based on the consumption speed of the lotus leaf finished products includes the following steps:

[0041] Statistically calculate the consumption time of a single processing device in the processing steps for processing one lotus leaf, and statistically calculate the transmission time for transferring the lotus leaf to the next processing step in the processing steps;

[0042] Divide 1 by the sum of the consumption time and the transmission time to obtain the single-piece processing speed;

[0043] Divide the consumption speed of the lotus leaf finished products by the single-piece processing speed to obtain the preliminary quantity;

[0044] Statistically analyze the average pass rate of all processing steps;

[0045] Number the processing steps in reverse order of the processing sequence. That is, the lotus leaf screening, lotus leaf cleaning, lotus leaf drying, and lotus leaf finishing are numbered 4, 3, 2, and 1 in sequence;

[0046] Use the processing scale formula to calculate the final quantity. The lotus leaf processing scale of the processing step is the parallel operation of the processing equipment with the final quantity in the processing step;

[0047] The processing scale formula is as follows:

[0048]

[0049] Where A is the final quantity, a is the preliminary quantity, b is the average pass rate, and n is the number of the processing step.

[0050] Preferably, the formation of the abnormal feature set for the processing step includes the following steps:

[0051] When the output speed of the lotus leaf in the processing step is less than the consumption speed of the finished lotus leaf, the processing step is regarded as an abnormal processing step;

[0052] Based on historical data, obtain the normal image of the processing equipment of the abnormal processing step, and obtain the action parameters of the processing equipment in the normal image;

[0053] Perform a difference analysis between the normal image of the processing equipment of the abnormal processing step and the actual image of the processing equipment of the abnormal processing step to obtain an abnormal feature set, and the abnormal feature set is composed of abnormal features;

[0054] Obtain at least one abnormal cause that causes the abnormality, statistically analyze the occurrence frequency of the abnormal cause that appears simultaneously with the abnormal feature, regard the abnormal cause with the highest occurrence frequency as the abnormal cause that causes the abnormal feature, and pair the abnormal feature with the abnormal cause that causes the abnormal feature.

[0055] Preferably, the recognition of the abnormal situation in the processing step based on the abnormal feature set includes the following steps:

[0056] Take the center of the processing equipment as the origin, and establish coordinate systems in the normal image and the abnormal image of the abnormal situation in the same way;

[0057] Perform a difference comparison between the normal image and the abnormal image to obtain the difference part;

[0058] Obtain the abnormal features that appear in the difference part as the target abnormal features.

[0059] Preferably, the formation of the scheduling plan based on the recognition result includes the following steps:

[0060] Based on historical data, obtain the target processing solution for the abnormal cause corresponding to the target abnormal feature;

[0061] Use the target processing solution to regulate the processing equipment;

[0062] Monitor the lotus leaf output speed of the regulated processing equipment. When the lotus leaf output speed is 0, it is determined that there is a fault in the lotus leaf transfer equipment corresponding to the processing equipment, and repair or replacement is carried out on the lotus leaf transfer equipment corresponding to the processing equipment;

[0063] When the lotus leaf output speed is less than the lotus leaf finished product consumption speed, taking the center of the processing equipment as the origin, establish coordinate systems in the normal image and the actual image after regulation in the same way;

[0064] Calculate the absolute value of the difference between the pixel values of the pixel points with the same coordinates in the processing equipment of the normal image and the processing equipment of the actual image after regulation, and accumulate to obtain the difference sum;

[0065] When the difference sum is greater than the preset gap, replace the processing equipment; otherwise, do nothing.

[0066] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0067] By obtaining the association requirements of the processing steps, correcting the processing steps, forming the lotus leaf processing scale of the processing steps, forming the abnormal feature set of the processing steps, and forming the scheduling plan, it is possible to identify and targetedly regulate most of the possible abnormalities involved in the large-scale processing of lotus leaves. Due to the setting of multiple layers of identification mechanisms, different attributes of abnormalities can be classified and identified, thereby ensuring the accuracy of identification, and then improving the accuracy of processing. Moreover, in order to ensure the processing speed, the lotus leaf processing scale of the processing steps is formed, and different processing scales are set according to different processing steps, so as to ensure that the overall processing speed is consistent with the lotus leaf finished product consumption speed, and then it is possible to meet the usage and consumption requirements of lotus leaves without adding too many devices. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] Figure 1 It is a flow chart of the management method for processing lotus leaves for glutinous rice chicken packaging based on artificial intelligence of the present invention;

[0069] Figure 2 It is a flow chart of decomposing the appearance requirements of the present invention to obtain the association requirements of the processing steps;

[0070] Figure 3 It is a flow chart of monitoring the processing process of the processing steps based on the association requirements of the processing steps of the present invention;

[0071] Figure 4Schematic flow diagram for forming a preset ratio range based on human eye recognition data in the present invention;

[0072] Figure 5 Schematic flow diagram for correcting processing steps based on monitoring results in the present invention;

[0073] Figure 6 Schematic flow diagram for forming the processing scale of lotus leaves in processing steps based on the consumption speed of finished lotus leaf products for glutinous rice chicken packaging in the present invention;

[0074] Figure 7 Schematic flow diagram for forming an abnormal feature set for processing steps in the present invention;

[0075] Figure 8 Schematic flow diagram for identifying abnormal situations occurring in processing steps based on the abnormal feature set in the present invention;

[0076] Figure 9 Schematic flow diagram for forming a scheduling plan based on the recognition result in the present invention. Detailed implementation manners

[0077] 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.

[0078] Refer to Figure 1 As shown, an artificial intelligence-based management method for processing lotus leaves for glutinous rice chicken packaging includes:

[0079] Decompose the lotus leaf processing process into at least one processing step, and process according to the order of at least one processing step to obtain lotus leaf finished products meeting the glutinous rice chicken packaging standard. The processing steps are lotus leaf screening, lotus leaf washing, lotus leaf drying, and lotus leaf sorting;

[0080] Obtain the consumption speed of lotus leaf finished products for glutinous rice chicken packaging, and obtain the appearance requirements of lotus leaf finished products for glutinous rice chicken packaging;

[0081] Decompose the appearance requirements to obtain the associated requirements for processing steps;

[0082] Based on the associated requirements for processing steps, monitor the processing process of processing steps, and based on the monitoring results, correct the processing steps;

[0083] Based on the consumption speed of lotus leaf finished products, form the processing scale of lotus leaves in processing steps;

[0084] Form an abnormal feature set for processing steps, based on the abnormal feature set, identify abnormal situations occurring in processing steps, and based on the recognition result, form a scheduling plan and perform abnormal repair according to the scheduling plan.

[0085] The appearance requirements for the finished lotus leaf used for wrapping glutinous rice chicken are as follows:

[0086] The appearance requirements for the finished lotus leaf include the target shape of the lotus leaf, the target color of the lotus leaf, the allowable proportion of impurities on the lotus leaf surface, the target dryness of the lotus leaf, and the allowable proportion of defects on the lotus leaf.

[0087] When conducting the management of lotus leaf processing for glutinous rice chicken packaging, if there is no problem with the production line for lotus leaf processing, no processing is required. If there is a problem with the production line for lotus leaf processing, processing is required. The problems with the production line for lotus leaf processing may be that there are deviations in the setting of the action parameters of the processing equipment for the lotus leaf, there may be problems with the transfer of the lotus leaf between processing steps, or there may be a malfunction in the processing equipment itself. Since different types of problems vary greatly, it is difficult to use the same mechanism for synchronous identification. Therefore, in this solution, corresponding steps are set to solve this problem specifically.

[0088] Refer to Figure 2 As shown, the appearance requirements are decomposed, and the associated requirements for the processing steps are as follows:

[0089] Take the target shape of the lotus leaf, the target color of the lotus leaf, and the allowable proportion of defects on the lotus leaf as the associated requirements for the processing steps of corresponding lotus leaf screening.

[0090] Take the target color of the lotus leaf, the allowable proportion of impurities on the lotus leaf surface, and the allowable proportion of defects on the lotus leaf as the associated requirements for the processing steps of corresponding lotus leaf cleaning.

[0091] Take the target dryness of the lotus leaf as the associated requirement for the processing step of corresponding lotus leaf drying.

[0092] Take the target shape of the lotus leaf as the associated requirement for the processing step of corresponding lotus leaf sorting.

[0093] During lotus leaf screening, it is necessary to ensure the consistency of the lotus leaves to be screened and ensure that the lotus leaves can fully wrap the glutinous rice chicken, thereby ensuring that its storage time can reach the required time. Also, since it needs to be bought and sold, it is necessary to ensure the appearance of the lotus leaves, mainly defects and color. Therefore, during lotus leaf screening, take the target shape of the lotus leaf, the target color of the lotus leaf, and the allowable proportion of defects on the lotus leaf as the associated requirements.

[0094] During lotus leaf cleaning, it is necessary to ensure thorough cleaning, and the cleaning effect can be evaluated from the color and impurities. Here, the consideration of lotus leaf defects is also because due to the coverage of mud or impurities, some defects may be covered, so before cleaning, the covered defects cannot be identified and need to be re-identified after cleaning.

[0095] The lotus leaves are dried to maintain their dryness. Wet lotus leaves are likely to cause the deterioration of the glutinous rice chicken, resulting in a shorter shelf life.

[0096] The lotus leaves are sorted to flatten them. When the lotus leaves are dried, wrinkles and the like are likely to appear, which will affect the appearance of the packaging.

[0097] Refer to Figure 3 As shown, based on the association requirements of the processing steps, monitoring the processing process of the processing steps includes the following steps:

[0098] Obtain a sample image of the finished lotus leaf that meets the association requirements. According to the presentation characteristics of the association requirements, extract features from the sample image to obtain the sample image features of the association requirements, where the dryness is identified using infrared imaging;

[0099] Perform image recognition on the lotus leaf to be tested obtained through the processing steps to obtain the actual image features of the association requirements;

[0100] Based on the human eye recognition data, form a preset proportion range;

[0101] When the difference ratio between the actual image features of the association requirements and the sample image features belongs to the preset proportion range, the lotus leaf to be tested is regarded as a non-conforming lotus leaf;

[0102] Calculate the proportion of non-conforming lotus leaves in the lotus leaves to be tested as the feature proportion;

[0103] When the feature proportion exceeds the preset value, the processing step corresponding to the association requirements is regarded as the processing step to be corrected. Otherwise, no processing is performed, where the preset value is the allowable error value of lotus leaf processing obtained based on experience.

[0104] The purpose of the monitoring here is to identify the processing effect of the lotus leaves. When its processing effect does not meet the association requirements, it indicates that there is a deviation in the parameter settings of the processing equipment. Therefore, it is necessary to adjust the parameters of the processing equipment. When performing the identification, it is necessary to consider the differences between machine recognition and human eye recognition, that is, the accuracy of human eye recognition is insufficient. Therefore, when the difference exceeds the recognition range of the human eye, the difference is ignored because it will not have a substantial impact on the appearance of the lotus leaves.

[0105] Refer to Figure 4 As shown, forming a preset proportion range based on the human eye recognition data includes the following steps:

[0106] Evenly divide the value range of the difference ratio between the actual image features and the sample image features to obtain at least one recognition point;

[0107] Obtain at least one image to be verified whose difference ratio from the sample image is equal to the value at the recognition point;

[0108] When the human eye cannot recognize the sample image from the mixed image of the sample image and at least one image to be verified, the recognition point is taken as the target recognition point;

[0109] The interval formed by the maximum value and the minimum value of the target recognition point is used as the preset ratio range.

[0110] Refer to Figure 5 As shown, based on the monitoring results, the correction of the processing steps includes the following steps:

[0111] Obtain the actual processing parameters of the processing step to be corrected, and take the average of the ratio of the difference between the actual image features of the associated requirements of the processing step to be corrected and the sample image features to obtain the average difference ratio;

[0112] Obtain at least one processing device in the processing step to be corrected, and each processing device performs automated processing on one lotus leaf;

[0113] Statistical proportion of unqualified lotus leaves processed by the processing equipment, as the target ratio, the processing equipment whose target ratio exceeds the preset value is used as the processing equipment to be corrected;

[0114] Accumulate the target ratios of the processing equipment to be corrected to obtain the comprehensive ratio;

[0115] Multiply the average difference ratio by the target ratio of the processing equipment to be corrected and divide by the comprehensive ratio to obtain the adjusted allocation ratio;

[0116] Subtract the product of the actual processing parameter and the adjusted allocation ratio from the actual processing parameter to obtain the corrected processing parameter;

[0117] Adjust the actual processing parameter of the processing equipment to be corrected to the corrected processing parameter.

[0118] During correction, different adjustments need to be loaded for different processing equipment to be corrected in the processing step to be corrected. Because the usage conditions of the processing equipment to be corrected are different, there are also slight differences in its maintenance and usage environment. Therefore, over time, there will be certain differences in the processing equipment to be corrected itself. The same parameter setting will show different effects on the processing equipment to be corrected. Therefore, differential adjustments need to be made for each processing equipment to be corrected to ensure that its final processing effects are nearly the same, thereby controlling the processing quality.

[0119] Refer to Figure 6 As shown, based on the consumption speed of the finished lotus leaves, the formation of the lotus leaf processing scale of the processing step includes the following steps:

[0120] Statistical time taken for a single processing device in the processing step to process one lotus leaf, and statistical transfer time for the lotus leaf to the next processing step in the processing step;

[0121] Divide 1 by the sum of the consumption time and the transmission time to obtain the single-piece processing speed;

[0122] Divide the consumption speed of the finished lotus leaf by the single-piece processing speed to obtain the preliminary quantity;

[0123] Statistically calculate the average qualified rate of all processing steps;

[0124] Number the processing steps in the reverse order of the processing sequence, that is, number the lotus leaf screening, lotus leaf cleaning, lotus leaf drying, and lotus leaf finishing as 4, 3, 2, and 1 in sequence;

[0125] Use the processing scale formula to calculate the final quantity. The lotus leaf processing scale of the processing step is that the processing equipment with the final quantity runs in parallel in the processing step;

[0126] The processing scale formula is as follows:

[0127]

[0128] Wherein, A is the final quantity, a is the preliminary quantity, b is the average qualified rate, and n is the number of the processing step.

[0129] In each processing step, defective products will occur. Therefore, before entering the next processing step, it is necessary to remove the defective products. Thus, it is necessary to set more processing equipment than the actual demand in each processing step to ensure that the quantity of the screened lotus leaves meets the demand. Therefore, the processing equipment in each process is calculated according to the screening residue ratio each time, and the calculation is carried out in a reverse order, that is, starting from the last processing step. Because the last processing step still needs to be screened once, the number of its processing equipment is Multiplied by the average qualified rate b to obtain a. And according to the way of obtaining a, the output speed of a processing equipment in the last processing step is equal to the consumption speed of the finished lotus leaf. The same is true for the penultimate processing step. It still needs to be screened twice. Therefore, the number of its processing equipment is After two screenings, it is thus multiplied by the square of the average qualified rate b to obtain a. And according to the way of obtaining a, the output speed of a processing equipment in the penultimate processing step is equal to the consumption speed of the finished lotus leaf. The same is true for the remaining steps. Thus, the set processing scale can just meet the production and screening requirements, and further no redundant equipment will be set, which can control the cost.

[0130] Refer to Figure 7 As shown, forming an abnormal feature set for the processing step includes the following steps:

[0131] When the speed of the lotus leaf produced by the processing step is less than the consumption speed of the finished lotus leaf, then the processing step is regarded as an abnormal processing step;

[0132] Based on historical data, obtain the normal images of the processing equipment for abnormal processing steps, and obtain the action parameters of the processing equipment in the normal images;

[0133] Perform a difference analysis on the normal images of the processing equipment for abnormal processing steps and the actual images of the processing equipment for abnormal processing steps to obtain an abnormal feature set, which is composed of abnormal features;

[0134] Obtain at least one abnormal cause that causes the abnormality, count the occurrence frequency of the abnormal causes that appear simultaneously with the abnormal features, take the abnormal cause with the largest occurrence frequency as the abnormal cause that causes the abnormal features, and pair the abnormal features with the abnormal causes that cause the abnormal features.

[0135] The abnormality here is for the identification of the processing equipment, and various faults may exist. Therefore, through the identification of its actions, its faults are determined, and thus targeted repairs and treatments are carried out. After the treatment, a secondary judgment is made. Because the actual abnormalities that occur may not only come from the processing equipment, the lotus leaf output speed of the repaired processing equipment is monitored. When the lotus leaf output speed is 0, it is judged that there is a fault in the lotus leaf transfer equipment corresponding to the processing equipment. Because a fault in the lotus leaf transfer equipment will cause the lotus leaves to be blocked and then accumulate, and as the accumulation progresses, it will gradually lead to the out-of-control of the production line and ultimately inevitably lead to a lotus leaf output speed of 0. In this regard, repairs or replacements of the lotus leaf transfer equipment are required. There is also a situation where the processing equipment cannot be repaired. Therefore, the operation of the processed and regulated processing equipment is monitored. When its operation deviates too much from the normal situation, it means that it cannot be repaired, and thus, the equipment is replaced.

[0136] Refer to Figure 8 As shown, based on the abnormal feature set, identifying the abnormal situations that occur in the processing steps includes the following steps:

[0137] Taking the center of the processing equipment as the origin, establish coordinate systems in the normal images and the abnormal images of the abnormal situations in the same way;

[0138] Perform a difference comparison between the normal images and the abnormal images to obtain the different parts;

[0139] Obtain the abnormal features that appear in the different parts as the target abnormal features.

[0140] Refer to Figure 9 As shown, based on the recognition results, forming a scheduling plan includes the following steps:

[0141] Based on historical data, obtain the target treatment plan for the abnormal cause corresponding to the target abnormal feature;

[0142] Use the target treatment plan to regulate the processing equipment;

[0143] Monitor the output speed of lotus leaves of the processed processing equipment. When the output speed of lotus leaves is 0, it is determined that there is a fault in the lotus leaf transfer equipment corresponding to the processing equipment, and repair or replacement is carried out on the lotus leaf transfer equipment corresponding to the processing equipment;

[0144] When the output speed of lotus leaves is less than the consumption speed of lotus leaf products, taking the center of the processing equipment as the origin, establish coordinate systems in the normal image and the actual image after regulation in the same way;

[0145] Calculate the absolute value of the difference between the pixel values of the pixel points with the same coordinates in the processing equipment of the normal image and the processing equipment of the actual image after regulation, and accumulate to obtain the difference sum;

[0146] When the difference sum is greater than the preset gap, replace the processing equipment; otherwise, do nothing.

[0147] Here, the preset gap can be obtained from empirical data, that is, obtain the processing equipment that can work normally, and get the maximum allowable value of the gap as the preset gap.

[0148] Furthermore, this solution also proposes a storage medium, on which a computer-readable program is stored. When the computer-readable program is called, it executes the above-mentioned management method for processing and managing lotus leaves for glutinous rice chicken packaging based on artificial intelligence.

[0149] It can be understood that the storage medium can be a magnetic medium, such as a floppy disk, a hard disk, a magnetic tape; an optical medium such as a DVD; or a semiconductor medium such as a solid-state drive (SSD).

[0150] In summary, the advantages of the present invention are as follows: By obtaining the associated requirements of the processing steps, correcting the processing steps, forming the lotus leaf processing scale of the processing steps, forming an abnormal feature set for the processing steps, and forming a scheduling plan, it is possible to identify and target most of the possible abnormalities involved in the large-scale processing of lotus leaves. Due to the setting of multiple layers of recognition mechanisms, different types of abnormalities can be classified and recognized, thereby ensuring the accuracy of recognition, and then improving the accuracy of processing. And, in order to ensure the processing speed, the lotus leaf processing scale of the processing steps is formed, and different processing scales are set according to different processing steps, so as to ensure that the overall processing speed is consistent with the consumption speed of lotus leaf products, and then, without adding too many devices, meet the usage and consumption requirements of lotus leaves.

[0151] The foregoing has shown and described the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments, and 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 fall within the scope of the present invention claimed. The scope of protection required by the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for managing lotus leaf processing for glutinous rice and chicken packaging based on artificial intelligence, characterized in that: include: The lotus leaf processing process is decomposed into at least one processing step, and the lotus leaf finished product that meets the glutinous rice chicken packaging standard is obtained according to the sequence of at least one processing step, wherein the processing steps are lotus leaf screening, lotus leaf cleaning, lotus leaf drying and lotus leaf arranging; Obtain the consumption rate of the finished lotus leaf product packaged with glutinous rice and chicken, and obtain the appearance requirements of the finished lotus leaf product packaged with glutinous rice and chicken; Decompose the appearance requirements to obtain the associated requirements of the processing steps; Based on the associated requirements of the processing steps, the processing process of the processing steps is monitored, and based on the monitoring results, the processing steps are corrected; Based on the consumption rate of lotus leaf finished products, the lotus leaf processing scale of the processing steps is formed; An abnormal feature set is formed for the processing steps. Based on the abnormal feature set, the abnormal situations occurring in the processing steps are identified. Based on the identification results, a scheduling plan is formed, and the abnormalities are repaired according to the scheduling plan.

2. The method for managing lotus leaf processing for glutinous rice and chicken packaging based on artificial intelligence according to claim 1 is characterized in that: The appearance requirements for obtaining the lotus leaf finished product packaged with glutinous rice and chicken are as follows: The appearance requirements of the finished lotus leaf are the target shape of the lotus leaf, the target color of the lotus leaf, the allowable proportion of impurities on the lotus leaf surface, the target dryness of the lotus leaf and the allowable proportion of lotus leaf defects.

3. The method for managing lotus leaf processing for glutinous rice and chicken packaging based on artificial intelligence according to claim 2 is characterized in that: Decomposing the appearance requirements to obtain the associated requirements of the processing steps includes the following steps: The lotus leaf target shape, lotus leaf target color and lotus leaf defect allowable ratio are used as the associated requirements of the processing steps corresponding to the lotus leaf screening; The target color of lotus leaves, the allowable proportion of impurities on the lotus leaf surface and the allowable proportion of lotus leaf defects are used as the associated requirements for the processing steps of lotus leaf cleaning; The target dryness of lotus leaves is used as a correlation requirement for the processing step of drying lotus leaves; The lotus leaf target shape is taken as the associated requirement of the processing steps corresponding to the lotus leaf finishing.

4. The method for managing lotus leaf processing for glutinous rice and chicken packaging based on artificial intelligence according to claim 3 is characterized in that: The monitoring of the processing of the processing steps based on the associated requirements of the processing steps includes the following steps: Obtain sample images of finished lotus leaves that meet the associated requirements, extract features from the sample images according to the presentation characteristics of the associated requirements, and obtain sample image features that meet the associated requirements, wherein the dryness is identified using infrared imaging; Performing image recognition on the lotus leaf to be tested obtained through the processing step to obtain actual image features required by the association; Based on human eye recognition data, a preset ratio range is formed; When the difference ratio between the actual image features required for association and the sample image features falls within the preset ratio range, the lotus leaf to be tested is regarded as an unqualified lotus leaf; Calculate the proportion of unqualified lotus leaves in the lotus leaves to be tested as the characteristic proportion; When the feature ratio exceeds a preset value, the processing step corresponding to the associated requirement is taken as the processing step to be corrected, otherwise, no processing is performed, wherein the preset value is the allowable error value of the lotus leaf processing obtained based on experience.

5. The method for managing lotus leaf processing for glutinous rice and chicken packaging based on artificial intelligence according to claim 4 is characterized in that: The forming of a preset ratio range based on human eye recognition data comprises the following steps: Evenly divide the range of the difference ratio between the actual image feature and the sample image feature to obtain at least one recognition point; Acquire at least one image to be verified, the difference between which and the sample image is equal to the value at the identification point; When the human eye cannot recognize the sample image from the mixed image of the sample image and at least one image to be verified, the recognition point is used as the target recognition point; The interval formed by the maximum and minimum values ​​of the target recognition points is used as the preset ratio range.

6. The method for managing lotus leaf processing for glutinous rice and chicken packaging based on artificial intelligence according to claim 5 is characterized in that: The process of modifying the processing steps based on the monitoring results comprises the following steps: Acquire actual processing parameters of the processing step to be corrected, average the difference ratio between the actual image features associated with the processing step to be corrected and the sample image features, and obtain an average difference ratio; Obtain at least one processing device in the processing step to be corrected, each processing device performs automated processing of one lotus leaf; The proportion of unqualified lotus leaves processed by the processing equipment is counted as the target proportion, and the processing equipment whose target proportion exceeds the preset value is regarded as the processing equipment to be corrected; Accumulate the target ratios of the processing equipment to be corrected to obtain a comprehensive ratio; The average gap ratio is multiplied by the target ratio of the processing equipment to be corrected and divided by the comprehensive ratio to obtain the adjustment allocation ratio; The actual processing parameters are subtracted from the product of the actual processing parameters and the adjusted allocation ratio to obtain the corrected processing parameters; The actual processing parameters of the processing equipment to be corrected are adjusted to the corrected processing parameters.

7. The method for managing lotus leaf processing for glutinous rice and chicken packaging based on artificial intelligence according to claim 6 is characterized in that: The lotus leaf processing scale of forming the processing steps based on the consumption rate of the lotus leaf finished product includes the following steps: Count the time it takes for a single processing device to process a lotus leaf in a processing step, and count the transfer time for a lotus leaf to be transferred to the next processing step in a processing step; 1 divided by the sum of the consumption time and the transmission time gives the monomer processing speed; The consumption rate of lotus leaf finished products is divided by the processing rate of the monomers to get the preliminary number; Calculate the average qualified rate of all processing steps; The processing steps are numbered in reverse order of the processing sequence, i.e., lotus leaf screening, lotus leaf cleaning, lotus leaf drying, and lotus leaf arranging are numbered 4, 3, 2, and 1 respectively; Using the processing scale formula, the final number is calculated. The lotus leaf processing scale of the processing step is the parallel operation of the processing equipment with the final number in the processing step; The processing scale formula is as follows: Among them, A is the final number, a is the preliminary number, b is the average qualified rate, and n is the number of the processing step.

8. The method for managing lotus leaf processing for glutinous rice and chicken packaging based on artificial intelligence according to claim 7 is characterized in that: The forming of an abnormal feature set for the processing step comprises the following steps: When the speed of producing lotus leaves in a processing step is lower than the speed of consumption of finished lotus leaves, the processing step is regarded as an abnormal processing step; Based on historical data, a normal image of the processing equipment of the abnormal processing step is obtained, and the action parameters of the processing equipment in the normal image are obtained; Performing difference analysis on a normal image of a processing device at an abnormal processing step and an actual image of a processing device at an abnormal processing step to obtain an abnormal feature set, wherein the abnormal feature set is composed of abnormal features; Obtain at least one abnormal cause causing the abnormality, count the frequency of occurrence of the abnormal cause that appears simultaneously with the abnormal feature, take the abnormal cause with the largest frequency of occurrence as the abnormal cause causing the abnormal feature, and pair the abnormal feature with the abnormal cause causing the abnormal feature.

9. The method for managing lotus leaf processing for glutinous rice and chicken packaging based on artificial intelligence according to claim 8 is characterized in that: The method of identifying abnormal situations occurring in the processing steps based on the abnormal feature set includes the following steps: Taking the center of the processing equipment as the origin, a coordinate system is established in the normal image and the abnormal image of the abnormal situation in the same manner; Performing a difference comparison between the normal image and the abnormal image to obtain a difference portion; The abnormal features appearing in the difference part are obtained as the target abnormal features.

10. The method for managing lotus leaf processing for glutinous rice and chicken packaging based on artificial intelligence according to claim 9, characterized in that: The forming of a scheduling scheme based on the identification result comprises the following steps: Based on historical data, obtain the target processing solution for the abnormal cause corresponding to the target abnormal feature; Use targeted treatment plans to control processing equipment; The lotus leaf output speed of the regulated processing equipment is monitored. When the lotus leaf output speed is 0, it is determined that a lotus leaf transfer device corresponding to the processing equipment has a fault, and the lotus leaf transfer device corresponding to the processing equipment is repaired or replaced. When the output speed of lotus leaves is lower than the consumption speed of finished lotus leaves, the center of the processing equipment is taken as the origin, and the coordinate system is established in the normal image and the actual image after adjustment in the same way; Calculate the absolute value of the difference between the pixel values ​​of the pixel points at the same coordinates in the normal image processing device and the actual image processing device after adjustment, and add them up to obtain the difference sum; When the difference is greater than the preset gap, the processing equipment is replaced, otherwise, no action is taken.

Images