Intelligent red date sorting and distributing system

Abstract

The invention provides an intelligent red date sorting and shunting method and system, and relates to the technical field of image data analysis and processing, and the method comprises the steps: collecting image data of red dates; extracting size and shape features of the red dates from the image data of the red dates, comparing the extracted features with a preset appearance standard, and preliminarily screening out the red dates meeting the requirements of the first red date processed product as raw materials of the first red date processed product; and the red dates which do not meet the requirements of the first red date processed product are classified as raw materials of the second red date processed product, so that the problems of low sorting efficiency and high labor cost in the prior art are solved, and the problems of large fluctuation of product taste and physical characteristics and unstable quality in the deep processing process of the red dates are solved.

Description

Technical Field

[0001] This application relates to the technical field of image data analysis and processing, and particularly to an intelligent red date sorting and diversion system. Background Art

[0002] In the field of red date processing, it is often necessary to classify and screen red dates according to the requirements of different processed products. For example, when producing products such as red date sandwiched with walnut and red date paste pie, it is necessary to divert a batch of obtained red dates according to specific standards. However, the applicant found during the production process that the following problems generally exist in the existing technologies: First: The existing sorting technologies mainly rely on manual classification according to the size, appearance and other characteristics of red dates. Long-term use not only leads to high labor costs, but also has the problem of low sorting efficiency; Second: In the production of processed products such as red date sandwiched with walnut and red date paste pie, even if the same processing technology is adopted to process the same batch of red dates, the final finished products will still show large differences in taste and physical properties, resulting in large fluctuations in product quality and poor stability, and it is difficult to meet the requirements of high-quality products. Summary of the Invention

[0003] The purpose of this application is to provide an intelligent red date sorting and diversion system to solve the problems of low sorting efficiency and high cost existing in the existing technologies, as well as the problems that the taste and physical properties of its products still have large fluctuations, resulting in unstable product quality; In view of the above problems, this application provides an intelligent red date sorting and diversion method and system. Through intelligent detection and diversion of red date pictures, and precise grading in combination with characteristics such as the water content, dryness degree, and size of red dates, it can dynamically adjust the proportion of red dates at each drying level according to the requirements of different processed products, realize efficient and stable sorting and ratio control, ensure stable product quality of processing, and improve the level of production automation. The specific solutions are as follows; An intelligent red date sorting and diversion method includes: Collect image data of red dates, which can be collected from the front, side and partial close-up angles of red dates by using a high-resolution industrial camera under standardized lighting and background conditions; Extract the size and shape characteristics of red dates from the image data of red dates, and compare the extracted characteristics with the preset appearance standards, and initially screen out the red dates that meet the requirements of the first red date processed product and classify them as the raw materials of the first red date processed product; Classify the red dates that do not meet the requirements of the first red date processed product as the raw materials of the second red date processed product.

[0004] Among them, the method for extracting the size and shape characteristics of red dates from the image data of red dates is as follows; Preprocess the image data of red dates through image preprocessing techniques (such as image segmentation and contrast enhancement) to eliminate the influence of uneven illumination and background interference in the production environment; Use image processing algorithms (convolutional neural network (CNN)) to automatically extract key features such as the size and shape of red dates.

[0005] Furthermore, the method also includes; Obtain the moisture content of the raw materials of the first red date processed product; Obtain the standard moisture content of the first red date processed product based on the first red date processed product; Classify the red dates in the raw materials of the first red date processed product that do not meet the standard moisture content of the first red date processed product as the raw materials of the second red date processed product.

[0006] Furthermore, the method for obtaining the standard moisture content of the first red date processed product based on the first red date processed product is as follows; Obtain the index data related to the taste of the first red date processed product, and this data can be obtained through laboratory sensory testing or consumer research; Perform multi-source data fusion on the moisture content data of the raw materials of the first red date processed product and the taste index data; Use a deep learning model (such as LSTM or Transformer) to establish a correlation model between the drying degree of red dates and the taste, determine the range of the optimal moisture content corresponding to the best taste of red dates, and use it as the standard moisture content of the first red date processed product.

[0007] Furthermore, the method also includes; Obtain the ratio between the raw materials of the first red date processed product and the raw materials of the second red date processed product; If the ratio is lower than the preset value, feedback the drying parameter deviation data to the drying process control module to automatically adjust the drying parameters of the next batch of red dates.

[0008] Furthermore, the method also includes; Classify the red dates in the raw materials of the second red date processed product according to their drying degree; Obtain the ratio of each red date drying level required for the production of the second red date processed product based on the second red date processed product; Extract the corresponding red dates in the raw materials of the second red date processed product according to the ratio for processing the second red date processed product.

[0009] Furthermore, the method for obtaining the ratio of each red date drying level required for the production of the second red date processed product based on the second red date processed product is as follows; Obtain the index data related to the taste of the second red date processed product and the moisture content of the second red date processed product; Establish a correlation model between the taste of the second product and the water content, and determine the optimal ratio of red dates at different drying levels required in the production of the second product.

[0010] According to the grading results and the preset ratio requirements, automatically extract and mix red dates at each level to ensure the stability of the taste and quality of the second product.

[0011] Furthermore, the first red date processed product is red date with walnut, and the second red date processed product is red date paste pie.

[0012] An intelligent red date sorting and diversion system, comprising: Red date information collection module: responsible for obtaining the key information of red dates; Data analysis and determination module: used to analyze the characteristic data of red dates and determine the drying level and sorting category of each red date; Intelligent diversion module: According to the analysis results of the data analysis and determination module, conduct diversion operations on red dates, so that red dates with different drying levels and sizes enter different processing lines respectively; Raw material ratio adjustment module: used to intelligently calculate the optimal ratio of red dates at different drying levels according to processing requirements and guide the diversion strategy; Production control and data recording module: used to monitor and record the entire sorting and diversion process.

[0013] One or more technical solutions provided in this application have at least the following technical effects or advantages: This application replaces the traditional manual sorting method through intelligent image recognition, water content prediction and automatic grading technology, realizes continuous and precise high-efficiency sorting, significantly reduces manual participation, and reduces labor costs; This application aims at the special needs of different red date processed products (such as red date with walnut, red date paste pie, etc.). By establishing a correlation model between the product taste and red date characteristics (such as water content, drying level), dynamically adjust the usage ratio of red dates at different drying levels to ensure consistent taste and stable structure of the finished product, and reduce quality fluctuations; This application realizes real-time monitoring and recording of red date characteristics through full-process data collection and feedback adjustment, and can continuously optimize the sorting model and diversion strategy to ensure product consistency and high-quality output in long-term batch production; The above description is only an overview of the technical solution of this application. In order to understand the technical means of this application more clearly, it can be implemented according to the content of the specification. And in order to make the above and other purposes, features and advantages of this application more obvious and understandable, the following specific embodiments of this application are specifically given. It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of this application, nor is it used to limit the scope of this application. Other features of this application will become easily understood through the following specification. Brief Description of the Drawings

[0014] In order to more clearly illustrate the technical solutions in this application or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only exemplary. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained according to the provided drawings.

[0015] Figure 1 It is a schematic flow chart of an intelligent red date sorting and diversion method of this application; Figure 2 It is a schematic structural diagram of an intelligent red date sorting and diversion system of this application; Description of the reference numerals: Red date information acquisition module 11, data analysis and determination module 12, intelligent diversion module 13, raw material ratio adjustment module 14, production control and data recording module 15. Detailed Embodiments

[0016] This application provides an intelligent red date sorting and diversion system to solve the problems of low sorting efficiency and high labor cost in the prior art, as well as the problems of large fluctuations in product taste and physical properties and unstable quality during the deep processing of red dates. By analyzing the images of red dates, it can accurately know the proportion of red dates of different sizes in this batch of red dates, and preliminarily screen the red dates according to their sizes. To save costs, the preliminary screening can also be carried out by using equipment such as vibrating screens or drum screens to classify the red dates with sieves of different apertures. After obtaining the raw materials for the first red date processing product that meet the size requirements, obtain the drying condition of the raw materials for the first red date processing product and the proportion of its drying compliance, and conduct a secondary screening, so that not only the consistency of red date product processing can be ensured through the screening of moisture content, but also the quality fluctuation of red date products can be greatly reduced, enabling better adjustment of other parameters in the red date products to further improve the quality of red date products. Next, the technical solutions in this application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of this application, rather than all embodiments of this application. It should be understood that this application is not limited by the example embodiments described here. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of this application. Additionally, it should be noted that for the sake of convenience of description, only parts related to this application are shown in the accompanying drawings, rather than all of them.

[0017] Example 1. Please refer to the attached Figure 1 , this application provides an intelligent red date sorting and diversion system. Among them, the intelligent red date sorting and diversion method includes: An intelligent red date sorting and diversion method includes: S100 Collect image data of red dates; Specifically, it can collect red date images from the front, side, and partial close-up angles simultaneously by using a high-resolution industrial camera under standardized lighting and background conditions to ensure that the overall appearance, size, shape, and local detail features of the red dates can be completely captured; Store the collected image data in a dedicated server or cloud platform, and record the detailed metadata of the collection (such as collection time, camera number, shooting angle, etc.) for subsequent data management and traceability; S200 Extract the size and shape features of the red dates from the image data of the red dates, and compare the extracted features with the preset appearance standards, and preliminarily screen out the red dates that meet the requirements of the first red date processing product and classify them as the raw materials for the first red date processing product.

[0018] Specifically: S200 includes the following steps: S210 preprocesses the image data of red dates through image preprocessing techniques (such as image segmentation and contrast enhancement) to eliminate the influence of uneven illumination and background interference in the production environment; That is: Segment the collected image by using edge detection or region segmentation techniques to clearly outline the contours of red dates, so that the red dates can be separated from the background to eliminate the parts in the background that may interfere with feature extraction; By using techniques such as contrast enhancement, the details of red dates can be made more obvious. The enhancement process can improve the image blurring problems caused by uneven illumination, shadows or reflections, making the edges, textures and shapes of red dates clearer; And perform illumination correction processing on the image to adjust the brightness and color balance. In addition, denoising algorithms are used to filter out random noise in the image to further improve the image quality; Normalize the preprocessed image. By scaling or cropping the image, make all images consistent in size and resolution to facilitate the stable operation of subsequent algorithms; S220 automatically extracts key features such as the size and shape of red dates by using image processing algorithms (convolutional neural network (CNN)); That is: This application can automatically extract key features in the red date image by using a pre-trained or specially fine-tuned CNN model. The model will identify the edges, contours, shapes and local detail features of red dates, thus forming a feature vector describing the size and morphology of red dates; Quantify the image features extracted by CNN and organize them into data records in a unified format. For example, record information such as the aspect ratio, contour curvature, and area size of red dates; Compare the extracted red date features with the standards of the previously set requirements for the first red date processed product (such as shape, size range, symmetry, etc.). During the comparison process, use the set threshold or scoring standard to determine whether the current red date meets the requirements of the first product; If the features of the red dates are within the allowable range, it is determined to meet the requirements of the first red date processed product, and it is classified as the raw material for the first product; If not, it is directly classified as the raw material for the second red date processed product, and further processing and grading will be carried out later.

[0019] S300 obtains the moisture content of the raw material for the first red date processed product; Specifically, the step S300 includes: S310 obtains the image information of the raw material for the first red date processed product and preprocesses its image; That is: obtain the image data of each red date from the raw material for the first red date processed product; Perform noise removal, illumination correction, contrast enhancement, and color balance processing on the collected images to ensure that information such as the color, texture, and reflectance of red dates in the images is clear and uniform, providing high-quality input data for subsequent feature extraction. The specific method can refer to step S100; S320 Extract color, texture, and reflectance features related to the moisture content from the image information of the preprocessed raw materials of the first red date processing product using a convolutional neural network (CNN); That is: Apply the CNN algorithm to the preprocessed images to automatically extract image features related to the moisture content. These features include the color distribution, texture details, and reflectance information on the surface of the red dates, all of which can reflect the change in the moisture content of the red dates; Perform normalization and standardization processing on the extracted features to ensure that the features between different samples are compared and analyzed on the same scale; S330 Train the corresponding regression model; That is: By collecting a certain number of red date samples and providing real moisture content data for these samples (which can obtain calibration data through precise detection methods in the laboratory); Adopt a combined structure of a convolutional neural network and a fully connected layer to construct a regression model for mapping the extracted image features to the actual moisture content value; Use the prepared sample data to train the regression model and evaluate the prediction accuracy of the model through methods such as cross-validation. To improve the robustness of the model, techniques such as data augmentation and regularization can be used to continuously optimize the model performance, enabling the model to adapt to batch data processing, detect the moisture content of each red date one by one on the production line, and ensure that each red date obtains a separate prediction value; S340 Use the trained regression model (such as CNN or its combination with a fully connected layer) to predict the moisture content of each red date according to the extracted features; That is: In this step, the trained regression model is used to perform real-time prediction on each red date image to generate the corresponding moisture content value; S350 Record the predicted moisture content of each red date, store the relevant data, and feedback it to the production control system; That is: In this step, the predicted moisture content data of each red date is stored in the database together with the corresponding image information, collection time, sample number, and other metadata for easy tracking and subsequent data analysis; Its detection data is real-time feedback to the production control system as the basis for drying process adjustment and diversion decision-making. If it is detected that the moisture content of a certain batch of red dates is too high or too low, the system can automatically issue a warning and adjust the subsequent drying parameters to ensure the stability of product quality; The S400 obtains the standard water content of the first red date processed product based on the first red date processed product; Among them, the step S400 includes the following steps; S410 obtains the index data related to the taste of the first red date processed product, and this data can be obtained through laboratory sensory testing or consumer research methods; Specifically: It can obtain data related to the taste evaluation of the first red date processed product by collecting data such as consumer tasting scores and sensory evaluation experiment data.

[0020] The taste-related index data can cover aspects such as sweetness, softness, chewiness, fineness, and moisture content, and is obtained through laboratory testers or user feedback surveys.

[0021] Synchronously record the water content data of the corresponding red dates in different batch samples; S420 performs multi-source data fusion on the water content data of the raw materials of the first red date processed product and the taste index data; Specifically: This step matches and fuses the taste-related index data with the water content detection data of the corresponding red date samples to establish a multi-source data set; Normalize and standardize the taste data and the water content data to ensure that the two types of data are effectively correlated on the same scale; In addition, background information such as the source of red dates, picking batches, and drying process parameters can be added during the data fusion process to further enhance the generalization ability of the model; S430 uses a deep learning model (such as LSTM or Transformer) to establish an association model between the drying degree of red dates and the taste, determine the range of the best water content corresponding to the best taste of red dates, and use it as the standard water content of the first red date processed product; Specifically, this step uses a deep learning algorithm (such as the long short-term memory network LSTM or the Transformer network with self-attention mechanism) to model the fused multi-source data; The model automatically extracts the optimal feature representation by learning the non-linear relationship between different water contents and taste evaluations, and finds the best water content interval; Through the results obtained from model training, identify the range of water content corresponding to the best taste of red dates, and set this range as the standard water content of the first red date processed product; S440 samples the red date products of different batches that reach the standard water content, and invites professional tasting personnel or consumers to conduct a re-sensory evaluation to obtain the sensory evaluation results, so as to verify whether the range of the standard water content obtained by the model is stable and effective; Dynamically correct the standard moisture content range according to user feedback or factors such as season, climate, and changes in the source of red dates, ensuring its continuous adaptation to market demand and actual production conditions; Regularly update the verification and correction results to the model to form a cyclic iterative optimization mechanism, maintaining the long-term accuracy and adaptability of the standard moisture content parameters; S500 classifies the red dates in the raw materials of the first red date processing product that do not meet the standard moisture content of the first red date processing product as the raw materials of the second red date processing product; Specifically, the step S500 further includes: S510 calls the standard moisture content range parameters of the first red date processing product for the predicted moisture content data of each red date obtained in the previous steps.

[0022] Judges the moisture content of each red date one by one to determine whether it falls between the upper and lower limits of the standard moisture content.

[0023] Mark the judgment results in the traceability information of each red date for subsequent diversion and recording; S520 classifies and identifies the red dates with moisture content lower than the lower limit of the standard range (over-dry) or higher than the upper limit of the standard range (over-wet).

[0024] And use mechanical or air separation devices to remove the non-conforming red dates from the raw materials of the first red date processing product according to the moisture content judgment results of the red dates, and convey them to the raw material channel of the second red date processing product to ensure the stability and efficiency of the entire removal and diversion process; S530 conducts quantity and quality statistics on the remaining red dates in the raw materials of the first red date processing product to obtain statistical results; S600 obtains the ratio between the raw materials of the first red date processing product and the raw materials of the second red date processing product; If the ratio is lower than the preset value, feedback the drying parameter deviation data to the drying process control module to automatically adjust the drying parameters of the next batch of red dates; Specifically, the step S600 further includes S610 respectively counts the quantity and total weight of the red dates entering the raw material area of the first red date processing product and the quantity and total weight of the red dates entering the raw material area of the second red date processing product according to the classification data of the red dates in the previous diversion process; S620 calculates the ratio between the raw materials of the first red date processing product and the raw materials of the second red date processing product, which can be based on the quantity ratio, weight ratio, or a combination of both; For example, the quantity ratio is the quantity of the raw materials of the first red date processing product divided by the total quantity of red dates, and the weight ratio is the weight of the raw materials of the first red date processing product divided by the total weight of red dates.

[0025] S630 compares the calculated ratio with the reasonable ratio threshold preset in the system; If the ratio is lower than the preset value, it indicates that the drying effect of the current batch of red dates is insufficient (possibly due to high water content leading to an increase in the unqualified rate); If the ratio is higher than or close to the preset value, it means that the drying process control is reasonable, and the existing parameters are maintained; S640 When the ratio is lower than the preset threshold, the drying deviation information (such as batch number, ratio data, abnormal type, time, etc.) is fed back to the drying process control module in real time.

[0026] Based on the feedback information, the drying process control module automatically adjusts the drying process parameters of the next batch of red dates, such as increasing the drying temperature, extending the drying time, optimizing the wind speed, or adjusting the humidity control, etc., to reduce the generation of red dates with excessive water content; S650 Synchronously stores information such as ratio data, feedback adjustment records, and drying process adjustment plans in the production control system database to form a complete production traceability chain, facilitating subsequent analysis of the relationship between the drying effect and the diversion ratio of each batch of red dates; S700 Classifies the red dates in the raw materials of the second red date processing product according to their drying degree; Obtains the ratio of each red date drying level required for the production of the second red date processing product; Extracts the corresponding red dates in the raw materials of the second red date processing product according to the ratio for processing the second red date processing product.

[0027] Specifically: The specific steps of step S700 are as follows; S710 Collects the image information of the raw materials of the second red date processing product and preprocesses the image information. The specific method can refer to the method in step S100; S720 Uses a convolutional neural network (CNN) or other deep learning algorithms to extract feature information related to the drying degree from the image, such as changes in epidermal color, shrinkage degree, glossiness, and surface texture state, etc.; S730 Obtains known sample data of drying degrees. Based on a large amount of known sample data of drying degrees, trains a drying degree determination model to classify red dates into multiple drying levels (such as first-level drying, second-level drying, third-level drying, etc.). The model can use algorithms such as classification networks, support vector machines, or decision trees to achieve rapid classification and determination of the drying state of red dates; S740 Applies the drying degree determination model to identify the drying level of each red date and classifies the red dates into the corresponding drying level areas or containers according to the determination results; S750 Analyzes the optimal ratio relationship of red dates at each drying level in the second red date processing product by combining historical product quality data, consumer taste feedback, laboratory tests, etc.

[0028] Based on the experience of multi-batch production, optimize and determine the drying level ratio standard for the second processed product (e.g., 60% for primary drying, 30% for secondary drying, and 10% for tertiary drying). S760 automatically calculates whether the proportion of red dates at each level available for distribution meets the preset proportion requirements based on the quantity and weight of the graded red dates.

[0029] If the proportion requirements are met, accurately extract red dates at each level according to the proportion for mixing and put them into the production of the second red date processed product.

[0030] If the proportion is insufficient, issue a replenishment or adjustment notice to prompt manual or system-based allocation of raw materials from other batches for supplementation; S770 can regularly analyze data such as the ex-factory quality, taste performance, and market feedback of the second red date processed product, and dynamically optimize the drying level ratio standard based on the preset proportion; When significant fluctuations occur in the characteristics of red date raw materials (such as origin differences and seasonal variations), promptly adjust the grading determination standard or processing proportion to ensure the stability of the final product; S780 completely records information such as the proportion, source batch, drying parameters, and grading data of red dates at each drying level used for each batch of the second red date processed product, establishes a digital traceability chain, so that if product quality anomalies occur subsequently, it can quickly trace back to the specific red date batch and grading status, and accordingly improve the production process or adjust the raw material selection.

[0031] Among them, the method for obtaining the proportion of each red date drying level required for the production of the second red date processed product is as follows; Obtain the actual production data of the second red date processed product, including data on taste-related indicators (such as softness, elasticity, moisture content, chewiness, etc.) of products in different batches and the corresponding overall moisture content information; Obtain the drying grading proportion information of the raw material red dates for each batch corresponding to the above production data, specifically including the usage proportion of red dates at different drying levels in each batch; Adopt data analysis methods to establish an association model between the taste indicators and overall moisture content of the second red date processed product and the usage proportion of red dates at different drying levels, forming a mathematical relationship or data model between the taste of the second red date processed product and the red date drying level ratio; Train and optimize the model, continuously correct the model parameters using multi-batch historical data to ensure that the model can accurately reflect the impact of the usage proportion of red dates at different drying levels on the taste of the second red date processed product; Determine the best drying level red date ratio plan for different taste requirements (such as softer or more chewy) based on the model results, providing a decision-making basis for subsequent adjustment of red date raw material allocation; Integrate the finally optimized model into the production control system to achieve intelligent recommendation and dynamic adjustment of the red date raw material ratio for the second red date processed product.

[0032] Furthermore, the first red date processed product is red date with walnut, and the second red date processed product is red date and walnut bar. The red date and walnut bar is a healthy snack with red dates as the outer layer and walnuts wrapped inside, and it has a distinct three-layer structure: Upper layer: red date layer, providing natural sweetness and a soft texture.

[0033] Middle layer: walnut layer, bringing the crispy taste of nuts and rich nutrition.

[0034] Lower layer: red date layer, jointly wrapping the walnuts with the upper layer to increase the overall coordination.

[0035] Example 2: Based on the same inventive concept as an intelligent red date sorting and shunting method in the foregoing example, as Figure 2 shown, the present application provides an intelligent red date sorting and shunting system, wherein the system includes: Red date information acquisition module 11: responsible for obtaining key information of red dates; The red date information acquisition module 11 includes: Image acquisition unit: used to obtain high-definition images of single red dates by using devices such as industrial cameras; Water content estimation unit: estimate the water content of each red date through image feature extraction and model analysis.

[0036] Data analysis and determination module 12: used to analyze the characteristic data of red dates and determine the drying level and sorting category of each red date; The data analysis and determination module 12 includes: Image preprocessing unit: used to perform denoising, correction, etc. on red date images; Feature extraction unit: extract water content-related features such as color, texture, and reflectivity by using a convolutional neural network (CNN); Water content prediction unit: predict the water content of single red dates based on a trained regression model; Drying level determination unit: classify red dates into different drying levels according to the water content range; Size level determination unit: classify red dates by size according to the image analysis results.

[0037] Intelligent shunting module 13: according to the analysis results of the data analysis and determination module 12, perform shunting operations on red dates so that red dates with different drying levels and sizes enter different processing lines respectively; The intelligent shunting module 13 includes: Shunt control unit: Receives the determination result and controls shunt mechanisms such as the robotic arm, slideway, air flow, etc.

[0038] Shunt execution unit: Actually completes the sorting and classification of red dates and sends them into the corresponding paths.

[0039] Raw material ratio adjustment module 14: Used to intelligently calculate the optimal ratio of red dates at different drying levels according to processing requirements and guide the shunt strategy; The raw material ratio adjustment module 14 includes: Product demand analysis unit: Used to receive demand parameters such as the taste target and water content target of the second red date processed product; Taste and ratio modeling unit: Based on historical data, establishes a relationship model between the taste of products such as date paste and the ratio of drying levels; Optimal ratio calculation unit: According to the model, outputs the target ratio of red dates at each drying level; Dynamic ratio feedback unit: Real-time feedbacks the ratio information to the shunt control unit to adjust the shunt ratio.

[0040] Production control and data recording module 15: Used to monitor and record the entire picking and shunting process; The production control and data recording module 15 further includes; Process monitoring unit: Real-time monitors the operating status of each unit to ensure system stability.

[0041] Data storage unit: Stores the detection data, grading results, shunt paths, etc. of each batch of red dates.

[0042] Analysis and optimization unit: Periodically analyzes the production data to optimize the sorting model and ratio model.

[0043] Feedback control unit: Adjusts parameters such as information collection, determination, and shunting according to the analysis results.

[0044] Any of the above methods or steps can be stored as computer instructions or programs in various types of computer memories, and the computer instructions or programs are recognized by various types of computer processors, thereby implementing any of the above methods or steps; Secondly, the above description of the disclosed embodiments enables those skilled in the art to implement or use this application. Various modifications to these embodiments will be obvious to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application will not be limited to these embodiments shown herein, but will be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0045] Obviously, those skilled in the art can make various changes and modifications to this application without departing from the spirit and scope of this application. Thus, if these modifications and variations of this application fall within the scope of this application and its equivalent technologies, this application is also intended to include these changes and modifications.

Claims

1. An intelligent red date picking and diversion method, characterized in that: include: Collect image data of red dates; Extracting size and shape features of the red dates from the image data of the red dates, and comparing the extracted features with preset appearance standards, preliminarily screening out the red dates that meet the requirements of the first red date processing product as raw materials for the first red date processing product; Red dates that do not meet the requirements of the first red date processing product will be classified as raw materials for the second red date processing product.

2. The intelligent red date picking and diversion method according to claim 1 is characterized in that: The method for extracting the size and shape features of red dates from the image data of red dates is as follows; The image data of red dates are preprocessed by using image preprocessing technology; Image processing algorithms are used to automatically extract key features of red dates such as size and shape.

3. The intelligent red date picking and diversion method according to claim 1 is characterized in that ； Obtaining the water content of the raw material of the first red date processing product; Obtaining the standard moisture content of the first processed red dates product according to the first processed red dates product; The red dates in the raw materials of the first red dates processed product that do not meet the moisture content standard of the first red dates processed product shall be classified as raw materials for the second red dates processed product.

4. The intelligent red date picking and diversion method according to claim 3 is characterized in that: The method for obtaining the standard moisture content of the first processed red dates product according to the first processed red dates product is as follows; Obtain the taste-related index data of the first red date processed products; Perform multi-source data fusion on the water content data and taste index data of the raw materials of the first red date processed product; Using a deep learning model, a correlation model between the dryness and taste of red dates was established, and the optimal moisture content range corresponding to the best taste of red dates was determined, which was used as the standard moisture content of the first red date processed product.

5. The intelligent red date picking and diversion method according to claim 3 is characterized in that ； Obtaining the ratio between the raw material of the first red date processed product and the raw material of the second red date processed product; If the ratio is lower than the preset value, the drying parameter deviation data will be fed back to the drying process control module to automatically adjust the drying parameters of the next batch of red dates.

6. The intelligent red date picking and diversion method according to claim 3 is characterized in that ； Classifying the red dates in the raw material of the second red date processed product according to their degree of dryness; The ratio of each red date drying level required for the production of the second red date processed product is obtained; The corresponding red dates in the raw materials of the second red date processed product are extracted according to the proportion to process the second red date processed product.

7. The intelligent red date picking and diversion method according to claim 3 is characterized in that: The method for obtaining the ratio of each red date drying level required for the production of the second red date processed product is as follows; Obtaining taste-related index data of the second processed red dates product and the water content of the second processed red dates product; Establish a correlation model between the taste and moisture content of the second product, and determine the optimal ratio of red dates of different drying levels in the production of the second product; According to the grading results and preset proportion requirements, red dates of different grades are automatically extracted and proportioned.

8. An intelligent red date picking and diversion method according to any one of claims 1 to 7, characterized in that: The first processed red dates product is walnut-stuffed dates, and the second processed red dates product is jujube kernel pie.

9. An intelligent red date picking and diversion system, the system is used to execute the method according to any one of claims 1 to 7, characterized in that: include: Red dates information collection module: responsible for obtaining key information of red dates; Data analysis and determination module: used to analyze the characteristic data of red dates and determine the drying level and sorting category of each red date; Intelligent diversion module: According to the analysis results of the data analysis and judgment module, the red dates are diverted so that the red dates of different drying levels and sizes enter different processing lines; Raw material ratio adjustment module: used to intelligently calculate the optimal ratio of red dates of different drying levels according to processing requirements and guide the diversion strategy; Production control and data recording module: used to monitor and record the entire picking and diversion process.

Images