Intelligent screening and extracting technology for functional food raw materials

Abstract

The invention discloses an intelligent screening and extracting technology for functional food raw materials, and through the intelligent screening and extracting technology, active ingredients in the functional food raw materials can be efficiently and accurately screened and extracted, the extracting efficiency is improved, the cost is reduced, and meanwhile damage to the active ingredients is reduced. Besides, through an intelligent control system and data analysis, the stability and repeatability of the extraction process can be ensured, the quality and reliability of the product are improved, and the problems of low intelligent degree, poor flexibility and weak data processing capability in the screening and extraction process of the existing functional food raw materials are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of food raw material screening and extraction, and in particular to a functional food raw material intelligent screening and extraction technology. Background Art

[0002] In recent years, as people's living standards have improved, they have become more and more concerned about health and nutrition, and the demand for functional foods has gradually increased. Functional food ingredients refer to those food ingredients that contain specific bioactive ingredients and can have a positive impact on human health. These ingredients not only provide basic nutritional needs, but also have additional functions such as preventing diseases and promoting health.

[0003] Traditional methods for screening and extracting functional food raw materials mainly rely on manual operation and empirical judgment, and have the following problems: low efficiency, high cost, poor accuracy, and poor repeatability. Although some automated equipment and technologies have been introduced into the screening and extraction process of food raw materials in recent years, these technologies still have many shortcomings: low intelligence, poor flexibility, and weak data processing capabilities.

[0004] Therefore, it is necessary to develop an efficient and accurate intelligent screening and extraction technology to improve the screening and extraction efficiency of functional food raw materials, reduce costs, and achieve more efficient and intelligent production. Summary of the invention

[0005] The purpose of this section is to summarize some aspects of embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the specification abstract and the invention title of this application to avoid blurring the purpose of this section, the specification abstract and the invention title, and such simplifications or omissions cannot be used to limit the scope of the present invention.

[0006] In view of the above-mentioned problems existing in the screening and extraction of existing functional food raw materials, the present invention is proposed.

[0007] Therefore, the technical problem solved by the present invention is to solve the problems of low intelligence, poor flexibility and weak data processing capability in the screening and extraction processes of existing functional food raw materials.

[0008] To solve the above technical problems, the present invention provides the following technical solutions: a functional food raw material intelligent screening and extraction technology, including intelligent screening, intelligent extraction, separation and purification, data analysis, and quality control, and its specific steps are as follows: S1: collecting food raw materials; S2: using intelligent means to quickly and accurately identify and intelligently screen food raw materials; S3: according to the intelligent screening results, using enzymatic hydrolysis technology to extract the raw materials; S4: efficiently separating and purifying the extracted functional food raw materials to determine their functionality and the content of active ingredients; S5: quality classification monitoring of the entire screening and extraction process.

[0009] As a preferred solution of the functional food raw material intelligent screening and extraction technology described in the present invention, the following steps are specifically included in the rapid and accurate identification and intelligent screening of food raw materials using intelligent means: S1: after collecting the food raw materials, real-time image collection of the food raw materials is performed through a high-resolution camera or image sensor; S2: acquiring point cloud data of the collected image, constructing a point cloud distribution rate value range acquisition model, and acquiring the distribution rate value range of each point cloud image; S3: acquiring the distribution rate value range standard value of each food raw material pre-stored in the memory; S4: comparing the point cloud distribution rate value range obtained in S2 with the distribution rate value range standard value obtained in S3; S5: identifying and intelligently screening the current food image based on the comparison result.

[0010] As a preferred solution of the functional food raw material intelligent screening and extraction technology described in the present invention, after real-time image acquisition of food raw materials by a high-resolution camera or image sensor, the acquired images are preprocessed; wherein the preprocessing steps specifically include: denoising, grayscale, binarization and normalization.

[0011] As a preferred solution of the functional food raw material intelligent screening and extraction technology of the present invention, the point cloud distribution rate range acquisition model constructed is specifically:

[0012]

[0013] Wherein, δ is the point cloud distribution rate range; S is the point cloud distribution area in the acquired point cloud distribution image, cm 2 ; G is the number of corners in the shape of the acquired point cloud distribution image, pieces; L is the total length of the point cloud in the acquired point cloud distribution image, cm; 1.41, 0.981, and 0.9811 are all adjustment constants; dx is the integral operation.

[0014] As a preferred solution of the functional food raw material intelligent screening and extraction technology described in the present invention, the point cloud distribution rate range obtained in S2 is compared with the distribution rate range standard value obtained in S3 according to the following formula:

[0015]

[0016] Wherein, η is the comparison result value; δ is the point cloud distribution rate range; δ 标 is the standard value of the distribution rate range; 2.61 is the adjustment constant.

[0017] As a preferred solution of the functional food raw material intelligent screening and extraction technology described in the present invention, when the comparison result value is less than or equal to a threshold value, it is determined that the category of the current food image is consistent.

[0018] As a preferred solution of the functional food raw material intelligent screening and extraction technology described in the present invention, the threshold value is defined as 10.12%.

[0019] Beneficial effects of the present invention: The present invention provides a functional food raw material intelligent screening and extraction technology, through which the active ingredients in the functional food raw materials can be efficiently and accurately screened and extracted, the extraction efficiency can be improved, the cost can be reduced, and the damage to the active ingredients can be reduced. In addition, through the intelligent control system and data analysis, the stability and repeatability of the extraction process can be ensured, the quality and reliability of the product can be improved, and the problems of low intelligence, poor flexibility, and weak data processing capabilities in the screening and extraction process of existing functional food raw materials can be solved. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. Among them:

[0021] Figure 1 This is a flow chart of the overall steps of the intelligent screening and extraction technology of functional food ingredients provided by the present invention.

[0022] Figure 2 A flowchart of the steps of using intelligent means to quickly and accurately identify and intelligently screen food raw materials provided by the present invention. DETAILED DESCRIPTION

[0023] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0024] Traditional methods for screening and extracting functional food raw materials mainly rely on manual operation and empirical judgment, and have the following problems: low efficiency, high cost, poor accuracy, and poor repeatability. Although some automated equipment and technologies have been introduced into the screening and extraction process of food raw materials in recent years, these technologies still have many shortcomings: low intelligence, poor flexibility, and weak data processing capabilities.

[0025] Therefore, please refer to Figure 1The present invention provides a functional food raw material intelligent screening and extraction technology, including intelligent screening, intelligent extraction, separation and purification, data analysis, and quality control. The specific steps are as follows:

[0026] S1: Collect food raw materials;

[0027] It should be noted that the collection of raw materials is a mature application of existing mature technologies and will not be elaborated on here.

[0028] S2: Use intelligent means to quickly and accurately identify and intelligently screen food ingredients;

[0029] Furthermore, the use of intelligent means to quickly and accurately identify and intelligently screen food raw materials specifically includes the following steps:

[0030] H1: After collecting food raw materials, real-time image acquisition of food raw materials is performed through a high-resolution camera or image sensor;

[0031] Furthermore, after real-time image acquisition of the food raw material by a high-resolution camera or image sensor, the acquired image is preprocessed;

[0032] The preprocessing steps specifically include: denoising, grayscale, binarization and normalization.

[0033] It should be noted that:

[0034] The collected images are preprocessed, including denoising, grayscale, binarization, normalization and other steps to improve image quality and processing speed.

[0035] Step 1: Denoising

[0036] Remove noise from the image and improve image quality. Common denoising methods include mean filtering, Gaussian filtering, median filtering, etc.

[0037] Mean filtering: By calculating the average value of each pixel in the image and the pixels in its neighborhood, the average value is used to replace the value of the original pixel to achieve the purpose of denoising;

[0038] Gaussian filtering: Use the Gaussian function to perform weighted averaging on the image, and perform weighted summation on the pixel and its neighborhood to reduce the impact of noise;

[0039] Median filtering: Select a window containing an odd number of points and use the median of the pixel grayscale values ​​in the window to replace the value of the central pixel. It is particularly suitable for removing salt and pepper noise.

[0040] Step 2: Grayscale

[0041] Convert color images to grayscale images to simplify image features and reduce data volume. Common grayscale conversion methods include average method, maximum method, minimum method and weighted average method.

[0042] Average value method: the average value of the three RGB components of each pixel of the color image is taken as the grayscale value;

[0043] Maximum value method: the maximum value among the three components of RGB is taken as the gray value;

[0044] Minimum method: take the minimum value of the three RGB components as the grayscale value;

[0045] Weighted average method: According to different application scenarios, different weights are assigned to the three components of RGB, and then the weighted average is calculated as the grayscale value.

[0046] Step 3: Binarization

[0047] Convert the image into a binary image, that is, a black and white image, to simplify image processing and analysis. Commonly used binarization methods include global thresholding, local thresholding, and adaptive thresholding.

[0048] Global threshold method: According to the overall characteristics of the image, a global threshold is selected to divide the pixel values ​​of the image into two levels, namely black and white;

[0049] Local threshold method: Calculate a local threshold for each pixel in the image, and divide the pixel value into two levels according to the local threshold;

[0050] Adaptive threshold method: The threshold of each pixel is dynamically calculated based on the pixel values ​​of the pixel and its neighborhood, thereby achieving adaptive binarization.

[0051] Step 4: Normalization

[0052] Select the normalization range.

[0053] Determines the normalized pixel value range, usually [0,1] or [0,255].

[0054] For grayscale images, [0,1] is usually selected as the normalization range; for color images, [0,255] can be selected as the normalization range.

[0055] Calculate the maximum and minimum pixel values:

[0056] Traverse each pixel in the image and calculate the maximum value (Max) and minimum value (Min) of the pixel.

[0057] These maxima and minima generally represent the overall contrast range of the image.

[0058] Calculate the normalization coefficient:

[0059] Calculate the normalization coefficient, which is the difference between the maximum value and the minimum value (Range = Max-Min).

[0060] The normalization coefficient is used to map the original pixel value to the normalized range.

[0061] Perform normalization calculations:

[0062] For each pixel in the image, subtract the minimum value from its original value and then divide it by the absolute value of the normalization coefficient (for symmetric normalization) or the difference between the maximum and minimum values ​​(for asymmetric normalization).

[0063] Use the following formula for normalization calculation:

[0064] \[NormalizedValue=\frac{OriginalValue-Min}{Range}\]

[0065] For asymmetric normalization, the following formula can be used:

[0066] \[NormalizedValue=\frac{OriginalValue-Min}{Max-Min}\]

[0067] Results check:

[0068] Check whether the normalized image meets the expected normalization range and make adjustments if necessary.

[0069] For asymmetric normalization, it is also necessary to check whether the maximum and minimum values ​​meet certain conditions, such as whether they are both outside the normalization range.

[0070] H2: Obtain the point cloud data of the collected image, build a point cloud distribution rate range acquisition model, and obtain the distribution rate range of each point cloud image;

[0071] The details are as follows:

[0072] Step 1: Point cloud data acquisition

[0073] ① Use laser scanners, structured light scanners, depth cameras or RGB-D cameras to scan the surface of objects and obtain information about the surface of objects;

[0074] ②Point cloud data usually contains a large number of points, each of which has its coordinates (X, Y, Z) in three-dimensional space, and possible color information (RGB).

[0075] Step 2: Point cloud preprocessing

[0076] ① Noise removal: remove noise points in the point cloud through filtering algorithms (such as mean filtering, median filtering, etc.);

[0077] ②Point cloud registration: multiple point clouds are registered through algorithms to merge point cloud data from multiple perspectives;

[0078] ③ Point cloud simplification: Reduce the amount of data by reducing the number of points in the point cloud while retaining the geometric features of the object as much as possible.

[0079] Step 3: Construct a point cloud distribution rate range acquisition model

[0080] A specific mathematical model is selected to analyze the distribution characteristics of point cloud data. The point cloud distribution rate range acquisition model is as follows:

[0081]

[0082] Among them, δ is the point cloud distribution rate range; S is the point cloud distribution area in the acquired point cloud distribution image, cm 2 ; G is the number of corners in the shape of the acquired point cloud distribution image, pieces; L is the total length of the point cloud in the acquired point cloud distribution image, cm; 1.41, 0.981, and 0.9811 are all adjustment constants; dx is the integral operation.

[0083] Step 4: Get the distribution rate range of the point cloud image

[0084] ① Input the preprocessed point cloud data into the constructed model, and the model outputs the distribution rate value range of each point cloud image;

[0085] ②The distribution rate value range reflects the distribution of each point in the point cloud data.

[0086] Step 5: Model optimization and adjustment

[0087] ① Optimize the prediction performance of the model by continuously iterating and adjusting model parameters;

[0088] ② Use actual point cloud data to verify and test the model to ensure that the model can accurately obtain the distribution rate value range of the point cloud image.

[0089] H3: Obtain the distribution rate range standard value of each food raw material pre-stored in the memory;

[0090] It should be noted that the pre-stored standard values ​​of the distribution rate range of each food raw material are the distribution rate ranges obtained in advance based on the standard raw materials, and the distribution rate ranges can be obtained by obtaining point cloud images of the raw materials in actual states in advance.

[0091] Obtain the distribution rate range standard value of each food raw material pre-stored in the memory, the specific details are as follows:

[0092] Step 1: Data collection and standardization

[0093] ① Collect standard point cloud data of various food ingredients and standardize these data through preprocessing steps (such as noise removal, registration, simplification, etc.);

[0094] ② Label and classify the standard point cloud data of each food raw material to ensure that each data sample has an accurate and specific distribution rate value range standard value.

[0095] Step 2: Data storage and management

[0096] ①Store the processed food raw material point cloud data and its distribution rate range standard value in a database or file system for quick retrieval and access;

[0097] ② Realize data security management and access control to ensure data security and integrity.

[0098] H4: Compare the point cloud distribution rate range obtained in S2 with the standard value of the distribution rate range obtained in S3;

[0099] Furthermore, the point cloud distribution rate range obtained in S2 is compared with the standard value of the distribution rate range obtained in S3 according to the following formula:

[0100]

[0101] Among them, η is the comparison result value; δ is the point cloud distribution rate range; δ 标 is the standard value of the distribution rate range; 2.61 is the adjustment constant.

[0102] When the comparison result value is less than or equal to the threshold, it is determined that the category of the current food image is consistent.

[0103] Specifically, the threshold is defined as 10.12%.

[0104] H5: Identify and intelligently screen the current food image based on the comparison results.

[0105] S3. Extract the raw materials using enzymatic hydrolysis technology according to the intelligent screening results;

[0106] It should be noted that the use of enzymatic hydrolysis technology to extract raw materials is a direct application of existing mature technology and will not be elaborated on here.

[0107] S4. Efficiently separate and purify the extracted functional food raw materials to determine the content of their functional and active ingredients;

[0108] It should be noted that separation and purification is a direct application of existing mature technologies and will not be elaborated on here.

[0109] S5. Perform quality classification monitoring on the entire screening and extraction process.

[0110] In order to verify the beneficial effects of the present invention, the following experiments are performed:

[0111] Purpose of the test:

[0112] Verify the effectiveness of intelligent screening and extraction technology for functional food raw materials, including screening accuracy, extraction efficiency and active ingredient retention rate.

[0113] Test Materials:

[0114] 1. Functional food ingredients: such as herbs, fruits, vegetables, etc.

[0115] 2. High-resolution camera or image sensor.

[0116] 3. Computer vision and machine learning algorithms.

[0117] 4. Extraction equipment: such as ultrasonic extractor, microwave extractor, etc.

[0118] 5. Analytical instruments: such as high performance liquid chromatography (HPLC), mass spectrometry (MS), etc.

[0119] Test method:

[0120] 1. Intelligent screening:

[0121] Use a high-resolution camera to capture images of the raw materials.

[0122] Computer vision and machine learning algorithms are used to process images, including preprocessing steps such as denoising, grayscale, binarization, and feature extraction.

[0123] Use the trained model to intelligently identify and screen raw materials.

[0124] 2. Extraction process:

[0125] The screened qualified raw materials are extracted using equipment such as ultrasonic extractors or microwave extractors.

[0126] 3. Active ingredient analysis:

[0127] The active ingredients in the extracts were identified and quantitatively analyzed using analytical instruments such as high performance liquid chromatography (HPLC) and mass spectrometry (MS).

[0128] Compare with standard samples to determine the content and purity of the active ingredients in the extract.

[0129] Data recording and analysis:

[0130] 1. Screening accuracy:

[0131] Record indicators such as error rate, recall rate, and precision during the screening process.

[0132] The calculation formula is as follows:

[0133] Error rate = (number of incorrectly screened raw materials / total number of screened raw materials) × 100%

[0134] Recall rate = (number of qualified raw materials correctly screened / actual number of qualified raw materials) × 100%

[0135] Accuracy = (number of qualified raw materials correctly screened / total number of screened raw materials) × 100%

[0136] 2. Extraction efficiency:

[0137] Record the time, energy consumption and extract yield during the extraction process.

[0138] The calculation formula is as follows:

[0139] Extraction time = total time of the extraction process (minutes)

[0140] Energy consumption = Energy consumed during extraction (kWh)

[0141] Extract yield = mass of active ingredient extracted (g)

[0142] 3. Active ingredient retention rate:

[0143] The content and purity of the active ingredient in the extract were recorded and compared with the content in the original raw material.

[0144] The calculation formula is as follows:

[0145] Active ingredient content retention rate = (active ingredient content in the extract / active ingredient content in the original raw material) × 100%

[0146] Active ingredient purity = Purity of active ingredient in extract (%)

[0147] Test results:

[0148] Through the above test, the following data were obtained, as shown in Table 1:

[0149] Table 1: Experimental data

[0150]

[0151] The present invention provides a functional food raw material intelligent screening and extraction technology, through which the active ingredients in the functional food raw materials can be efficiently and accurately screened and extracted, the extraction efficiency can be improved, the cost can be reduced, and the damage to the active ingredients can be reduced. In addition, through the intelligent control system and data analysis, the stability and repeatability of the extraction process can be ensured, the quality and reliability of the product can be improved, and the problems of low intelligence, poor flexibility and weak data processing capability in the screening and extraction process of existing functional food raw materials can be solved.

[0152] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A functional food raw material intelligent screening and extraction technology, characterized in that: It includes intelligent screening, intelligent extraction, separation and purification, data analysis, and quality control. The specific steps are as follows: S1: Collect food raw materials; S2: Use intelligent means to quickly and accurately identify and intelligently screen food ingredients; S3: According to the intelligent screening results, the raw materials are extracted using enzymatic hydrolysis technology; S4: Efficiently separate and purify the extracted functional food raw materials to determine the content of their functional and active ingredients; S5: Quality classification monitoring of the entire screening and extraction process.

2. The intelligent screening and extraction technology of functional food raw materials according to claim 1, characterized in that: The specific steps of using intelligent means to quickly and accurately identify and intelligently screen food raw materials include the following: S1: After collecting the food raw materials, real-time image acquisition of the food raw materials is performed through a high-resolution camera or image sensor; S2: Obtain the point cloud data of the collected image, build a point cloud distribution rate range acquisition model, and obtain the distribution rate range of each point cloud image; S3: Obtaining the distribution rate range standard value of each food raw material pre-stored in the memory; S4: comparing the point cloud distribution rate range obtained in S2 with the distribution rate range standard value obtained in S3; S5: Identify and intelligently screen the current food image based on the comparison results.

3. The intelligent screening and extraction technology of functional food raw materials according to claim 2, characterized in that: After real-time image acquisition of food raw materials by a high-resolution camera or image sensor, the acquired images are pre-processed; The preprocessing steps specifically include: denoising, grayscale, binarization and normalization.

4. The intelligent screening and extraction technology of functional food raw materials according to claim 3, characterized in that: The point cloud distribution rate range acquisition model constructed is specifically as follows: Wherein, δ is the point cloud distribution rate range; S is the point cloud distribution area in the acquired point cloud distribution image, cm 2 ; G is the number of corners in the shape of the acquired point cloud distribution image, pieces; L is the total length of the point cloud in the acquired point cloud distribution image, cm; 1.41, 0.981, and 0.9811 are all adjustment constants; dx is the integral operation.

5. The intelligent screening and extraction technology of functional food raw materials according to claim 4, characterized in that: The point cloud distribution rate range obtained in S2 is compared with the standard value of the distribution rate range obtained in S3 according to the following formula: Wherein, η is the comparison result value; δ is the point cloud distribution rate range; δ 标 is the standard value of the distribution rate range; 2.61 is the adjustment constant.

6. The intelligent screening and extraction technology of functional food raw materials according to claim 5, characterized in that: When the comparison result value is less than or equal to the threshold, it is determined that the category of the current food image is consistent.

7. The intelligent screening and extraction technology of functional food raw materials according to claim 6, characterized in that: The threshold value is defined as 10.12%.

Images