Marine product biological sample pretreatment method

Abstract

The invention belongs to the technical field of biological sample treatment, and discloses a marine product biological sample pretreatment method which comprises the following steps: removing an outer package of a sample; putting into a clean stainless steel container; performing targeted treatment on different samples, and repeatedly cleaning the samples with distilled water; performing targeted decomposition treatment on different types of marine product samples; laying a sample in a net manner; an analysis and recognition module of the monitoring mechanism recognizes the type and the size of the received sample; identifying an abnormal condition; a drying time and temperature calculation module is used for calculating an optimal drying time and temperature combination through drying equipment; the sample is dried through a drying oven; performing powdering by using a food-grade powdering machine; and packaging the powder sample. Different decomposition modes are adopted for fishes of different sizes, drying can be accurately controlled, the drying time and temperature calculation module establishes an evaluation model based on a large amount of historical data and sample parameters, and the optimal drying time and temperature combination is intelligently calculated.

Description

Technical Field

[0001] The present application relates to the technical field of biological sample processing, and more specifically, to a method for preprocessing marine biological samples. Background Art

[0002] Radioactive substances are transferred through the ecology and food chain, resulting in significant enrichment, which affects the radioactive content of seafood. Some marine organisms such as fish, shellfish and crustaceans may be affected by radioactive substances in nuclear contaminated water, posing a potential risk to human consumers. In order to monitor the impact of radioactive substances on human health through seafood, it is necessary to test the seafood collected before it is discharged into the sea for radioactive substances. The gamma energy spectrum analysis method of radionuclides in biological samples and the determination of strontium-89 and strontium-90 and other radionuclide determination technical methods all require the collected fresh biological samples to be pre-treated, dried and ground. Through cleaning, splitting, drying and powdering, a usable powdered dry sample is generated, and information such as sample weight changes is recorded to prepare the sample for the next step of analysis.

[0003] The document with the prior art publication number CN104392653A provides a shaping method for fish biological specimens, which relates to fish biological specimens. Fine sand is screened, washed with water, floating soil is removed, and dried for standby use; the fish biological sample to be shaped is killed, the internal organs are removed, excess water is wiped off after washing, and then a clean absorbent cotton ball is filled, while the fish body is kept moist; a layer of fine sand is spread at the bottom of a container, the treated fish biological sample is placed on the fine sand, the tail fin of the fish biological sample is grabbed and dragged in the fine sand in reverse, the pectoral fin, dorsal fin, pelvic fin, etc. of the fish biological sample are unfolded under the resistance of the fine sand, the gaps are naturally adhered and filled with fine sand, so that the fin rays no longer shrink, and the shaping process is completed, and then all are buried with fine sand, the container and the fish biological sample buried in the fine sand are taken out after drying, and the fine sand and dust on the surface of the fish biological sample are removed to obtain the shaped fish biological sample; a transparent coating is sprayed on the surface of the shaped fish biological sample, and the shaping of the fish biological specimen is completed after drying.

[0004] Although the above-mentioned prior art solutions can achieve relevant beneficial effects through the structure of the prior art, they still have the following defects: 1. It is not possible to adopt different treatment methods for different seafood products. 2. It is not possible to timely discover abnormal conditions during sample pretreatment; 3. It is not possible to effectively adjust and control the drying temperature and time of the sample.

[0005] In view of this, we propose a pretreatment method for marine biological samples. Summary of the invention

[0006] 1. Technical issues to be solved

[0007] The purpose of the present application is to provide a method for preprocessing marine biological samples, which solves the technical problems raised in the above-mentioned background technology, realizes different decomposition methods for fish of different sizes, and reasonably decomposes or processes crustaceans, crabs, shrimps, shellfish and algae according to their own characteristics, which is helpful for the consistency and accuracy of subsequent analysis and detection; the drying can be accurately controlled, and the drying process is carried out in two steps, and the temperature and time are strictly controlled, which effectively prevents the sample composition from changing due to improper drying, and ensures the quality of the sample and the accuracy of the detection results; the drying time and temperature calculation module establishes an evaluation model based on a large amount of historical data and sample parameters, intelligently calculates the optimal drying time and temperature combination, and adjusts and optimizes according to real-time data during the drying process; the technical effect of precise control of the drying process is achieved.

[0008] 2. Technical solution

[0009] The technical solution of the present application provides a method for pretreatment of marine biological samples, comprising the following steps:

[0010] S1. Receive samples and remove the outer packaging of all samples; if they are frozen samples (including dried samples of fish, crustaceans, and algae that need to be soaked), place them in a clean stainless steel container for natural thawing. For dried algae samples, obtain the fresh weight after soaking; for non-frozen fresh samples (fresh samples of fish, crustaceans, and algae), place them directly in a clean stainless steel container;

[0011] S2, cleaning samples;

[0012] Fish: On a plastic cutting board, carefully cut open the fish from the head to the midline of the abdomen to the vent, remove the internal organs to prevent them from rupturing and contaminating the fish meat, and then remove the gills. After removing the internal organs, drizzle with distilled water.

[0013] Crustaceans: If the shellfish is edible, do not process it; if it is a complete shellfish, use stainless steel utensils to remove the shell and keep only the shellfish meat. Pour with distilled water. For farmed shrimps and crabs, pay attention to remove impurities such as nylon ropes.

[0014] Algae: For fresh samples, wash repeatedly with distilled water to remove impurities such as insects, sand, nylon ropes, etc.

[0015] S3, decompose the sample;

[0016] Fish: For very small fish, only remove the viscera, no need to cut into sections; for some small and medium-sized fish, remove the viscera and cut in half; for large fish, remove the viscera first, cut off the fish head and keep it separately, and then slice it perpendicular to the fish bones from the neck, with a thickness of less than 5mm. At the same time, keep the fish head, fins, scales, and tail. Weigh after disassembly and record it as the fresh sample weight.

[0017] Crustaceans: Cut crabs in half and smash their claws; for shellfish and shrimp, just drain the water and do no other processing.

[0018] Algae: Extra-long algae samples can be cut into pieces, while short algae samples do not need to be cut.

[0019] S4. Spread samples on separate nets: Spread the processed fish, crustacean and algae samples evenly and flatly on the stainless steel mesh, taking individual samples as units, to ensure that the samples are evenly distributed and avoid accumulation or overlap. Place a mesh between each sample, spread a plastic film, and stick it with tape to prevent cross contamination. Label each sample separately, including the sample source, number, processing time, etc.

[0020] S4. Drying samples: Clean the drying equipment and use a high-pressure air gun to blow around the upper blower of the drying tunnel to remove dust. Place the sample in the drying tunnel for about 24 hours to reduce the moisture content of the sample to 20%. This process strictly controls the temperature and time to prevent changes in the sample composition.

[0021] S6. Drying samples: Send the preliminarily dried samples to the laboratory for weighing and recording the weight.

[0022] The fish and crustacean samples were placed in a laboratory oven and dried at 105°C for 8-20 hours; the algae samples were placed in a laboratory oven and dried at 70°C for 8-20 hours.

[0023] During the drying process, the sample is weighed once every 1 hour. When the weight change of the sample is within 1% for two consecutive weighings, the sample is deemed to have reached the dried and moisture-free state.

[0024] S7. Sample powdering: Use a food-grade powdering machine to powder the dried fish, crustacean and algae samples.

[0025] S8. Powder sample packaging: Place the powder sample in a room temperature environment and cool it to room temperature. Weigh it for the last time and record the weight as dry weight. Calculate parameters such as dry-to-fresh ratio based on the dry weight and the previously recorded fresh sample weight. Package the sample and put it into a special bag and seal it to prevent it from being affected by external air, moisture, etc. Attach a label to the bag, indicating the sample name, source, number, processing time, dry weight, dry-to-fresh ratio and other information for easy storage, management and subsequent use. Send the sample for testing.

[0026] As an optional solution of the present invention, in the process of separating and laying the fish, crustacean and algae samples, a monitoring mechanism is used to collect high-definition images of the fish, crustacean and algae samples, analyze their types, analyze the size of the samples after decomposition, and calculate the temperature and time of sample drying and baking. The monitoring mechanism includes:

[0027] Data collection module: Collect a large number of images of seafood such as fish, crustaceans and algae samples, covering samples at different growth stages, species characteristics and in different environments. Label the images (including species) as reference samples; collect a large number of experimental data of fish, crustaceans and algae samples (including drying time and temperature data); comprehensively collect experimental data of fish, crustaceans and algae samples during pretreatment, such as temperature and time curves of different samples during drying, and weight and size change data of samples in various processing links. These data will provide a solid foundation for subsequent analysis and model building.

[0028] Image acquisition module: Install high-definition cameras at key locations such as the sample receiving area, sample decomposition area, and inside the drying and drying equipment to ensure that the status of the sample at each stage can be clearly captured.

[0029] Image preprocessing module: preprocess the collected images, including filtering and denoising, image enhancement, grayscale and normalization;

[0030] Feature extraction module: extracts features from the preprocessed image. The extracted features include color, texture and shape.

[0031] Color feature extraction: By analyzing the color histogram, color moment and other features of the image, the color information of the sample, such as the main color tone, color distribution, etc., is extracted. These color features can be used as one of the important bases for distinguishing different types of seafood.

[0032] Texture feature extraction: Use texture analysis algorithms, such as gray-level co-occurrence matrix, local binary pattern, etc., to extract texture features of the sample surface, such as texture thickness, direction, repeatability, etc. Texture features play an important role in distinguishing different types of seafood with similar appearance.

[0033] Shape feature extraction: Through edge detection, contour extraction and other technologies, the shape information of the sample is obtained, such as area, perimeter, aspect ratio, circularity, etc. Shape features are key indicators for identifying seafood types and judging whether the sample decomposition meets the requirements.

[0034] Analysis and recognition module: Analyze and recognize the image after feature extraction, identify the type of received samples, identify the size of decomposed samples; identify abnormal situations (sample damage or over-drying, etc.) in time;

[0035] Temperature sensing module: Install high-precision temperature sensors at different locations inside the drying box and oven to ensure that the temperature data of each area in the box can be collected in real time and accurately. The collected temperature data is transmitted to the PLC control module in real time to provide a basis for subsequent drying time and temperature calculations and abnormal situation judgments.

[0036] Drying time and temperature calculation module: Based on the large amount of historical data accumulated in the data collection module, combined with the type, size, weight and other parameters of fish, crustacean and algae samples, the drying time and temperature evaluation model is established using data mining and machine learning algorithms. The model can intelligently calculate the optimal drying time and temperature combination based on the input sample information to ensure that the sample can fully remove moisture during the drying process while retaining the composition and characteristics of the sample to the greatest extent. During the sample drying process, the drying time and temperature are adjusted and optimized in real time based on the temperature data fed back in real time by the temperature sensing module and the sample weight and size change information provided by the analysis and recognition module to ensure accurate control of the drying process.

[0037] Alarm module: includes an alarm, which will issue an alarm in time when an abnormal situation is detected.

[0038] PLC control module: connected with the data collection module, image acquisition module, image preprocessing module, feature extraction module, analysis and recognition module, temperature sensing module, alarm module, drying box and drying time and temperature calculation module. According to the various data information received, the PLC control module performs intelligent control and decision-making on the entire preprocessing process.

[0039] As an optional solution of the present invention, the drying equipment is a tunnel-type air duct low-temperature rapid drying equipment.

[0040] The tunnel-type air duct low-temperature rapid drying equipment includes: steam heater, steam heater, exhaust fan and closed air duct;

[0041] A steam heater is installed at the air inlet of the closed air duct, hot dry air (38-42°C) heated by steam is input, and the air is forced to flow by a blower; an exhaust fan is used to extract air at the outlet of the closed air duct to discharge the air into the atmosphere.

[0042] In the closed air duct, hot dry air continuously passes through the objects to be dried, and the air continuously absorbs moisture from the objects. The air changes from dry air to high-humidity air and is discharged from the air outlet, achieving the purpose of drying the objects.

[0043] 3. Beneficial effects

[0044] One or more technical solutions provided in the technical solution of this application have at least the following technical effects or advantages:

[0045] 1. The present invention can adopt different decomposition methods for fish of different sizes, and reasonably decompose or process crustaceans, crabs, shrimps, shellfish and algae according to their own characteristics, which is helpful for the consistency and accuracy of subsequent analysis and detection.

[0046] 2. When laying samples on the mesh, take a single sample as a unit and lay it evenly and flatly on the stainless steel mesh. Place mesh between samples and spread plastic film and stick it with tape to effectively avoid cross contamination between different samples, ensure the independence of each sample and the reliability of the test results. Each sample is marked with information such as sample source, number, processing time, etc., for easy subsequent traceability and management.

[0047] 3. Drying can be precisely controlled. The drying process is carried out in two steps, and the sample is weighed every 1 hour. When the weight change of the sample is within 1% after two consecutive weighings, the drying is considered complete. The temperature and time are strictly controlled to effectively prevent the sample composition from changing due to improper drying, thus ensuring the quality of the sample and the accuracy of the test results.

[0048] 4. Install high-precision temperature sensors at different locations in the drying box and oven to collect temperature data in real time and accurately, providing a reliable basis for drying time and temperature calculation and abnormal judgment. The drying time and temperature calculation module establishes an evaluation model based on a large amount of historical data and sample parameters, intelligently calculates the optimal drying time and temperature combination, and adjusts and optimizes according to real-time data during the drying process. It achieves precise control of the drying process, which can fully remove moisture and retain sample composition and characteristics to the greatest extent. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 The figure is a schematic flow chart of a method for pretreating marine biological samples disclosed in a preferred embodiment of the present application.

[0050] Figure 2 This is a weight record sheet for the air-drying experiment using air-dried Spanish mackerel as a sample.

[0051] Figure 3 These are the images of the sample after drying and the image of powdering using air-dried Spanish mackerel as the sample for the air-drying experiment.

[0052] Figure 4 These are comparison images of samples 1-6 being powdered and bagged in an air-drying experiment using air-dried Spanish mackerel as samples. DETAILED DESCRIPTION

[0053] The present application is further described in detail below in conjunction with the accompanying drawings.

[0054] Reference Figure 1 Embodiment 1: The present invention provides a method for pretreating fish biological samples, comprising the following steps:

[0055] 1. Receiving samples: After receiving the samples, remove the outer packaging and place the frozen samples in a clean stainless steel container for natural thawing; for non-frozen fresh samples, place them in a clean stainless steel container; natural thawing can maintain the cell structure and component integrity of the samples to the greatest extent, and prevent nutrient loss or tissue damage caused by rapid thawing.

[0056] 2. Clean the samples: Remove the internal organs on a plastic cutting board. Starting from the fish head, cut along the midline of the abdomen and accurately cut to the excretion hole. This process requires caution to prevent the internal organs from rupturing and contaminating the fish meat. Then, completely remove the gills and internal organs. If all the samples are relatively clean and there is no obvious dirt or impurities, no additional cleaning operation is required, but it is important to ensure that the cutting board and knife are strictly cleaned before use to ensure that there are no residual contaminants on their surface. After completing the evisceration operation, pour distilled water on the sample. The purity of distilled water can effectively avoid the introduction of new impurities and ensure the cleanliness of the sample.

[0057] S3. Decompose the sample: Decompose according to the size of the fish. Different decomposition methods are used according to the size of the fish.

[0058] For very small fish, simply remove the viscera without cutting them into pieces to preserve their overall shape for subsequent analysis.

[0059] For some small and medium-sized fish, after removing the internal organs, they were cut in half. This processing method can meet the sample size requirements of subsequent experiments without excessively damaging the fish tissues.

[0060] For large fish, remove the internal organs first and then cut them into pieces. When cutting, cut off the fish head first. The fish head has unique analytical value and needs to be kept separately. Then, start slicing from the neck. The slicing direction should be perpendicular to the fish bones, and the thickness should be strictly controlled to be less than 5mm. Such thickness can not only ensure the uniformity of the sample in subsequent drying and other processing processes, but also facilitate analysis. At the same time, the fish head, fins, scales, and tail are retained. These parts contain specific biomarkers or have different composition in the experiment, which is of great significance for the comprehensive study of the biological characteristics of fish. After the decomposition is completed, weigh each part of the sample and record it in detail as the fresh sample weight of the sample.

[0061] S4. Spread samples on separate meshes: Spread samples using individual samples as units. Spread the samples evenly and flatly on the stainless steel mesh to ensure that the samples are evenly distributed and avoid accumulation or overlap, so as to ensure the consistency of subsequent drying and other treatments. To prevent the juice of the upper sample from dripping onto the lower sample before drying and causing cross contamination, place a mesh between each sample and spread plastic film on the mesh. Use tape to firmly adhere the plastic film to form an effective isolation layer. At the same time, clearly mark each sample separately. The marking content should include key information such as the source, number, and processing time of the sample, so that each sample can be accurately traced and identified during subsequent experiments.

[0062] S5. Drying samples: Before drying the samples, clean the drying equipment first. Use a high-pressure air gun to blow around the upper fan of the drying tunnel to completely remove the dust stuck to the fan and the four walls. The dust may contain various microorganisms, impurities, etc. If not removed, they may fall on the samples during the drying process and cause cross contamination. After cleaning, place the samples in the drying tunnel for drying. The drying time is about 24 hours, so that the moisture content of the samples is reduced to 20%. This process requires strict control of temperature and time to ensure uniform evaporation of water, while avoiding changes in the composition of the samples due to excessive temperature or excessive time.

[0063] S6. Drying samples: Send the preliminarily dried samples to the laboratory, weigh them again, and record the weight at this time. Then, place the samples in a laboratory oven and dry them at 105°C for 8-20 hours. During the drying process, weigh the samples every 1 hour. When the sample weight changes within 1% after two consecutive weighings, it can be determined that the sample has reached a dry and moisture-free state. This precise weighing and judgment process can ensure that the moisture in the sample is completely removed, providing an accurate basis for subsequent analysis.

[0064] S7. Sample powdering: Use a food-grade powdering machine to powder the dried samples. The food-grade powdering machine can ensure that no harmful chemicals are introduced during the powdering process, ensuring the purity and safety of the samples. After the powdering is completed, the powdering machine cup is thoroughly cleaned and rinsed with distilled water several times to remove the sample powder remaining in the cup. After cleaning, place the cup in a clean environment to dry naturally to prevent secondary contamination during the drying process and prepare for the next use.

[0065] S8. Powder sample packaging: Place the powder sample at room temperature to cool, and after it has completely cooled to room temperature, weigh it for the last time and record the weight as the dry weight. According to the dry weight and the previously recorded weight of the fresh sample, calculate the dry-to-fresh ratio and other related parameters. These parameters are of great significance for analyzing the component content and nutritional composition of the sample. Subsequently, the sample is placed in a special bag and sealed to ensure that the sample is not affected by external air, moisture and other factors. Attach a label to the bag. The label should specify the name, source, number, processing time, dry weight, dry-to-fresh ratio and other information of the sample to facilitate the storage, management and subsequent use of the sample.

[0066] Example 2: The key points of crustacean processing vary according to different species. For shellfish, it is necessary to remove the shells and clean the sand in the shellfish meat. For farmed shrimps and crabs, it is necessary to remove artificial impurities such as nylon ropes mixed in during the farming and fishing process. The overall oil content of crustaceans is not high, but the water content is high, so more raw materials are needed to produce sufficient dry powder. The crustacean processing process is basically similar to that of fish, with slight differences as follows:

[0067] In the second step: if the sample is only edible shellfish, do not process it; if it is a whole shellfish, use stainless steel utensils to remove the shell and keep only the shellfish meat;

[0068] In step 3: cut the crab in half and smash the crab claws; for shellfish and shrimp, just drain the water and do no other processing.

[0069] The present invention provides a method for pretreating crustacean biological samples, comprising the following steps:

[0070] 1. Receiving samples: After receiving the samples, remove the outer packaging and place the frozen samples in a clean stainless steel container for natural thawing; for non-frozen fresh samples, place them in a clean stainless steel container;

[0071] 2. Cleaning samples: If the sample is only edible shellfish meat, no processing is done; if it is a complete shellfish, use stainless steel utensils to remove the shell and only keep the shellfish meat; if all the samples are relatively clean and have no obvious dirt or impurities, no additional cleaning operation is required. Pour the samples with distilled water. The purity of distilled water can effectively avoid the introduction of new impurities and ensure the cleanliness of the samples.

[0072] 3. Decompose the sample: Cut the crab in half and smash the crab claws; for shellfish and shrimp, only drain the water and do no other processing.

[0073] 4. Spread samples on separate meshes: Spread samples on a single unit. Spread the processed samples evenly and flatly on the stainless steel mesh to ensure that the samples are evenly distributed and avoid accumulation or overlap. Clearly mark each sample, and the marking content should include key information such as the source, number, and processing time of the sample.

[0074] 5. Drying samples: Before drying the samples, clean the drying equipment. After cleaning, place the samples in the drying tunnel for drying. The drying time is about 24 hours, so that the moisture content of the samples is reduced to 20%.

[0075] 6. Drying samples: Send the preliminarily dried samples to the laboratory, weigh them again, and record the weight at this time. Then, put the samples into the laboratory oven and dry them at 105°C for 8-20 hours. During the drying process, weigh the samples every 1 hour. When the weight change of the sample is within 1% after two consecutive weighings, it can be determined that the sample has reached the state of drying and moisture-free.

[0076] 7. Sample powdering: Use a food-grade powdering machine to powder the dried samples.

[0077] 8. Powder sample packaging: Place the powder sample at room temperature to cool, and after it has completely cooled to room temperature, weigh it for the last time and record the weight as the dry weight. According to the dry weight and the previously recorded fresh sample weight, calculate the dry-to-fresh ratio and other related parameters, which are important for analyzing the component content and nutritional composition of the sample. Subsequently, put the sample into a special bag, seal it, and affix a label to the bag. The label should specify the sample name, source, number, processing time, dry weight, dry-to-fresh ratio and other information in detail to facilitate the storage, management and subsequent use of the sample.

[0078] Example 3: The original storage state of algae samples is divided into dried samples (e.g. kelp) and fresh samples. The dried samples need to be soaked in distilled water to obtain the fresh weight, and then processed according to the fresh sample process. During the fresh sample processing, it is necessary to remove impurities such as insects, sand, nylon ropes, etc. during cleaning. The drying temperature of algae is 70°C, which is different from fish and crustaceans.

[0079] The present invention provides an algae biological sample pretreatment method, comprising the following steps:

[0080] 1. Receive samples: After receiving the samples, remove the outer packaging, soak the dry samples with distilled water, and obtain the fresh weight; place the fresh samples in a clean stainless steel container;

[0081] 2. Cleaning samples: When handling fresh samples, remove impurities such as insects, sand, nylon ropes, etc., wash them repeatedly with distilled water to remove impurities, dry them, weigh them, and record them as the weight of fresh samples.

[0082] 3. Decompose the samples: Extra-long algae samples can be cut into pieces, while short algae samples do not need to be cut.

[0083] 4. Spread samples on mesh: Spread samples on a single unit. Spread the processed samples evenly and flatly on the stainless steel mesh to ensure that the samples are evenly distributed and avoid accumulation or overlap to ensure the consistency of subsequent drying and other treatments. Clearly label each sample, and the label content should include key information such as the source, number, and processing time of the sample, so that each sample can be accurately traced and identified in the subsequent experimental process.

[0084] 5. Drying samples: Before drying the samples, clean the drying equipment. After cleaning, place the samples in the drying tunnel for drying.

[0085] 6. Drying samples: Send the preliminarily dried samples to the laboratory, weigh them again, and record the weight at this time. Then, put the samples in the laboratory oven and dry them at 70°C for 8-20 hours. During the drying process, weigh the samples every 1 hour. When the weight change of the sample is within 1% after two consecutive weighings, it can be determined that the sample has reached the state of drying and moisture-free.

[0086] 7. Sample powdering: Use a food-grade powdering machine to powder the dried samples.

[0087] 8. Powder sample packaging: Place the powder sample at room temperature to cool. After it has completely cooled to room temperature, weigh it for the last time and record the weight as dry weight. According to the dry weight and the previously recorded fresh sample weight, calculate the dry-to-fresh ratio and other related parameters. These parameters are of great significance for analyzing the component content and nutritional composition of the sample. Put the sample into a special bag, seal it, and put a label on the bag. The label should specify the sample name, source, number, processing time, dry weight, dry-to-fresh ratio and other information in detail to facilitate the storage, management and subsequent use of the sample.

[0088] Example 4: In the process of separating and laying fish, crustacean and algae samples, high-definition images of fish, crustacean and algae samples are collected by a monitoring mechanism to analyze their types, analyze the size of the samples after decomposition, and calculate the temperature and time of sample drying and baking. The monitoring mechanism includes:

[0089] Data collection module: Collect a large number of images of seafood such as fish, crustaceans and algae samples, covering samples at different growth stages, species characteristics and in different environments. Label the images (including species) as reference samples; collect a large number of experimental data of fish, crustaceans and algae samples (including drying time and temperature data); comprehensively collect experimental data of fish, crustaceans and algae samples during pretreatment, such as temperature and time curves of different samples during drying, and weight and size change data of samples in various processing links. These data will provide a solid foundation for subsequent analysis and model building.

[0090] Image acquisition module: Install high-definition cameras in key locations such as the sample receiving area, sample decomposition area, and inside the drying and drying equipment to ensure that the status of the samples at each stage can be clearly captured. Collect high-definition images of the received samples to facilitate the subsequent analysis of the sample types; collect high-definition images of the decomposed samples to facilitate the calculation of the size and weight of the decomposed samples;

[0091] Image preprocessing module: preprocess the collected images, including filtering and denoising, image enhancement, grayscale and normalization;

[0092] Filtering and denoising: Use advanced filtering algorithms, such as Gaussian filtering and median filtering, to remove noise generated by environmental interference, equipment noise and other factors during the image acquisition process, making the image clearer and smoother and improving image quality.

[0093] Image enhancement: Use image enhancement techniques such as histogram equalization and contrast stretching to enhance the contrast and clarity of the image, highlight the detailed features of the sample, and facilitate subsequent feature extraction and analysis.

[0094] Grayscale: Convert color images into grayscale images to reduce the amount of image data, while highlighting the brightness information of the image and simplifying subsequent image processing.

[0095] Normalization: Normalize the image and uniformly map the pixel values ​​of the image to a specific range to make different images comparable and improve the accuracy and stability of image processing.

[0096] Feature extraction module: extracts features from the preprocessed image. The extracted features include color, texture and shape.

[0097] Color feature extraction: By analyzing the color histogram, color moment and other features of the image, the color information of the sample, such as the main color tone, color distribution, etc., is extracted. These color features can be used as one of the important bases for distinguishing different types of seafood.

[0098] Texture feature extraction: Use texture analysis algorithms, such as gray-level co-occurrence matrix, local binary pattern, etc., to extract texture features of the sample surface, such as texture thickness, direction, repeatability, etc. Texture features play an important role in distinguishing different types of seafood with similar appearance.

[0099] Shape feature extraction: Through edge detection, contour extraction and other technologies, the shape information of the sample is obtained, such as area, perimeter, aspect ratio, circularity, etc. Shape features are key indicators for identifying seafood types and judging whether the sample decomposition meets the requirements.

[0100] Analysis and recognition module: Analyze and recognize the image after feature extraction, identify the type of received samples, identify the size of the decomposed samples; identify abnormal situations (sample damage or over-drying, etc.) in a timely manner.

[0101] Temperature sensing module: Install high-precision temperature sensors at different locations inside the drying box and oven to ensure that the temperature data of each area in the box can be collected in real time and accurately. The collected temperature data is transmitted to the PLC control module in real time to provide a basis for subsequent drying time and temperature calculations and abnormal situation judgments.

[0102] Drying time and temperature calculation module: Based on the large amount of historical data accumulated in the data collection module, combined with the type, size, weight and other parameters of fish, crustacean and algae samples, the drying time and temperature evaluation model is established using data mining and machine learning algorithms. The model can intelligently calculate the optimal drying time and temperature combination based on the input sample information to ensure that the sample can fully remove moisture during the drying process while retaining the composition and characteristics of the sample to the greatest extent. During the sample drying process, the drying time and temperature are adjusted and optimized in real time based on the temperature data fed back in real time by the temperature sensing module and the sample weight and size change information provided by the analysis and recognition module to ensure accurate control of the drying process.

[0103] Alarm module: including alarm, which will sound an alarm in time when abnormal conditions are detected; equipped with various alarms such as sound and light to ensure that alarm signals can be issued in time and effectively when abnormal conditions are detected. When the analysis and identification module detects abnormal conditions such as sample damage and over-drying, or the drying time and temperature calculation module finds that the actual drying temperature deviates too much from the set value, or the drying time exceeds the expected range, the alarm module will be triggered immediately to remind the operator to handle it to avoid affecting the experimental results.

[0104] PLC control module: connected to the data collection module, image acquisition module, image preprocessing module, feature extraction module, analysis and recognition module, temperature sensing module, alarm module, drying box and drying time and temperature calculation module. Based on the various data information received, the PLC control module intelligently controls and makes decisions on the entire preprocessing process. For example, according to the optimal drying time and temperature provided by the drying time and temperature calculation module, the heating element of the drying box is controlled to achieve precise control of the drying process; when the alarm module triggers the alarm, the PLC control module can automatically take corresponding measures according to the preset emergency plan, such as stopping the operation of the drying equipment, starting the ventilation and heat dissipation device, etc., to ensure the safety and smooth progress of the experiment.

[0105] Furthermore, the analysis and recognition module analyzes and recognizes the image after feature extraction, identifies the type of the received sample, identifies the size of the decomposed sample, and promptly identifies abnormal conditions (such as sample damage or over-drying). The following steps are included:

[0106] 1. Sample type identification:

[0107] 1.1. Feature data preparation: Get the extracted image features from the feature extraction module, including color features (such as color histogram, color moment), texture features (such as gray-level co-occurrence matrix, local binary pattern) and shape features (such as area, perimeter, aspect ratio, circularity), etc. Make sure that these feature data have been organized into a format suitable for machine learning algorithm processing, such as combining different types of features into a feature vector.

[0108] 1.2. Database loading: Load the established reference sample database from the data collection module. The database should contain the characteristic data of different types of fish, crustaceans and algae samples, and the characteristic data of each sample should be accurately labeled with its species information. The storage structure of the database should facilitate fast search and comparison, such as using a database management system or file storage system, and using an index structure to improve retrieval efficiency.

[0109] 1.3. Build a species recognition model: Select the support vector machine (SVM) algorithm model. Use the sample features in the database as training data and the sample types as labels to train the model. Normalize the sample features to ensure that features of different dimensions are on the same scale. Adjust the parameters of SVM, such as kernel functions (linear, polynomial, radial basis function, etc.) and regularization parameters, and find the optimal parameter combination through cross-validation and other methods to improve the classification performance of the model. Input the extracted features of the sample to be identified into the trained SVM model to obtain the species prediction result of the sample. To verify the species recognition results, the model performance can be evaluated by calculating indicators such as accuracy, recall, and F1 value.

[0110] For misclassified samples, the reasons for analysis may be insufficient feature extraction, unbalanced sample data, or inappropriate model parameters. The feature extraction method or model should be adjusted and optimized based on the analysis results.

[0111] 2. Size calculation:

[0112] 2.1. Image scale acquisition: In the image acquisition module, by placing a standard reference object of known size in the image, the scale of the image and the actual size is calculated. A ruler of known length is placed next to the sample, and the scale is calculated based on its pixel length in the image and the actual length. The scale information is stored for subsequent calculations.

[0113] 2.2. Size calculation: For the decomposed sample image, use image processing technology to detect edges, such as using the Canny edge detection algorithm to accurately extract the sample's outline. According to the edge outline, calculate the sample's length, width, thickness and other size information. For length and width, the distance between the farthest two points on the outline can be calculated; for thickness, it can be estimated based on images from multiple angles or 3D reconstruction technology. Convert the calculated pixel size to the actual size according to the image scale.

[0114] 3. Abnormal situation identification: For sample damage, set the threshold of appearance features. For example, for shape features, if the integrity of the contour is lower than a certain threshold, it is considered that there may be damage; for texture features, if the uniformity of the texture is lower than a certain threshold, it may also indicate damage. For over-drying, set the threshold of weight and size change. According to historical data, when the weight change of a sample within a certain period of time is less than a minimum value, or the size shrinkage ratio exceeds a certain threshold, it may indicate over-drying.

[0115] For deformation, the threshold of shape change is set according to different types of samples, such as the allowable range of change of shape features such as aspect ratio and circularity.

[0116] Obtain weight and size data from the analysis and recognition module in real time, and continuously observe its changes during the drying process. Observe the shape, texture and other features obtained by the feature extraction module to check whether they exceed the set threshold range.

[0117] 4. Abnormal handling: When the weight, size or characteristics are detected to be beyond the set threshold range, the alarm module is triggered immediately. The time when the abnormality occurred, the sample number involved, the type of abnormality and other information are recorded to facilitate the subsequent analysis of the abnormality. The alarm module can notify relevant personnel through sound, light or send messages to monitoring personnel, so that they can take timely measures, such as adjusting drying parameters, checking samples, etc.

[0118] In this technical solution, the analysis and identification module can conduct comprehensive analysis and identification of seafood samples, ensure the accuracy and safety of the sample processing process, and provide reliable data support and early warning of abnormal situations for subsequent experimental analysis and processing.

[0119] Furthermore, the drying time and temperature calculation module uses data mining and machine learning algorithms to establish an evaluation model for drying time and temperature based on a large amount of historical data accumulated in the data collection module, combined with parameters such as the type, size, and weight of fish, crustacean, and algae samples, and intelligently calculates the optimal drying time and temperature combination, including the following steps;

[0120] 1. Data preprocessing: The historical drying data of fish, crustacean and algae samples are obtained from the data collection module. These data include information such as sample type, size (length, width, thickness), weight, and corresponding drying time and drying temperature.

[0121] Clean the data to remove outliers and missing values. For example, for weight data, if there is a maximum or minimum value that is significantly deviated from other data, it can be compared with the weight range of samples of the same type to determine whether it is an outlier. If so, it can be corrected or deleted. For missing values, methods such as mean filling and regression prediction can be used to supplement them.

[0122] Normalize the data and convert the data of different dimensions to the same scale range.

[0123] 2. Model construction: Select the support vector machine (SVM) model and use the preprocessed historical data to train the model. The sample characteristics (type, size, weight, etc.) are used as input, and the corresponding drying time and temperature are used as output labels. During the training process, the weights and thresholds are continuously adjusted through the back propagation algorithm to minimize the error between the predicted value and the true value. For example, the mean square error (MSE) is used as the loss function.

[0124] The performance of the model is evaluated through cross-validation and other methods, and the model parameters are continuously adjusted until the model achieves good prediction accuracy. The drying time is calculated according to the following formula:

[0125] t=∑ N i＝1 [(a i -a i * )K(x,x i )+b]+∑ N i＝1 ∑ N j＝1 [(a i -a i * )K(xi ,x j )];

[0126] K(x,x i )=exp{-∑ m k＝1 {[w k (x k -x ik ) 2 ] / (2σ 2 )}};

[0127] K(x i ,x j )=exp{-∑ m k＝1 {[w k (x ik -x jk ) 2 ] / (2σ 2 )}};

[0128] Where t represents the drying time, which is the target variable to be predicted. x is the input feature vector, which contains various characteristics of seafood, including length, width, thickness, weight and seafood type. i is the input feature vector of the th sample in the training dataset, and has the same structure as a. i and a i * are Lagrange multipliers, which are optimization variables obtained by solving a quadratic programming problem. In support vector machines, these multipliers are used to determine which samples are support vectors (i.e., samples that play a key role in determining the decision boundary) and determine how much they contribute to the final prediction. K(x,x i ) is the kernel function. k is the weight factor for different features. Different features can contribute to the kernel function to different degrees. For example, if we think that weight has a greater impact on drying time or temperature than length, we can set the corresponding weight component to be relatively large. When calculating the kernel function, the difference in weight will have a greater impact on the result. By adjusting w k , can emphasize or weaken the importance of different features in constructing high-dimensional space, so that the model pays more attention to the features that have a greater impact on the drying process. k In the context of drying time and temperature prediction, multiple features are combined into an input feature vector x to describe the information of seafood. ik represents the kth feature of the i-th sample. K(x i ,x j ) is the kernel function. jk represents the kth feature of the jth sample.

[0129] Calculate the drying temperature according to the following formula:

[0130] T 烘干 =∑ N i＝1 [(β i -β i * )K(x,x i )+b^]+∑ N i＝1 ∑ N j＝1 [(β i -β i * )(β j -β j * )K(x i ,x j )];

[0131] Where, T 烘干 represents the drying temperature, which is another target variable to be predicted. i and β i * represents the Lagrange multiplier, which is the optimization variable obtained by solving the quadratic programming problem and is used to determine the samples that play a key role in the drying temperature prediction and their contribution. b^ is the bias term, which is used to translate the drying temperature prediction results.

[0132] 3. Model application and real-time adjustment: When new samples need to be dried, the type, size, weight and other information of the samples are input into the trained model, and the model outputs the predicted optimal drying time and temperature.

[0133] During the sample drying process, the temperature sensing module provides real-time feedback of the current drying temperature, and the analysis and recognition module provides real-time weight and size change information of the sample. Based on the real-time feedback data, the weight change rate and size change rate are calculated.

[0134] If the weight change rate or size change rate deviates from the expected range, or the current temperature deviates greatly from the predicted temperature, the drying time and temperature are adjusted according to the preset adjustment rules. For example, if the weight changes too slowly, it means that the drying speed may be too slow, and the temperature can be appropriately increased; if the temperature is too high and the size changes too quickly, it may affect the sample composition and characteristics, and the temperature can be appropriately lowered and the drying time can be extended. The specific calculation formula and rules for adjustment can be determined based on experiments and experience.

[0135] Through the above drying time and temperature calculation module, the drying time and temperature can be intelligently calculated and adjusted in real time according to the sample information, ensuring accurate control of the drying process and retaining the composition and characteristics of the sample to the greatest extent.

[0136] Furthermore, the drying equipment is a tunnel-type air duct low-temperature rapid drying equipment.

[0137] The tunnel-type air duct low-temperature rapid drying equipment includes: steam heater, steam heater, exhaust fan and closed air duct;

[0138] A steam heater is installed at the air inlet of the closed air duct, hot dry air (38-42°C) heated by steam is input, and the air is forced to flow by a blower; an exhaust fan is used to extract air at the outlet of the closed air duct to discharge the air into the atmosphere.

[0139] In the closed air duct, hot dry air continuously passes through the objects to be dried, and the air continuously absorbs moisture from the objects. The air changes from dry air to high-humidity air and is discharged from the air outlet, achieving the purpose of drying the objects.

[0140] After low-temperature drying (~40°C, ~24h), the moisture content of seafood can be reduced to 20%, and the oil content of the sample can be effectively reduced.

[0141] Example 5: Reference Figure 2 , Figure 3 and Figure 4 , air-dried Spanish mackerel was used as a sample (sample weight 400 g) for air-drying experiment:

[0142] Test plan: Divide the samples into samples 1-6, of which samples 1-3 are dried at 105℃, samples 4-6 are dried at 150℃, samples 1 and 4 are dried for 2 hours, samples 2 and 5 are dried for 4 hours, and samples 3 and 6 are dried for 6 hours, and then weighed and broken separately. Weight records are shown in Figure 2 Table 1:

[0143] After 2 hours of drying, the samples 1-3 were obviously damp and had oily skin, which was mainly attached to the surface of the samples, with a small amount falling into the tray. The on-site judgment showed that the humidity was still too high and no powdering was necessary.

[0144] Samples 4-6 had a burnt aroma, a crispy appearance, and a large amount of oil dripping into the tray. Sample 4 was tried for powdering, but the amount of oil was too large, resulting in fine particles.

[0145] Preliminary summary: The fish produces too much oil; Idea 1, lower the temperature and extend the drying time to avoid oil production; Idea 2, increase the temperature (150°) and increase the time to reduce the oil content as much as possible.

[0146] Situation after 4 hours of drying: Weight updated in the table above; Sample No. 2 has little change in appearance and weight, and it is impossible to determine whether the weight loss is water or oil. There is obvious grease on the surface, and a small amount drips into the tray; during the breaking and powdering process, the fish bones are crispy and the fish meat still has a certain toughness, and obvious grease can be seen in the thicker part of the fish meat (0.5cm). The powdering result is still in the shape of meat floss. There is no obvious grease on the surface of Sample No. 5, and the weight does not change much. The powdering result is obviously finer than Sample No. 4, and the appearance is similar to medium silt sand, but there is grease adhesion, and it is difficult to pass the 100 mesh sieve.

[0147] After 6 hours of drying: Sample No. 3 has little change in appearance, little change in weight, and less grease on the surface. The powdered product is like meat floss. Sample No. 6 has little change in weight, darker color, no grease on the surface, and still has a lot of oil when pinched by hand. The powdered product is still mainly fine particles, and the particles are slightly fine. The adhesion phenomenon is larger in scope but less severe. It is still unlikely to pass through a 100-mesh sieve, so it was not tried.

[0148] Summary: According to relevant requirements, the temperature of fish samples cannot exceed 105℃; experimental results show that the degree of drying of fish samples at 105℃ for 6 hours is close to that of drying at 150℃ for 4 hours, and basically meets the requirements for powder sample preparation. Therefore, it is preliminarily determined that the subsequent air-dried samples are based on 105℃ and drying for 8 hours, and the fresh samples are based on 105℃ and drying for 24+8 hours, and the drying process is adjusted according to the appearance characteristics of the samples.

[0149] The present invention can adopt different decomposition methods for fish of different sizes, and reasonably decompose or process crustaceans, crabs, shrimps, shellfish and algae according to their own characteristics, which is helpful for the consistency and accuracy of subsequent analysis and detection. When laying samples on the net, a single sample is used as a unit, and is evenly and flatly laid on a stainless steel mesh. A mesh is placed between samples and a plastic film is spread and glued with tape, which effectively avoids cross contamination between different samples and ensures the independence of each sample and the reliability of the test results. Each sample is marked with information such as the source, number, and processing time of the sample, which is convenient for subsequent tracing and management. Drying can be accurately controlled. The drying process is carried out in two steps, and the sample is weighed every 1 hour during the drying process. When the weight change of the sample weighed twice in a row is within 1%, the drying is considered to be completed. The temperature and time are strictly controlled, which effectively prevents the sample composition from changing due to improper drying, and ensures the quality of the sample and the accuracy of the test results. High-precision temperature sensors are installed at different positions in the drying box and the oven to accurately collect temperature data in real time, providing a reliable basis for drying time and temperature calculation and abnormal judgment. The drying time and temperature calculation module establishes an evaluation model based on a large amount of historical data and sample parameters, intelligently calculates the optimal drying time and temperature combination, and adjusts and optimizes according to real-time data during the drying process. It achieves precise control of the drying process, which can fully remove moisture and retain the sample composition and characteristics to the greatest extent.

[0150] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for pretreating marine biological samples, characterized in that: The following steps are involved: S1. Receive samples and remove the outer packaging of samples; for frozen samples, place them in a clean stainless steel container for natural thawing, and measure the fresh weight of the dried algae samples after they are soaked; non-frozen fresh samples are directly placed in a clean stainless steel container; S2. Cleaning samples: Carry out targeted treatment on different samples and repeatedly clean the samples with distilled water; S3. Decompose samples: Carry out targeted decomposition treatment on different types of seafood samples; S4. Spread samples on separate nets: Spread the processed fish, crustacean and algae samples evenly and flatly on the stainless steel mesh as a unit. Use mesh and plastic film to separate and stick the samples together. Label each sample with its source, number and processing time information. The type and size of the received samples are identified through the analysis and identification module of the monitoring organization; abnormal situations are identified in a timely manner; S5. Drying samples: Drying the samples through drying equipment to reduce the moisture content of the samples to 20%; the drying time and temperature calculation module calculates the optimal drying time and temperature combination; S6. Drying samples: Send the preliminarily dried samples to the laboratory for weighing and recording the weight; dry the samples in an oven; S7. Sample powdering: Use a food-grade powdering machine to powder the dried fish, crustacean and algae samples; S8. Powder sample packaging: Cool the powder sample to room temperature and weigh it to record the dry weight. Calculate the dry-to-fresh ratio based on this weight and the weight of the fresh sample. Put the sample into a bag, seal it, and affix a label containing the sample name and source information.

2. The method for pretreating marine biological samples according to claim 1, characterized in that: Step S2 includes the following steps: S21. Cleaning of fish samples: On a plastic cutting board, carefully cut open the fish from the head along the midline of the abdomen to the excretion hole, remove the internal organs and gills, and after the internal organs are removed, pour distilled water over it; S22. Cleaning of crustacean samples: For edible shellfish, do not process; for intact shellfish, use stainless steel utensils to remove the shell, leaving only the shellfish meat, and drench it with distilled water; for farmed shrimps and crabs, remove nylon rope impurities; S23. Algae sample cleaning: For fresh samples, wash repeatedly with distilled water to remove insects, sand, stones and nylon rope impurities.

3. The method for pretreating marine biological samples according to claim 1, characterized in that: Step S3 includes the following steps: S31. Fish sample disassembly: For very small fish, only the internal organs are removed without cutting into sections; for small and medium fish, the internal organs are removed and then cut in half; for large fish, the internal organs are removed first, the fish head is cut off and kept separately, and then the fish is sliced ​​perpendicular to the fish bones starting from the neck, retaining the fish head, fins, scales and tail; after disassembly, the fish is weighed and recorded as the fresh sample weight; S32. Disintegration of crustacean samples: Cut crabs in half and break their claws; only drain the water from shellfish and shrimps without any other treatment; S33. Algae sample decomposition: Extra-long algae samples are cut into pieces, while short algae samples do not need to be cut.

4. The method for pretreating marine biological samples according to claim 3, characterized in that: Step S6 includes the following steps: S61, fish and crustacean samples were placed in a laboratory oven and dried at 105°C for 8-20 hours; S62, the algae sample was placed in a laboratory oven and dried at 70°C for 8-20 hours; S63. During the drying process, the sample is weighed once every 1 hour. When the weight change of the sample after two consecutive weighings is within 1%, the sample is deemed to have reached the dried and moisture-free state.

5. The method for pretreating marine biological samples according to claim 1, characterized in that: In step S4, the analysis and recognition module analyzes and recognizes the image after feature extraction, identifies the type and size of the received sample, and promptly identifies abnormal conditions; the steps include: S41. Sample type identification: S41.

1. Feature data preparation: obtaining extracted image features from the feature extraction module, including color features, texture features, and shape features; S41.2, database loading: loading the established reference sample database from the data collection module; S41.

3. Construct a species recognition model: select a support vector machine (SVM) algorithm model, use the sample features in the database as training data, and the sample types as labels to train the model; adjust the parameters of the SVM, input the extracted features of the sample to be identified into the trained SVM model, obtain the sample type prediction result, and verify the type recognition result; S42, size calculation: S42.1, image scale acquisition: in the image acquisition module, by placing a standard reference object of known size in the image, the scale of the image and the actual size is calculated; S42.2, size calculation: for the decomposed sample image, use image processing technology to perform edge detection, accurately extract the outline of the sample, and calculate the length, width and thickness of the sample based on the edge outline; S43, abnormal situation identification: set thresholds for sample damage, over-drying and deformation abnormalities respectively, obtain the weight and size data of the analysis and identification module in real time, observe the shape and texture features of the feature extraction module, and check whether it exceeds the threshold range to identify the abnormality; S44, exception handling: When the monitored weight, size or characteristics exceed the set threshold range, the alarm module is immediately triggered.

6. The method for pretreating marine biological samples according to claim 1, characterized in that: Step S5 comprises the following steps: S51, data preprocessing: obtaining historical drying data of fish, crustacean and algae samples from the data collection module, and preprocessing the data; S52, model construction: select the support vector machine (SVM) model, train the model with the preprocessed historical data, take the type, size and weight characteristics of the sample as input, and the corresponding drying time and temperature as output labels; use the back propagation algorithm in training, take the mean square error (MSE) as the loss function, continuously adjust the weights and thresholds to reduce the prediction error, evaluate the model performance through cross-validation, and adjust the parameters until the model achieves good prediction accuracy; S53. Model application and real-time adjustment: When drying a new sample, input its type, size and weight information into the trained model to obtain the predicted optimal drying time and temperature. During the drying process, use the temperature sensing module and the analysis and recognition module to obtain the temperature, weight and size change information in real time, and calculate the change rate. If the weight or size change rate deviates from the expected value, or the current temperature deviates greatly from the predicted temperature, adjust the drying time and temperature according to the preset rules.

7. The method for pretreating marine biological samples according to claim 6, characterized in that: In step S52, the drying time is calculated according to the following formula: t=∑ N i＝1 [(a i -a i * )K(x,x i )+b]+∑ N i＝1 ∑ N j＝1 [(a i -a i * )K(x i ,x j )]; K(x,x i )=exp{-∑ m k＝1 {[w k (x k -x ik ) 2 ] / (2σ 2 )}}; K(x i ,x j )=exp{-∑ m k＝1 {[w k (x ik -x jk ) 2 ] / (2σ 2 )}}; Where t represents the drying time, which is the target variable to be predicted; x is the input feature vector, which contains various characteristics of seafood, including length, width, thickness, weight and seafood type; x i is the input feature vector of the th sample in the training data set; a i and a i * is the Lagrange multiplier, which is the optimization variable obtained by solving the quadratic programming problem; K(x,x i ) is the kernel function; w k is the weight factor for different features; x k represents the kth feature of the input feature vector x; x ik represents the kth feature of the i-th sample; K(x i ,x j ) is the kernel function; x jk represents the kth feature of the jth sample.

8. The method for pretreating marine biological samples according to claim 7, characterized in that: In step S52, the drying temperature is calculated according to the following formula: T 烘干 =∑ N i＝1 [(b i -b i * )K(x,x i )+b^]+∑ N i＝1 ∑ N j＝1 [(b i -b i * )(b j -b j * )K(x i ,x j )]; where T 烘干 Indicates the drying temperature; β i and β i * represents the Lagrange multiplier, which is the optimization variable obtained by solving the quadratic programming problem; b^ is the bias term, which is used to translate the drying temperature prediction result.

9. The method for pretreating marine biological samples according to claim 1, characterized in that: Monitoring agencies include: Data collection module: collect a large number of images of seafood, covering samples at different growth stages, species characteristics and in different environments, and annotate the images as reference samples; Image acquisition module: Install high-definition cameras at key locations in the sample receiving area, sample decomposition area, and drying and drying equipment to capture clear images; Image preprocessing module: preprocess the collected images, including filtering and denoising, image enhancement, grayscale and normalization; Feature extraction module: extracts features from the preprocessed image. The extracted features include color, texture and shape. Analysis and recognition module: Analyze and recognize the image after feature extraction, identify the type and size of the received sample, and identify abnormal situations in time; Temperature sensing module: Install high-precision temperature sensors at different locations inside the drying oven and the oven to collect temperature data of each area in the oven in real time; Drying time and temperature calculation module: Using historical data, combined with the types, sizes and weight parameters of fish, crustaceans and algae samples, an evaluation model is constructed through data mining and machine learning algorithms; the optimal drying time and temperature combination is calculated. During drying, the drying time and temperature are adjusted and optimized in real time based on the temperature feedback from the temperature sensing module and the sample weight and size change information provided by the analysis and recognition module; Alarm module: including alarm, which will send out alarm in time when abnormal situation is detected; PLC control module: connected to the data collection module, image acquisition module, image preprocessing module, feature extraction module, analysis and recognition module, temperature sensing module, alarm module, drying box and drying time and temperature calculation module.

10. The method for pretreating marine biological samples according to claim 1, characterized in that: The drying equipment is a tunnel-type air duct low-temperature rapid drying equipment, which includes: a steam heater, a steam heater, an exhaust fan and a closed air duct; A steam heater is installed at the air inlet of the closed air duct to input hot dry air heated by steam, and a blower is used to force the air to be delivered; an exhaust fan is used at the outlet of the closed air duct to exhaust the air into the atmosphere.

Images