A method, apparatus and storage medium for predicting the freshness of fish meat

Abstract

This invention provides a method, apparatus, and storage medium for predicting the freshness of fish meat, belonging to the field of agricultural and livestock product quality testing technology. The method for predicting fish meat freshness includes: determining the target fish meat part, the target constant temperature value, and the target duration; inputting the target fish meat part, the target constant temperature value, and the target duration into a fish meat freshness prediction model, respectively, to obtain the freshness index value output by the fish meat freshness prediction model; wherein, the fish meat freshness prediction model is trained using preset sample fish meat parts, constant sample temperature values, preset sample durations, and freshness sample index values. It is suitable for large-scale and rapid applications requiring the detection of fish meat freshness.

Description

Technical Field

[0001] This invention relates to the field of agricultural and livestock product quality testing technology, and in particular to a method, apparatus and storage medium for predicting the freshness of fish. Background Technology

[0002] Fish meat has a high water and protein content, making it highly susceptible to microbial and enzymatic activity during storage, which can cause it to become stale or even spoil. Methods for assessing fish freshness generally rely on sensory evaluation, chemical analysis, and microbial detection. While these methods provide accurate freshness indicators, they often require specialized equipment, are time-consuming, and costly. Given my country's high annual freshwater fish production, developing a large-scale and rapid method for assessing fish freshness is a crucial issue that the industry urgently needs to address. Summary of the Invention

[0003] To address the problems existing in the prior art, the present invention provides a method, apparatus and storage medium for predicting the freshness of fish meat.

[0004] This invention provides a method for predicting the freshness of fish meat, comprising: determining a target fish meat part, a target constant temperature value, and a target duration; inputting the target fish meat part, the target constant temperature value, and the target duration into a fish meat freshness prediction model, respectively, to obtain a freshness index value output by the fish meat freshness prediction model; wherein, the fish meat freshness prediction model is trained by preset sample fish meat parts, constant sample temperature values, preset sample durations, and freshness sample index values.

[0005] According to the present invention, a method for predicting the freshness of fish meat is provided. The fish meat freshness prediction model includes an input unit, a processing unit, and an output unit. The training method of the fish meat freshness prediction model includes: inputting the sample fish meat part, a constant sample temperature value, and a preset sample duration into the input unit, and processing them sequentially through the processing unit and the output unit to output a freshness sample prediction value; comparing the freshness sample prediction value with the freshness sample index value, and adjusting the parameters of the input unit, the processing unit, and the output unit according to the comparison result until the difference between the freshness sample prediction value and the freshness sample index value is less than a threshold.

[0006] According to the present invention, a method for predicting the freshness of fish meat is provided, wherein the freshness sample index values ​​include at least the total viable bacteria count sample index value, the volatile basic nitrogen content sample index value, the freshness K value sample index value, the sensory evaluation total score sample index value, and the sensory evaluation overall acceptability index value; the processing unit includes: a fish meat total viable bacteria count prediction subunit, a fish meat volatile basic nitrogen content prediction subunit, a fish meat freshness K value prediction subunit, a fish meat sensory evaluation total score prediction subunit, and a fish meat sensory evaluation overall acceptability prediction subunit; the output unit includes... The system includes sub-units for outputting total viable bacteria count in fish, volatile basic nitrogen content in fish, freshness K-value in fish, overall sensory evaluation score of fish, and overall sensory acceptability of fish. The system inputs the sample fish part, constant sample temperature, and preset sample duration into an input unit, which then processes the data sequentially through a processing unit and an output unit to output a freshness sample prediction value. This includes inputting the sample fish part, constant sample temperature, and preset sample duration into an input unit, and then processing the data sequentially through a processing unit and an output unit. The fish meat total viable bacteria count output subunit processes the data and outputs the predicted total viable bacteria count value. The sample fish meat part, constant sample temperature, and preset sample duration are input to the input unit, and then processed sequentially by the fish meat volatile basic nitrogen content prediction subunit and the fish meat volatile basic nitrogen content output subunit to output the predicted fish meat volatile basic nitrogen content value. The sample fish meat part, constant sample temperature, and preset sample duration are input to the input unit, and then processed sequentially by the fish meat freshness K value prediction subunit and the fish meat freshness K value output subunit to output the fish meat freshness K value. The sample predicts the overall sensory evaluation score. The sample fish part, constant sample temperature, and preset sample duration are input into the input unit, and then processed sequentially by the fish meat sensory evaluation total score prediction subunit and the fish meat sensory evaluation total score output subunit to output the fish meat sensory evaluation total score sample predictive value. The sample fish part, constant sample temperature, and preset sample duration are input into the input unit, and then processed sequentially by the fish meat sensory evaluation overall acceptability prediction subunit and the fish meat sensory evaluation overall acceptability output subunit to output the fish meat sensory evaluation overall acceptability sample predictive value.

[0007] According to a method for predicting fish freshness provided by the present invention, the predicted value of a freshness sample is compared with the index value of a freshness sample, and the parameters of the input unit, processing unit, and output unit are adjusted according to the comparison result. This includes: comparing the predicted value of a total viable bacteria count sample with the index value of a total viable bacteria count sample, and adjusting the parameters of the input unit, the fish meat total viable bacteria count prediction subunit, and the fish meat total viable bacteria count output subunit according to the comparison result; comparing the predicted value of a volatile basic nitrogen content sample with the index value of a volatile basic nitrogen content sample, and adjusting the parameters of the input unit, the fish meat volatile basic nitrogen content prediction subunit, and the fish meat volatile basic nitrogen content output subunit according to the comparison result; and comparing the predicted value of a freshness K-value sample with the index value of a volatile basic nitrogen content sample. The predicted values ​​are compared with the freshness K-value sample index values, and the parameters of the input unit, the fish freshness K-value prediction subunit, and the fish freshness K-value output subunit are adjusted according to the comparison results; the predicted values ​​of the sensory evaluation total score sample are compared with the sensory evaluation total score sample index values, and the parameters of the input unit, the fish sensory evaluation total score prediction subunit, and the fish sensory evaluation total score output subunit are adjusted according to the comparison results; the predicted values ​​of the sensory evaluation overall acceptability sample are compared with the sensory evaluation overall acceptability sample index values, and the parameters of the input unit, the fish sensory evaluation overall acceptability prediction subunit, and the fish sensory evaluation overall acceptability output subunit are adjusted according to the comparison results.

[0008] According to a method for predicting fish freshness provided by the present invention, the freshness index values ​​include at least the total viable bacteria count index value, the volatile basic nitrogen content index value, the freshness K-value index value, the sensory evaluation total score index value, and the sensory evaluation overall acceptability index value. The method involves inputting the target fish meat part, the target constant temperature value, and the target duration into a fish freshness prediction model to obtain the freshness index values ​​output by the model. This includes: inputting the target fish meat part, the target constant temperature value, and the target duration into an input unit, and processing them through a fish meat total viable bacteria count prediction subunit and a fish meat total viable bacteria count output subunit to obtain the total viable bacteria count index value; inputting the target fish meat part, the target constant temperature value, and the target duration into the input unit, and processing them through a fish meat volatile basic nitrogen content prediction subunit... The system processes the target fish meat volatile basic nitrogen content and fish meat volatile basic nitrogen content output subunit to obtain the volatile basic nitrogen content index value; the target fish meat part, target constant temperature value, and target duration are input into the input unit, and processed by the fish meat freshness K value prediction subunit and fish meat freshness K value output subunit to obtain the freshness K value; the target fish meat part, target constant temperature value, and target duration are input into the input unit, and processed by the fish meat sensory evaluation total score prediction subunit and fish meat sensory evaluation total score output subunit to obtain the sensory evaluation total score; the target fish meat part, target constant temperature value, and target duration are input into the input unit, and processed by the fish meat sensory evaluation overall acceptability prediction subunit and fish meat sensory evaluation overall acceptability output subunit to obtain the sensory evaluation overall acceptability.

[0009] The present invention also provides a fish freshness prediction device, comprising: an input acquisition module for determining a target fish part, a target constant temperature value, and a target duration; and a freshness index value output module for inputting the target fish part, the target constant temperature value, and the target duration into a fish freshness prediction model to obtain a freshness index value output by the fish freshness prediction model; wherein the fish freshness prediction model is trained using preset sample fish parts, constant sample temperature values, preset sample durations, and freshness sample index values.

[0010] The present invention also provides an electronic device, including a storage device, a processor, and a computer program stored in the storage device and executable on the processor, wherein the processor executes the program to implement the fish freshness prediction method as described above.

[0011] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the fish freshness prediction method as described above.

[0012] The present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the fish freshness prediction method as described above.

[0013] The fish meat freshness prediction method, device and storage medium provided by the present invention can directly obtain the freshness index value of the target fish meat part by using the target fish meat part, the target constant temperature value and the target time, and can detect the freshness of fish meat on a large scale and quickly. Attached Figure Description

[0014] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0015] Figure 1 This is a flowchart illustrating the fish freshness prediction method provided in this embodiment of the invention.

[0016] Figure 2 This is a schematic diagram of the operation of the fish freshness prediction model provided in this embodiment of the invention.

[0017] Figure 3 This is one of the structural schematic diagrams of the fish freshness prediction model provided in the embodiments of the present invention.

[0018] Figure 4 This is the second schematic diagram of the fish freshness prediction model provided in the embodiments of the present invention.

[0019] Figure 5A This is a schematic diagram of the total colony count sample index value of grass carp back meat after storing the sample at a constant sample temperature for a preset sample duration, as provided in an embodiment of the present invention.

[0020] Figure 5B This is a schematic diagram of the total colony count sample index value of grass carp abdominal meat after storing the sample at a constant sample temperature for a preset sample duration, as provided in an embodiment of the present invention.

[0021] Figure 5C This is a schematic diagram of the total bacterial count sample index value of grass carp belly meat after storing the sample at a constant sample temperature for a preset sample duration, as provided in this embodiment of the invention.

[0022] Figure 5D This is a schematic diagram of the total bacterial count sample index value of grass carp tail meat after storing the sample at a constant sample temperature for a preset sample duration, as provided in this embodiment of the invention.

[0023] Figure 6A This is a schematic diagram of the sensory evaluation total score sample index values ​​of grass carp back meat after storing the constant sample temperature value for a preset sample time according to an embodiment of the present invention.

[0024] Figure 6BThis is a schematic diagram of the sensory evaluation total score sample index values ​​of grass carp abdominal meat after storing the constant sample temperature value for a preset sample time according to an embodiment of the present invention.

[0025] Figure 6C This is a schematic diagram of the sensory evaluation total score sample index value of grass carp belly meat after storing the constant sample temperature value for a preset sample time according to an embodiment of the present invention.

[0026] Figure 6D This is a schematic diagram of the sensory evaluation total score sample index value of grass carp tail meat after storing the constant sample temperature value for a preset sample time according to an embodiment of the present invention.

[0027] Figure 7A This is a schematic diagram of the overall acceptability sample index values ​​of the sensory evaluation of grass carp back meat after storing the sample at a constant sample temperature for a preset sample duration, as provided in an embodiment of the present invention.

[0028] Figure 7B This is a schematic diagram of the overall acceptability sample index values ​​of grass carp abdominal meat after storing samples at a constant sample temperature for a preset sample duration, as provided in an embodiment of the present invention.

[0029] Figure 7C This is a schematic diagram of the overall acceptability sample index values ​​of grass carp belly meat after storing samples at a constant sample temperature for a preset sample time, as provided in an embodiment of the present invention.

[0030] Figure 7D This is a schematic diagram of the overall acceptability sample index values ​​of grass carp tail meat after storing samples at a constant sample temperature for a preset sample time, as provided in an embodiment of the present invention.

[0031] Figure 8A This is a schematic diagram of the TVN-B content sample index values ​​of grass carp back meat after storing samples at a constant sample temperature for a preset sample duration, as provided in this embodiment of the invention.

[0032] Figure 8B This is a schematic diagram of the TVN-B content sample index values ​​of grass carp abdominal meat after storing samples at a constant sample temperature for a preset sample duration, as provided in this embodiment of the invention.

[0033] Figure 8C This is a schematic diagram of the TVN-B content sample index value of grass carp belly meat after storing the sample at a constant sample temperature for a preset sample duration, as provided in this embodiment of the invention.

[0034] Figure 8D This is a schematic diagram of the TVN-B content sample index value of grass carp tail meat after storing the sample at a constant sample temperature for a preset sample duration, as provided in an embodiment of the present invention.

[0035] Figure 9AThis is a schematic diagram of the freshness K-value of grass carp back meat after storing samples at a constant sample temperature for a preset sample duration, as provided in an embodiment of the present invention.

[0036] Figure 9B This is a schematic diagram of the freshness K-value of grass carp abdominal meat after storing it at a constant sample temperature for a preset sample duration, as provided in an embodiment of the present invention.

[0037] Figure 9C This is a schematic diagram of the freshness K-value sample index value of grass carp belly meat after storing the sample at a constant sample temperature for a preset sample time, as provided in an embodiment of the present invention.

[0038] Figure 9D This is a schematic diagram of the freshness K-value sample index value of grass carp tail meat after storing the sample at a constant sample temperature for a preset sample duration, as provided in an embodiment of the present invention.

[0039] Figure 10 This is a schematic diagram of the fish freshness prediction device provided in an embodiment of the present invention.

[0040] Figure 11 This is a schematic diagram of the structure of the electronic device provided in an embodiment of the present invention.

[0041] Reference numerals: 201: Input acquisition module; 202: Freshness index value output module; 310: Processor; 320: Communication interface; 330: Memory; 340: Communication bus. Detailed Implementation

[0042] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0043] The following is combined with Figures 1-11 This invention describes a method, apparatus, and storage medium for predicting the freshness of fish meat.

[0044] Fish, rich in protein and water, is prone to spoilage, and consuming spoiled fish can harm human health. Therefore, how to detect the freshness of fish and ensure it is in good condition for sale and consumption is a problem that needs to be solved in the fish sales process. Existing methods for detecting fish freshness generally employ chemical analysis and microbiological testing, requiring sampling and experimental procedures to obtain results, resulting in relatively low efficiency. Given my country's large fish production and sales volume, there is a need to improve the efficiency of fish freshness detection. Establishing a reliable and accurate method for predicting the freshness of different parts of fish by storing different parts of the fish at different constant temperatures for different durations, and measuring the changes in freshness index values ​​during these storage periods, is an urgent problem to be solved.

[0045] Therefore, embodiments of the present invention provide a method for predicting the freshness of fish meat. Figure 1 This is a flowchart illustrating the fish freshness prediction method provided in an embodiment of the present invention, as shown below. Figure 1 As shown, the method includes:

[0046] Step 101: Determine the target fish meat part, the target constant temperature value, and the target duration.

[0047] Specifically, the target fish fillet is a small portion of product obtained by cutting different parts of a whole fish. For example, the target fish fillet may include back fillet, belly fillet, flank fillet, and tail fillet. For instance, when grass carp is divided and stored, the obtained grass carp flank fillet is the target fish fillet.

[0048] The target constant temperature value refers to maintaining a constant storage temperature for the target fish fillet. This can be achieved through refrigeration or micro-freezing. When using refrigeration, the target fish fillet can be placed in a freezer. For example, the target constant temperature value could be 0-12 degrees Celsius.

[0049] The target duration refers to the storage time of a target fish fillet at a target constant temperature. For example, the target duration can be the general time from the acquisition of the target fish fillet to its sale. For instance, the target duration could be 0-10 days.

[0050] Step 102: Input the target fish meat part, the target constant temperature value, and the target duration into the fish meat freshness prediction model to obtain the freshness index value output by the fish meat freshness prediction model; wherein, the fish meat freshness prediction model is trained by preset sample fish meat parts, constant sample temperature values, preset sample durations, and freshness sample index values.

[0051] Specifically, the freshness index value is used to reflect the quality and preservation status of fish. For example, the freshness grade of fish can be determined based on the freshness index value, and fish with different freshness grades have different qualities and preservation status.

[0052] Before performing step 102, a fish freshness prediction model can be pre-trained. Specifically, this can be achieved by: first, collecting a large number of sample fish parts, constant sample temperature values, sample durations, and freshness index values; and then manually labeling these data to obtain multiple sets of sample fish parts, constant sample temperature values, sample durations, and freshness index values, with each set having a corresponding relationship. Next, inputting each set of sample fish parts, constant sample temperature values, sample durations, and freshness index values ​​into the initial model for training, thereby obtaining the fish freshness prediction model.

[0053] The fish freshness prediction method provided in this invention can directly obtain the freshness index value of the target fish part by using the target fish part, the target constant temperature value, and the target time, and can detect the freshness of fish on a large scale and quickly.

[0054] Compared to using chemical analysis and microbial testing to obtain the freshness index value of the target fish meat, using a fish meat freshness prediction model to obtain the freshness index value of the target fish meat does not require sampling of the target fish meat, and can obtain the freshness index value of the target fish meat without damage, without affecting the sales of the target fish meat.

[0055] The composition of fish meat varies across different parts of the fish. Therefore, pre-training a fish freshness prediction model using samples from a specific part of the fish can lead to significant biases when predicting the freshness of other target parts. For example, the belly and abdominal areas of grass carp have higher fat content than the back and tail areas, while the belly and abdominal areas have lower protein content than the back and tail areas. Protein content affects the rate of spoilage in different parts of the fish. Thus, using a fish freshness prediction model pre-trained on the back of a grass carp will result in a significant bias when predicting the freshness index of the belly. In contrast, pre-training a fish freshness prediction model using a large number of samples from different parts of the fish can reduce the bias in predicting the freshness index of different target parts and improve the accuracy of the predicted freshness index.

[0056] Based on the above embodiments, the fish freshness prediction model includes: an input unit, a processing unit, and an output unit. The training method for the fish freshness prediction model includes:

[0057] Step 201: Input the sample fish meat part, constant sample temperature value, and preset sample duration into the input unit, and process them sequentially through the processing unit and the output unit to output the freshness sample prediction value.

[0058] Specifically, the input unit is the part of the fish freshness prediction model that receives raw data, the processing unit is the part of the fish freshness prediction model that performs data processing and feature extraction, and the output unit is the part of the fish freshness prediction model that generates predicted freshness index values.

[0059] Understandably, at least the input sample fish meat parts, constant sample temperature values, and preset sample durations need to undergo data standardization processing to convert these data into the data format required by the processing unit. For example... Figure 2 and Figure 3 As shown, data standardization can be performed through a separate data standardization unit, or it can be performed in the input unit or the processing unit.

[0060] Step 202: Compare the predicted value of the freshness sample with the freshness sample index value, and adjust the parameters of the input unit, processing unit and output unit according to the comparison results until the difference between the predicted value of the freshness sample and the freshness sample index value is less than the threshold.

[0061] The difference between the predicted value and the freshness sample index value can be either the difference between them or the ratio between them. The threshold can be a preset tolerance range; an example threshold could be -20% to +20%.

[0062] When the difference between the predicted value and the index value of a freshness sample is the ratio between the two, the difference between the predicted value and the index value of a freshness sample can be obtained by the following formula, for example.

[0063]

[0064] Where RE is the difference between the predicted value and the index value of the freshness sample; PV is the predicted value of the freshness sample; and EV is the index value of the freshness sample.

[0065] The fish freshness prediction method provided in this invention can evaluate the model's performance in the current state by comparing the predicted value of the freshness sample with the freshness sample index value, quantify the model's prediction deviation in the current state, and control the prediction deviation of the trained fish freshness prediction model by making the difference between the predicted value of the freshness sample and the freshness sample index value less than a threshold, thereby improving the accuracy of predicting the freshness index value of the target fish part.

[0066] After step 102, the method further includes inputting the fish meat part of the test sample, the constant temperature value of the test sample, and the preset test sample duration into the input unit, and processing them sequentially through the processing unit and the output unit to output the predicted value of the test freshness sample; comparing the predicted value of the test freshness sample with the test freshness sample index value, and adjusting the parameters of the input unit, processing unit, and output unit according to the comparison result until the difference between the predicted value of the test freshness sample and the test freshness sample index value is less than a threshold.

[0067] Specifically, the range of the constant sample temperature value tested does not overlap with the range of the aforementioned constant sample temperature values. In this way, by comparing the predicted value of the test freshness sample under a test constant sample temperature value that is at least different from the constant sample temperature value with the test freshness sample index value, the performance of the model under the current state can be better evaluated, and the prediction bias of the model under the current state can be quantified, so as to improve the accuracy of predicting the freshness index value of the target fish meat part.

[0068] like Figure 4 As shown, based on any of the above embodiments, the freshness sample index values ​​include at least the total viable count (TVC) sample index value, the total volatile base nitrogen (TVB-N) content sample index value, the freshness K value sample index value, the sensory evaluation total score sample index value, and the sensory evaluation overall acceptability index value.

[0069] Specifically, regarding the total viable count sample index, for example, 5 grams of fish meat sample can be weighed in a sterile laminar flow hood. The fish meat sample is placed in a sterile homogenization bag or sterile sampling bag, and 45 ml of sterile saline solution with a mass concentration of 0.85% (w / v) is added to the bag. The fish meat sample is tapped continuously for 15 seconds to ensure thorough mixing and homogenization with the sterile saline solution. The homogenized liquid from the fish meat sample is obtained, and 2-3 appropriate dilution gradients are selected based on the microbial growth of the sample. The homogenized liquid is then serially diluted 10-fold using sterile physiological saline. 100 μL of the diluted solution is added to the surface of a plate counting agar and spread evenly, with two plates for each dilution. The plates are then incubated aerobically at 30°C for 48 hours. Plates with colony counts between 30 and 300 are selected for colony counting to obtain the total viable count of the fish meat sample. The colony count unit is log CFU / g, and the sample fish meat in the bag can be patted using a patting homogenizer.

[0070] For example, regarding the volatile basic nitrogen content sample index value, 2.5 grams of sample fish meat can be weighed, and the minced sample fish meat can be mixed with 25 ml of distilled water in a centrifuge tube. Homogenization can be performed for 1 minute, followed by low-speed shaking for 30 minutes and centrifugation at 4000 rpm for 5 minutes. 5 ml of the supernatant can be mixed with a 10 g / L magnesium oxide solution in an equal proportion. The mixture can be distilled using a Kjeldahl nitrogen analyzer for 5 minutes, and the distillate can be absorbed in an Erlenmeyer flask containing 10 ml of 20 g / L boric acid solution and 100 μL of 2 g / L methyl red-methylene blue indicator. The absorbent solution can be titrated with a 0.02 mol / L hydrochloric acid standard solution until it turns blue-purple, and the volume of hydrochloric acid standard solution used can be recorded. Distilled water can be used as a blank control instead of the supernatant. The volatile basic nitrogen content sample index value of the sample fish meat can then be obtained.

[0071] For example, regarding the K-value of freshness, 1 gram of sample fish meat can be weighed and homogenized with 2 mL of 10% (w / w) perchloric acid solution, followed by centrifugation at 5000 rpm for 3 minutes. The supernatant is collected in a centrifuge tube, and 2 mL of 5% (w / w) PCA solution is added to the precipitate. The mixture is then centrifuged at 5000 rpm for 3 minutes, and this process is repeated twice. The supernatants are combined. The pH of the supernatant is adjusted to 6.4 ± 0.05 with 10 mol / L KOH and centrifuged at 5000 rpm for 3 minutes. The supernatant is then collected. The volume is brought to 10 mL with neutralized PCA solution to obtain the ATP-related extract. The ATP-related extract and ATP-related standards are filtered through a 0.22 μm aqueous ultrafiltration membrane and analyzed by high-performance liquid chromatography (HPLC). HPLC was used to monitor ATP-related compounds in the solution using a COSMOSIL 5C18-PAQ liquid chromatography column (4.6ID × 250 mm × 5 μm). The chromatographic conditions were as follows: detection wavelength 254 nm; column temperature 30 °C; injection volume 50 μL; mobile phase A: 0.05 mol / L Na₂HPO₄-NaH₂PO₄ buffer (pH 6.8); mobile phase B: chromatographically pure methanol; flow rate: mL / min. The content of each related compound in the sample was calculated using a standard curve of peak area versus concentration of standards. The freshness K value can be calculated using the following formula.

[0072]

[0073] Among them, H X Hypoxanthine, H X R stands for inosine nucleoside, ATP stands for adenosine triphosphate (ATP), ADP stands for adenosine diphosphate (ADP), AMP stands for adenosine monophosphate (AMP), and IMP stands for inosine monophosphate (IMP).

[0074] The freshness K-value is obtained by measuring the changes in the content of ATP and its decomposition products in the fish meat of a sample. For example, fish meat with a freshness K-value of less than 20% is fresh, while fish meat with a freshness K-value of more than 60% is not fresh.

[0075] For example, the total sensory evaluation score sample index value and the overall sensory evaluation acceptability index value can be provided by 6 women and 6 men who have undergone sensory training, based on the texture, color, smell and tissue morphology of the sample fish meat.

[0076] For example, the sensory evaluation scoring rules for the sample fish meat parts can be found in Table 1.

[0077] Table 1. Sensory evaluation scoring rules for fish meat parts of the samples.

[0078]

[0079] The processing unit includes: a sub-unit for predicting the total viable bacteria count of fish meat, a sub-unit for predicting the volatile basic nitrogen content of fish meat, a sub-unit for predicting the K-value of fish meat freshness, a sub-unit for predicting the total sensory evaluation score of fish meat, and a sub-unit for predicting the overall acceptability of fish meat sensory evaluation.

[0080] The output units include: total viable bacteria count of fish meat output subunit, volatile basic nitrogen content of fish meat output subunit, freshness K value of fish meat output subunit, total sensory evaluation score of fish meat output subunit, and overall sensory acceptance of fish meat output subunit.

[0081] The sample fish meat part, constant sample temperature value, and preset sample duration are input into the input unit, and then processed sequentially through the processing unit and the output unit to output the freshness sample prediction value. This includes: inputting the sample fish meat part, constant sample temperature value, and preset sample duration into the input unit, and then processing sequentially through the fish meat total viable bacteria count prediction subunit and the fish meat total viable bacteria count output subunit to output the total viable bacteria count sample prediction value.

[0082] The sample fish meat part, constant sample temperature value, and preset sample duration are input into the input unit, and then processed by the fish meat volatile basic nitrogen content prediction subunit and the fish meat volatile basic nitrogen content output subunit to output the sample predicted value of fish meat volatile basic nitrogen content.

[0083] The sample fish meat part, constant sample temperature value, and preset sample duration are input into the input unit, and then processed by the fish meat freshness K value prediction subunit and the fish meat freshness K value output subunit in sequence to output the fish meat freshness K value sample prediction value.

[0084] The sample fish meat part, constant sample temperature value, and preset sample duration are input into the input unit, and then processed by the fish meat sensory evaluation total score prediction subunit and the fish meat sensory evaluation total score output subunit to output the fish meat sensory evaluation total score sample prediction value.

[0085] The sample fish meat part, constant sample temperature value, and preset sample duration are input into the input unit, and then processed by the overall acceptability prediction subunit and the overall acceptability output subunit of fish meat sensory evaluation in sequence, and the overall acceptability sample value of fish meat sensory evaluation is output.

[0086] The sample fish meat parts, constant sample temperature, and preset sample duration are the same input values; however, the total viable bacteria count, volatile basic nitrogen content, freshness K value, sensory evaluation total score, and sensory evaluation overall acceptability are different freshness sample index values. Training a fish freshness prediction model with a single processing unit using the same sample fish meat parts, constant sample temperature, preset sample duration, and different freshness sample index values ​​requires overcoming the conflicts in the process of predicting different freshness sample index values, which increases the training difficulty and complexity of the fish freshness prediction model.

[0087] The fish freshness prediction method provided in this invention predicts the corresponding freshness sample index values ​​for different parts of the fish, constant sample temperature, and preset sample duration through different freshness sample index value prediction sub-units. This can reduce the training difficulty and complexity of the fish freshness prediction model.

[0088] Furthermore, using the same sample fish meat parts, constant sample temperature values, preset sample durations, and different freshness sample index values, the fish meat freshness prediction model with a single processing unit trained may reduce the accuracy of the fish meat freshness prediction model in predicting the value of a single freshness sample in order to balance the predicted values ​​of different freshness samples. In order to ensure the accuracy of the fish meat freshness prediction model in comprehensively predicting the value of each freshness sample, the fish meat freshness prediction method provided in this embodiment of the invention can avoid the above-mentioned problems.

[0089] like Figure 3 As shown, based on any of the above embodiments, the predicted value of the freshness sample is compared with the freshness sample index value, and the parameters of the input unit, processing unit and output unit are adjusted according to the comparison result, including: comparing the predicted value of the total viable bacteria sample with the total viable bacteria sample index value, and adjusting the parameters of the input unit, the fish meat total viable bacteria prediction subunit and the fish meat total viable bacteria output subunit according to the comparison result.

[0090] The predicted values ​​of volatile basic nitrogen content samples are compared with the index values ​​of volatile basic nitrogen content samples, and the parameters of the input unit, the fish meat volatile basic nitrogen content prediction subunit, and the fish meat volatile basic nitrogen content output subunit are adjusted according to the comparison results.

[0091] The predicted values ​​of the freshness K-value samples are compared with the sample index values ​​of the freshness K-value samples, and the parameters of the input unit, the fish meat freshness K-value prediction subunit, and the fish meat freshness K-value output subunit are adjusted according to the comparison results.

[0092] The predicted value of the total sensory evaluation score sample is compared with the total sensory evaluation score sample index value, and the parameters of the input unit, the fish meat sensory evaluation total score prediction sub-unit, and the fish meat sensory evaluation total score output sub-unit are adjusted according to the comparison results.

[0093] The predicted values ​​of the overall acceptability of sensory evaluation samples are compared with the index values ​​of the overall acceptability of sensory evaluation samples, and the parameters of the input unit, the prediction sub-unit of the overall acceptability of fish sensory evaluation, and the output sub-unit of the overall acceptability of fish sensory evaluation are adjusted according to the comparison results.

[0094] The method of comparing the predicted value of the total viable bacteria sample with the index value of the total viable bacteria sample to obtain the degree of difference between the predicted value of the total viable bacteria sample and the index value of the total viable bacteria sample is basically the same as the aforementioned method of comparing the predicted value of the freshness sample with the index value of the freshness sample to obtain the degree of difference between the predicted value of the freshness sample and the index value of the freshness sample, and will not be repeated here.

[0095] The fish freshness prediction method provided in this invention, which evaluates the performance of the fish total viable bacteria count prediction subunit and the fish total viable bacteria count output subunit under the current state by using the predicted value and index value of the total viable bacteria count sample, quantifies the prediction deviation of the fish total viable bacteria count prediction subunit under the current state, and adjusts the parameters of the fish total viable bacteria count prediction subunit and the fish total viable bacteria count output subunit by using the prediction deviation, so that the difference between the predicted value and the index value of the total viable bacteria count sample is less than a threshold, thereby improving the accuracy of predicting the total viable bacteria count of the target fish part.

[0096] The working principle and technical effect of the fish meat volatile basic nitrogen content prediction subunit and fish meat volatile basic nitrogen content output subunit, fish meat volatile basic nitrogen content prediction subunit and fish meat volatile basic nitrogen content output subunit, fish meat sensory evaluation total score prediction subunit and fish meat sensory evaluation total score output subunit, fish meat sensory evaluation overall acceptability prediction subunit and fish meat sensory evaluation overall acceptability output subunit are basically the same as those of the fish meat total viable bacteria count prediction subunit and fish meat total viable bacteria count output subunit, and will not be repeated here.

[0097] Based on any of the above embodiments, the freshness index values ​​include at least the total viable bacteria count index value, the volatile basic nitrogen content index value, the freshness K value index value, the sensory evaluation total score index value, and the sensory evaluation overall acceptability index value.

[0098] The target fish meat part, target constant temperature value, and target duration are respectively input into the fish meat freshness prediction model to obtain the freshness index value output by the fish meat freshness prediction model. This includes: inputting the target fish meat part, target constant temperature value, and target duration into the input unit, and processing them through the fish meat total viable bacteria count prediction subunit and the fish meat total viable bacteria count output subunit to obtain the total viable bacteria count index value.

[0099] The target fish meat part, target constant temperature value, and target duration are input into the input unit, and then processed by the fish meat volatile basic nitrogen content prediction subunit and the fish meat volatile basic nitrogen content output subunit to obtain the volatile basic nitrogen content index value.

[0100] The target fish part, target constant temperature value, and target duration are input into the input unit, and then processed by the fish freshness K value prediction subunit and the fish freshness K value output subunit to obtain the freshness K value.

[0101] The target fish meat part, target constant temperature value, and target duration are input into the input unit, and processed by the fish meat sensory evaluation total score prediction subunit and the fish meat sensory evaluation total score output subunit to obtain the sensory evaluation total score.

[0102] The target fish meat part, target constant temperature value, and target duration are input into the input unit, and then processed by the overall acceptability prediction subunit and the overall acceptability output subunit to obtain the overall acceptability of the sensory evaluation.

[0103] The fish freshness prediction method provided in this invention can directly obtain various freshness index values ​​of the target fish part by using the target fish part, the target constant temperature value, and the target time, and can detect the freshness of fish on a large scale and quickly.

[0104] Compared to using a fish freshness prediction model with a single processing unit to predict different freshness index values, the fish freshness prediction provided in this embodiment of the invention uses a fish freshness prediction model with multiple processing units having different sub-units. By using different sub-units to predict the corresponding freshness index values, better performance can be obtained and the accuracy of predicting each freshness index value of the target fish part can be improved.

[0105] To illustrate the fish freshness prediction method provided in this embodiment, a specific example is given below. Before training the grass carp freshness prediction model for different target fish parts, it is necessary to first obtain the sample fish meat parts, constant sample temperature values, preset sample duration, and freshness sample prediction values. The sample fish meat parts correspond to different parts of the grass carp.

[0106] A fish freshness prediction model can be trained using Support Vector Regression (SVR), based on the principles of Support Vector Machine (SVM), by considering the sample fish meat parts, constant sample temperature, preset sample duration, and freshness sample prediction values. SVR, by finding an optimal function to predict continuous variables with the smallest possible error, exhibits good generalization ability. Furthermore, SVR maintains good predictive performance even with limited sample sizes, making it suitable for small datasets. The SVR model also demonstrates good predictive performance for indicators such as the overall sensory evaluation score, total viable bacteria count (TVC), volatile basic nitrogen (TVB-N), freshness K-value, pH value, and overall sensory acceptance index during the storage of aquatic products.

[0107] Specifically, 96 live grass carp weighing 2.01-2.07 kg and measuring 53.8-58 cm in length were transported live from the seafood market to the laboratory using oxygenated transport bags. After the fish stabilized somewhat, they were stunned by tapping their heads, and then scaled, gutted, and cleaned. Subsequently, the heads were removed, and the fish were cut into four parts: back, belly, belly, and tail, to obtain samples of the fish meat.

[0108] Set 0°C, 3°C, 6°C, 9°C, and 12°C as constant sample temperatures. For example, constant temperature environmental conditions can be obtained through 5 freezers, and during constant temperature storage, each sample of grass carp flesh is placed in a separate storage drawer.

[0109] At 0 degrees Celsius, the preset sample duration can be set to 0 days, 2 days, 4 days, 6 days, 8 days, and 10 days; at 3 degrees Celsius, the preset sample duration can be set to 0 days, 1 day, 2 days, 4 days, 6 days, and 8 days; at 6 degrees Celsius, the preset sample duration can be set to 0 days, 1 day, 2 days, 3 days, 4 days, and 5 days; at 9 degrees Celsius, the preset sample duration can be set to 0 days, 1 day, 2 days, 3 days, 4 days, and 5 days; at 12 degrees Celsius, the preset sample duration can be set to 0 days, 0.5 days, 1 day, 1.5 days, 2 days, 2.5 days, 3 days, and 4 days.

[0110] After storing the grass carp meat at a constant temperature for a predetermined period of time, the total viable bacteria count, sensory evaluation score, overall sensory acceptance, volatile basic nitrogen content, and freshness K value were measured for the back meat, belly meat, belly meat, and tail meat of the grass carp. The freshness index data for different parts of the grass carp at a constant storage temperature were obtained, as shown in Figures 5-9. The methods for detecting the total viable bacteria count, sensory evaluation score, overall sensory acceptance, volatile basic nitrogen content, and freshness K value can be found in the aforementioned implementation method and will not be repeated here.

[0111] Specifically, as shown in Figure 5, when stored at 0–12 degrees Celsius, the initial TVC of each sample fish meat portion was 4.03–5.06 log CFU / g, and it showed a continuous increasing trend with the extension of storage time. Compared with 0 degrees Celsius, the growth rate of microorganisms at 12 degrees Celsius was significantly faster. The higher the temperature, the faster the growth rate of microorganisms. When the TVC in the sample fish meat portion exceeded 7.0 log CFU / g, it meant that the number of microorganisms had exceeded the standard, and the sample fish meat portion had spoiled and was inedible.

[0112] As shown in Figures 6 and 7, the initial total sensory evaluation scores for each sample fish meat portion ranged from 89.00 to 96.70, indicating that the sensory quality of each sample fish meat portion was excellent before storage. During storage, the total sensory evaluation scores of the sample fish meat portions decreased with increasing storage temperature, showing a trend of decreasing faster at higher storage temperatures. When the total sensory evaluation score was <40, it indicated that the average scores for texture, color, odor, tissue morphology, and overall acceptability of the sample fish meat portion were already low, and its edibility was low; while when the total sensory evaluation score was <20, the sensory condition of the sample fish meat portion was extremely poor, and its edibility was extremely low, meaning the fish meat was inedible. According to the sensory evaluation table for sample fish meat portions, when the overall acceptability score was below 8, the overall sensory quality of the sample fish meat portion was considered poor, and its edibility was low; while when the overall acceptability score was below 4, the overall sensory quality of the sample fish meat portion was considered extremely poor, and its edibility was extremely low. The overall acceptability of the initial sensory evaluation for each group ranged from 17.30 to 19.30, indicating that the overall sensory quality of the fish meat in the samples was good and the edibility was high before storage. During storage, the overall sensory acceptability of each group decreased rapidly with the increase of storage temperature.

[0113] As shown in Figure 8, the TVB-N content increased at a faster rate with increasing storage temperature, and the order and severity of spoilage in different parts of the fish meat varied at different temperatures. The grass carp belly meat showed the most severe spoilage when stored at 0 degrees Celsius.

[0114] As shown in Figure 9, the changes in the freshness K-value of the fish meat samples during storage at 0–12 degrees Celsius are illustrated in Figures 5 and 6. The initial freshness K-values ​​of the fish meat samples ranged from 15.65% to 19.99%, indicating high freshness. With increasing storage time, the freshness K-values ​​of each group increased significantly. After 10 days of storage at 0 degrees Celsius, the freshness K-values ​​of the grass carp belly meat, grass carp belly meat, and grass carp tail meat all exceeded 60%, reaching 63.56%, 60.55%, and 61.91%, respectively. At 3 degrees Celsius, the freshness K-values ​​of the grass carp belly meat, grass carp belly meat, and grass carp tail meat exceeded the limit after 6 days of storage, while after 8 days of storage, the freshness K-values ​​of the grass carp belly meat, grass carp belly meat, and grass carp tail meat reached 66.05%, 68.32%, and 65.00%, respectively. When stored at 6 degrees Celsius, the freshness K-value of grass carp belly meat and grass carp belly meat exceeded the limit after 4 days of storage, and subsequently, the freshness K-value of grass carp tail meat exceeded the limit after 5 days of storage. When stored at 9 degrees Celsius, the freshness K-value of grass carp tail meat and grass carp belly meat exceeded the limit after 4 days of storage, and after 5 days of storage, the freshness K-values ​​of grass carp back meat, grass carp belly meat, grass carp belly meat, and grass carp tail meat all exceeded the limit, at 70.55%, 70.94%, 62.84%, and 63.94%, respectively. When stored at 12 degrees Celsius, the freshness K-value of grass carp tail meat exceeded the limit after 2.5 days (60 hours) of storage. After 3 days (72 hours) of storage, the freshness K-values ​​of grass carp back meat, grass carp belly meat, and grass carp belly meat all exceeded the limit. On the 4th day (96 hours) of storage, the freshness K-values ​​of grass carp back meat, grass carp belly meat, grass carp belly meat, and grass carp tail meat were 70.06%, 71.02%, 68.48%, and 69.53%, respectively.

[0115] The sample fish meat part, constant sample temperature value, and preset sample duration are input into the input unit, and each prediction subunit and output subunit outputs the corresponding freshness index value in sequence. The parameters of the input unit, each prediction subunit, and the output subunit are adjusted according to the comparison results until the difference between the predicted value of each freshness sample and the corresponding freshness sample index value is less than the threshold.

[0116] For example, with a constant sample temperature of 4 degrees Celsius, as shown in Table 2, the predicted REs for TVC are all between -10% and 10%, indicating that the constructed SVR model has high prediction accuracy for TVC of the target fish meat and can accurately predict the total number of colonies of the target fish meat during storage at 0 to 12 degrees Celsius.

[0117] Table 2 shows the total bacterial count of samples stored at a constant temperature of 4 degrees Celsius.

[0118]

[0119] As shown in Tables 3 and 4, the REs for the overall sensory evaluation score and the overall sensory acceptability are both between -10% and 10%, indicating that the constructed SVR model can accurately predict the overall sensory evaluation score and overall acceptability of the target fish meat during storage at 0 to 12 degrees Celsius.

[0120] Table 3. Sensory evaluation total scores during storage at a constant sample temperature of 4 degrees Celsius.

[0121]

[0122] Table 4. Sensory evaluation of overall acceptability during storage at a constant sample temperature of 4 degrees Celsius.

[0123] As shown in Table 5, the prediction accuracy (RE) for TVB-N content was between -20% and 20%, with 33.33% of the TVB-N content predictions having an RE between -10% and 10%. This indicates that the constructed SVR model has high prediction accuracy for the TVB-N content of the target fish meat.

[0124] Table 5. TVB-N content during storage at a constant sample temperature of 4 degrees Celsius.

[0125]

[0126] As shown in Table 6, the RE values ​​predicted by the freshness K value are all between -20% and 20%, of which 62.50% of the freshness K value RE values ​​are between -10% and 10%, indicating that the constructed SVR model can accurately predict the freshness K value of the target fish meat part during storage at 0 to 12 degrees Celsius.

[0127] Table 6. Freshness K-values ​​of samples stored at a constant temperature of 4 degrees Celsius.

[0128]

[0129] In training a grass carp freshness prediction model for different target parts of the fish using Support Vector Regression (SVR) based on Support Vector Machine (SVM) principles, necessary libraries can be imported first. These include pandas for data processing and analysis, the function `train_test_split` for splitting data into training and test sets, and `OneHotEncoder` for one-hot encoding of categorical variables (such as storage location). Pandas facilitates reading and manipulating data, such as loading Excel files containing preset sample fish parts, constant sample temperature values, preset sample durations, and freshness index values. `train_test_split` facilitates model training and evaluation. OneHotEncoder facilitates the transformation of input into a numerical format suitable for machine learning algorithms. For example, initializing OneHotEncoder with the sparse parameter set to False allows it to return a dense array. The encoder function encoder.fit_transform performs one-hot encoding on the parts column (storage location) of all data, converting it to a numerical format. Let parts_encoded = encoder.fit_transform(storage location); parts_encoded is the one-hot encoded numerical format. For instance, if there are three types of fish meat in a sample—back, belly, and tail—they will be converted into three separate columns, each corresponding to a location. The encoded array is then converted into a pandas dataframe for easier subsequent processing. The storage location can also be referred to as the type of fish meat in the sample.

[0130] The pandas.concat function can be used to merge one-hot encoded site features with other numerical features, such as storage temperature and storage time, to form a complete feature set. For example, features_encoded = pd.concat(storage site, storage temperature, storage time).

[0131] More function libraries can also be imported, such as SVR, mean_squared_error, and sqrt. SVR is a support vector regression model that can predict continuous values; mean_squared_error can evaluate the mean squared error of the model's prediction results; sqrt can calculate the root mean square error (RMSE), which can evaluate the prediction accuracy.

[0132] The `train_evaluate_svr` function can be used to train and evaluate a support vector regression model, splitting the features and target variables into a training set (80%) and a test set (20%). The SVR model is initialized and trained using a radial basis function (RBF). Predictions are made on the test set, and the root mean square error (RMSE), the square root of the error between the predicted and actual values, is calculated to measure the model's performance.

[0133] The `targets` list contains multiple freshness metrics. An empty dictionary `models` can be initialized to store the trained model and evaluation results. The following operations can be performed on each target variable: filter out rows containing missing target values; extract the corresponding features and target variable; call the `train_evaluate_svr` function for training and evaluation; and store the trained model and RMSE in the `models` dictionary.

[0134] The `predict_new_data` function can be used to use a trained model to predict the target fish part, the target constant temperature value, and the target duration, and output the corresponding freshness index value.

[0135] The grass carp freshness prediction method for different target parts provided in this invention uses support vector regression to train a grass carp freshness prediction model for different target parts. This model can handle high-dimensional data and nonlinear relationships, thereby providing high-precision freshness prediction results. The rate and pattern of spoilage of grass carp meat in different parts of the fish vary under storage conditions. This grass carp freshness prediction model can perform refined freshness prediction for different parts of the grass carp, rather than just the back meat or the whole fish. This increases the accuracy of the prediction results and provides a comprehensive assessment of the quality of different parts of the grass carp meat.

[0136] The fish freshness prediction device provided by the present invention is described below. The fish freshness prediction device described below and the fish freshness prediction method described above can be referred to in correspondence.

[0137] Figure 10 This is a schematic diagram of the fish freshness prediction device provided by the present invention, as shown below. Figure 10 The device shown includes an input acquisition module 201, used to determine the target fish meat part, the target constant temperature value, and the target duration; and a freshness index value output module 202, used to input the target fish meat part, the target constant temperature value, and the target duration into the fish meat freshness prediction model, and obtain the freshness index value output by the fish meat freshness prediction model; wherein, the fish meat freshness prediction model is trained by preset sample fish meat parts, constant sample temperature values, preset sample durations, and freshness sample index values.

[0138] The fish freshness prediction device provided by this invention can directly obtain the freshness index value of the target fish part by using the target fish part, the target constant temperature value and the target time, and can detect the freshness of fish on a large scale and quickly.

[0139] It is understood that the fish freshness prediction device provided by the present invention corresponds to the fish freshness prediction device provided in the above embodiments. The relevant technical features of the fish freshness prediction device provided by the present invention can be referred to the relevant technical features of the fish freshness prediction device provided in the above embodiments, and will not be repeated here.

[0140] Figure 11 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 11 As shown, the electronic device may include a processor 310, a communication interface 320, a memory 330, and a communication bus 340. The processor 310, communication interface 320, and memory 330 communicate with each other via the communication bus 340. The processor 310 can call logical instructions in the memory 330 to execute a fish freshness prediction method. This method includes: determining the target fish part, the target constant temperature value, and the target duration; inputting the target fish part, the target constant temperature value, and the target duration into a fish freshness prediction model to obtain a freshness index value output by the fish freshness prediction model; wherein the fish freshness prediction model is trained using preset sample fish parts, constant sample temperature values, preset sample durations, and freshness sample index values.

[0141] Furthermore, the logical instructions in the aforementioned storage device 330 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, essentially, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0142] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the fish freshness prediction method provided by the above methods. The method includes: determining a target fish part, a target constant temperature value, and a target duration; inputting the target fish part, the target constant temperature value, and the target duration into a fish freshness prediction model, respectively, to obtain a freshness index value output by the fish freshness prediction model; wherein the fish freshness prediction model is trained by preset sample fish parts, constant sample temperature values, preset sample durations, and freshness sample index values.

[0143] In another aspect, the present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon. When executed by a processor, the computer program implements the fish freshness prediction method provided by the above methods. The method includes: determining a target fish part, a target constant temperature value, and a target duration; inputting the target fish part, the target constant temperature value, and the target duration into a fish freshness prediction model, respectively, to obtain a freshness index value output by the fish freshness prediction model; wherein the fish freshness prediction model is trained using preset sample fish parts, constant sample temperature values, preset sample durations, and freshness sample index values.

[0144] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0145] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0146] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for predicting the freshness of fish meat, characterized in that, include: Determine the target fish fillet part, the target constant temperature value, and the target duration; The target fish meat part, the target constant temperature value, and the target duration are respectively input into the fish meat freshness prediction model to obtain the freshness index value output by the fish meat freshness prediction model. The fish freshness prediction model is trained using preset sample fish parts, constant sample temperature values, preset sample duration, and freshness sample index values. The fish freshness prediction model includes: an input unit, a processing unit, and an output unit; The training method for the fish freshness prediction model includes: The sample fish meat part, constant sample temperature value, and preset sample duration are input into the input unit, and then processed by the processing unit and the output unit in sequence to output the freshness sample prediction value. The predicted value of the freshness sample is compared with the freshness sample index value, and the parameters of the input unit, processing unit and output unit are adjusted according to the comparison result until the difference between the predicted value of the freshness sample and the freshness sample index value is less than the threshold. The freshness sample index values ​​include at least the total viable bacteria count sample index value, the volatile basic nitrogen content sample index value, the freshness K value sample index value, the sensory evaluation total score sample index value, and the sensory evaluation overall acceptability index value. The processing unit includes: a sub-unit for predicting the total viable bacteria count of fish meat, a sub-unit for predicting the volatile basic nitrogen content of fish meat, a sub-unit for predicting the K-value of fish meat freshness, a sub-unit for predicting the total sensory evaluation score of fish meat, and a sub-unit for predicting the overall acceptability of fish meat sensory evaluation. The output unit includes: a sub-unit for outputting the total viable bacteria count of fish meat, a sub-unit for outputting the volatile basic nitrogen content of fish meat, a sub-unit for outputting the K-value of fish meat freshness, a sub-unit for outputting the total sensory evaluation score of fish meat, and a sub-unit for outputting the overall acceptability of fish meat sensory evaluation. The sample fish meat portion, constant sample temperature value, and preset sample duration are input into the input unit, and then processed sequentially by the processing unit and the output unit to output a freshness sample prediction value, including: The sample fish meat part, constant sample temperature value, and preset sample duration are input into the input unit, and then processed by the fish meat total viable bacteria count prediction subunit and the fish meat total viable bacteria count output subunit in sequence to output the total viable bacteria count sample prediction value. The sample fish meat part, constant sample temperature value, and preset sample duration are input into the input unit, and then processed sequentially by the fish meat volatile basic nitrogen content prediction subunit and the fish meat volatile basic nitrogen content output subunit to output the sample predicted value of fish meat volatile basic nitrogen content. The sample fish meat part, constant sample temperature value, and preset sample duration are input into the input unit, and then processed sequentially by the fish meat freshness K value prediction subunit and the fish meat freshness K value output subunit to output the fish meat freshness K value sample prediction value. The sample fish meat part, constant sample temperature value, and preset sample duration are input into the input unit, and then processed by the fish meat sensory evaluation total score prediction subunit and the fish meat sensory evaluation total score output subunit in sequence to output the fish meat sensory evaluation total score sample prediction value. The sample fish meat part, constant sample temperature value, and preset sample duration are input into the input unit, and then processed sequentially by the fish meat sensory evaluation overall acceptability prediction subunit and the fish meat sensory evaluation overall acceptability output subunit to output the fish meat sensory evaluation overall acceptability sample prediction value. The predicted value of the freshness sample is compared with the freshness sample index value, and the parameters of the input unit, processing unit, and output unit are adjusted according to the comparison result, including: The predicted value of the total viable bacteria count sample is compared with the index value of the total viable bacteria count sample, and the parameters of the input unit, the fish meat total viable bacteria count prediction subunit, and the fish meat total viable bacteria count output subunit are adjusted according to the comparison results. The predicted value of volatile basic nitrogen content sample is compared with the index value of volatile basic nitrogen content sample, and the parameters of the input unit, the fish meat volatile basic nitrogen content prediction subunit, and the fish meat volatile basic nitrogen content output subunit are adjusted according to the comparison results. The predicted value of the freshness K value sample is compared with the freshness K value sample index value, and the parameters of the input unit, the fish freshness K value prediction subunit, and the fish freshness K value output subunit are adjusted according to the comparison results. The predicted value of the total sensory evaluation score sample is compared with the index value of the total sensory evaluation score sample, and the parameters of the input unit, the fish meat sensory evaluation total score prediction subunit, and the fish meat sensory evaluation total score output subunit are adjusted according to the comparison results. The predicted value of the overall acceptability of sensory evaluation samples is compared with the index value of the overall acceptability of sensory evaluation samples, and the parameters of the input unit, the prediction subunit of overall acceptability of fish sensory evaluation, and the output subunit of overall acceptability of fish sensory evaluation are adjusted according to the comparison results.

2. The method for predicting fish freshness according to claim 1, characterized in that, The freshness index values ​​include at least the total viable bacteria count index value, the volatile basic nitrogen content index value, the freshness K value index value, the sensory evaluation total score index value, and the sensory evaluation overall acceptability index value. The target fish fillet part, target constant temperature value, and target duration are respectively input into the fish freshness prediction model to obtain the freshness index value output by the fish freshness prediction model, including: The target fish meat part, target constant temperature value, and target duration are input into the input unit, and processed by the fish meat total viable bacteria count prediction subunit and the fish meat total viable bacteria count output subunit to obtain the total viable bacteria count index value. The target fish meat part, target constant temperature value, and target duration are input into the input unit, and processed by the fish meat volatile basic nitrogen content prediction subunit and the fish meat volatile basic nitrogen content output subunit to obtain the volatile basic nitrogen content index value. The target fish meat part, target constant temperature value, and target duration are input into the input unit, and processed by the fish meat freshness K value prediction subunit and the fish meat freshness K value output subunit to obtain the freshness K value; The target fish meat part, target constant temperature value, and target duration are input into the input unit, and processed by the fish meat sensory evaluation total score prediction subunit and the fish meat sensory evaluation total score output subunit to obtain the sensory evaluation total score; The target fish meat part, target constant temperature value, and target duration are input into the input unit, and then processed by the overall acceptability prediction subunit and the overall acceptability output subunit to obtain the overall acceptability of the sensory evaluation.

3. A device for predicting the freshness of fish meat, characterized in that, include: The input acquisition module is used to determine the target fish meat part, the target constant temperature value, and the target duration; The freshness index value output module is used to input the target fish meat part, the target constant temperature value and the target duration into the fish meat freshness prediction model to obtain the freshness index value output by the fish meat freshness prediction model. The fish freshness prediction model is trained using preset sample fish parts, constant sample temperature values, preset sample duration, and freshness sample index values. The fish freshness prediction model includes an input unit, a processing unit, and an output unit. The training method for the fish freshness prediction model includes: The sample fish meat part, constant sample temperature value, and preset sample duration are input into the input unit, and then processed by the processing unit and the output unit in sequence to output the freshness sample prediction value. The predicted value of the freshness sample is compared with the freshness sample index value, and the parameters of the input unit, processing unit and output unit are adjusted according to the comparison result until the difference between the predicted value of the freshness sample and the freshness sample index value is less than the threshold. The freshness sample index values ​​include at least the total viable bacteria count sample index value, the volatile basic nitrogen content sample index value, the freshness K value sample index value, the sensory evaluation total score sample index value, and the sensory evaluation overall acceptability index value. The processing unit includes: a sub-unit for predicting the total viable bacteria count of fish meat, a sub-unit for predicting the volatile basic nitrogen content of fish meat, a sub-unit for predicting the K-value of fish meat freshness, a sub-unit for predicting the total sensory evaluation score of fish meat, and a sub-unit for predicting the overall acceptability of fish meat sensory evaluation. The output unit includes: a sub-unit for outputting the total viable bacteria count of fish meat, a sub-unit for outputting the volatile basic nitrogen content of fish meat, a sub-unit for outputting the K-value of fish meat freshness, a sub-unit for outputting the total sensory evaluation score of fish meat, and a sub-unit for outputting the overall acceptability of fish meat sensory evaluation. The sample fish meat portion, constant sample temperature value, and preset sample duration are input into the input unit, and then processed sequentially by the processing unit and the output unit to output a freshness sample prediction value, including: The sample fish meat part, constant sample temperature value, and preset sample duration are input into the input unit, and then processed by the fish meat total viable bacteria count prediction subunit and the fish meat total viable bacteria count output subunit in sequence to output the total viable bacteria count sample prediction value. The sample fish meat part, constant sample temperature value, and preset sample duration are input into the input unit, and then processed sequentially by the fish meat volatile basic nitrogen content prediction subunit and the fish meat volatile basic nitrogen content output subunit to output the sample predicted value of fish meat volatile basic nitrogen content. The sample fish meat part, constant sample temperature value, and preset sample duration are input into the input unit, and then processed sequentially by the fish meat freshness K value prediction subunit and the fish meat freshness K value output subunit to output the fish meat freshness K value sample prediction value. The sample fish meat part, constant sample temperature value, and preset sample duration are input into the input unit, and then processed by the fish meat sensory evaluation total score prediction subunit and the fish meat sensory evaluation total score output subunit in sequence to output the fish meat sensory evaluation total score sample prediction value. The sample fish meat part, constant sample temperature value, and preset sample duration are input into the input unit, and then processed sequentially by the fish meat sensory evaluation overall acceptability prediction subunit and the fish meat sensory evaluation overall acceptability output subunit to output the fish meat sensory evaluation overall acceptability sample prediction value. The predicted value of the freshness sample is compared with the freshness sample index value, and the parameters of the input unit, processing unit, and output unit are adjusted according to the comparison result, including: The predicted value of the total viable bacteria count sample is compared with the index value of the total viable bacteria count sample, and the parameters of the input unit, the fish meat total viable bacteria count prediction subunit, and the fish meat total viable bacteria count output subunit are adjusted according to the comparison results. The predicted value of volatile basic nitrogen content sample is compared with the index value of volatile basic nitrogen content sample, and the parameters of the input unit, the fish meat volatile basic nitrogen content prediction subunit, and the fish meat volatile basic nitrogen content output subunit are adjusted according to the comparison results. The predicted value of the freshness K value sample is compared with the freshness K value sample index value, and the parameters of the input unit, the fish freshness K value prediction subunit, and the fish freshness K value output subunit are adjusted according to the comparison results. The predicted value of the total sensory evaluation score sample is compared with the index value of the total sensory evaluation score sample, and the parameters of the input unit, the fish meat sensory evaluation total score prediction subunit, and the fish meat sensory evaluation total score output subunit are adjusted according to the comparison results. The predicted value of the overall acceptability of sensory evaluation samples is compared with the index value of the overall acceptability of sensory evaluation samples, and the parameters of the input unit, the prediction subunit of overall acceptability of fish sensory evaluation, and the output subunit of overall acceptability of fish sensory evaluation are adjusted according to the comparison results.

4. An electronic device comprising a storage device, a processor, and a computer program stored in the storage device and executable on the processor, characterized in that, When the processor executes the program, it implements the fish freshness prediction method as described in any one of claims 1 to 2.

5. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the fish freshness prediction method as described in any one of claims 1 to 2.

6. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the fish freshness prediction method as described in any one of claims 1 to 2.

Images