Preparation method of colorimetric sensing array capable of intelligently detecting meat freshness

Abstract

The invention discloses a preparation method of a colorimetric sensing array capable of intelligently detecting meat freshness, and belongs to the technical field of meat freshness intelligent detection systems. The preparation method of the colorimetric sensor array comprises the following steps: dissolving carboxymethyl cellulose in water to prepare an aqueous solution, then adding anthocyanin, then dividing the solution into N equal parts, taking 1 part as a blank, and respectively adding the rest N-1 parts into corresponding ionic liquids to form anthocyanin solutions with different color gamuts; the preparation method comprises the following steps: respectively casting in a culture dish, drying, uncovering a film, carrying out hydrophobic coating on the film by using a hydrophobic coating, drying, cutting into wafers with the same scale by using a puncher, and fixing in a pore plate to obtain the colorimetric sensor array. The colorimetric sensing array has the advantages of portability, visualization, no damage, stability, biodegradability, simplicity in preparation and operation, reusability and the like, can monitor and predict the freshness of meat in real time, and can solve the problems and defects of a traditional detection method.

Description

Technical Field

[0001] The invention belongs to the technical field of meat freshness intelligent detection system, and specifically relates to a method for preparing a colorimetric sensor array capable of intelligently detecting meat freshness. Background Art

[0002] Animal-derived meat products play a vital role in the human diet and are an excellent source of high-quality protein for the human body. However, meat is extremely susceptible to spoilage due to microbial contamination during processing, transportation, and storage. In addition, lipid oxidation and protein degradation in meat can lead to the generation of volatile amines such as ammonia, dimethylamine, and trimethylamine. Traditional methods for assessing the freshness of meat include gas chromatography-mass spectrometry, liquid chromatography, and spectrophotometry. Although these methods are highly selective and sensitive, they have the limitations of expensive instruments and time-consuming sample preparation. Therefore, there is an urgent need to develop a simple and rapid method for detecting the freshness of meat.

[0003] Colorimetric sensor array is a detection technology that simulates the mammalian olfactory system and is composed of multiple sensor units. It can produce unique color cross-response signals for different analytes. It has the characteristics of portability, visualization, rapidity and non-destructiveness. It can be used to detect pesticide residues, animal residues, heavy metal ions and food freshness. The design of colorimetric sensor arrays usually focuses on the selection of responsive dyes and solid carriers. Existing sensor arrays often use toxic chemical synthetic dyes (porphyrins, metalloporphyrins and pH-sensitive synthetic dyes) as response dyes. Natural dyes have become a more promising response dye because of their non-toxicity and biodegradability. In recent years, natural dye anthocyanins have been favored by researchers because of their non-toxicity and wide color gamut. However, individual anthocyanins have a single color response under specific pH conditions and cannot form a cross-response of the array. Therefore, constructing an anthocyanin-based colorimetric sensor array is still a challenge currently faced.

[0004] The solid matrix is ​​the carrier and supporting platform of the dye in the colorimetric sensor array, ensuring the stability of the structure and function of the sensor element. Carboxymethyl cellulose is a water-soluble derivative of cellulose partially replaced by carboxymethyl groups. It is widely used in the fields of medicine, food, paper and packaging because of its non-toxicity, tastelessness, biodegradability and excellent film-forming ability. Therefore, carboxymethyl cellulose can be used as a solid matrix of the colorimetric sensor array to replace the traditional non-degradable solid matrix (C2 reversed-phase silica gel plate, polytetrafluoroethylene plate and quartz, etc.). However, fresh meat products contain a high amount of water, which may cause the dye in the carboxymethyl cellulose-based sensor array to migrate, thereby affecting its sensing accuracy and stability. In summary, in practical applications, colorimetric sensor arrays often require both green and safe dyes and matrices as well as excellent hydrophobic stability under high humidity conditions to achieve the purpose of real-time and accurate monitoring of the freshness of fresh meat.

[0005] Although anthocyanin / carboxymethyl cellulose-based colorimetric labels have been reported, a single colorimetric label has a single color when monitoring the freshness of fresh meat in real time, which can easily lead to misjudgment of the freshness of meat. In addition, the integrity of a single colorimetric label is easily destroyed under high humidity conditions, resulting in dye leakage, which ultimately affects its sensing accuracy and service life.

[0006] Therefore, it is of great application significance to develop a sensitive, stable, reusable and biodegradable colorimetric sensing array for real-time and accurate detection of the freshness of fresh meat. Summary of the invention

[0007] In view of the defects and shortcomings of the prior art, the present invention provides a method for preparing a colorimetric sensor array with multifunctional characteristics for accurately detecting the freshness of fresh meat in real time. The colorimetric sensor array prepared by the present invention has the advantages of portability, visualization, non-destructiveness, stability, biodegradability, simple preparation and operation, and reusability, and can monitor and predict the freshness of meat in real time, and can solve the problems and shortcomings of traditional detection methods and existing detection methods.

[0008] In order to achieve the above object, the present invention adopts the following technical solution:

[0009] The present invention provides a method for preparing a colorimetric sensor array capable of intelligently detecting the freshness of meat, the preparation method specifically comprising the following steps:

[0010] (1) dissolving carboxymethyl cellulose in water to prepare an aqueous solution, then adding anthocyanin, and then dividing the solution into N equal parts, 1 part as a blank, and adding corresponding ionic liquids to the remaining N-1 parts, respectively, to form anthocyanin solutions of different color ranges;

[0011] (2) The anthocyanin solutions of different color ranges in step (1) are respectively cast in culture dishes, and after drying, the membrane is peeled off, and then the membrane is hydrophobically coated with a hydrophobic coating. After drying, a puncher is used to cut out discs of the same size and fix them in a well plate, thereby preparing a colorimetric sensor array.

[0012] In one embodiment of the present invention, the mass fraction of carboxymethyl cellulose in the aqueous solution in step (1) is 1-6 wt %, preferably 1-2 wt %.

[0013] In one embodiment of the present invention, the mass fraction of anthocyanin in step (1) is 0.1-2 wt %, based on the mass of carboxymethyl cellulose; preferably 0.5-1 wt %.

[0014] In one embodiment of the present invention, in step (1), N is a natural number selected from 2 to 10, preferably 2 to 6; and more preferably 3 to 5.

[0015] In one embodiment of the present invention, the anthocyanidin in step (1) is any one of blueberry anthocyanidin, blackberry anthocyanidin, purple cabbage anthocyanidin, purple potato anthocyanidin, and grape anthocyanidin.

[0016] In one embodiment of the present invention, the ionic liquid in step (1) is one or more of 1-butyl-3-methylimidazolium bromide, 1-butyl-3-methylimidazolium chloride, 1-hexyl-3-methylimidazolium chloride and ionic liquid analogs (deep eutectic solvents).

[0017] In one embodiment of the present invention, the mass fraction of the ionic liquid in step (1) is 8-40 wt %, based on the mass of carboxymethyl cellulose in each solution; preferably 8-24 wt %.

[0018] In one embodiment of the present invention, in step (1), N is 3, and the corresponding ionic liquids are 1-butyl-3-methylimidazolium chloride and two ionic liquid analogs (analog 1: choline chloride and anhydrous glucose, with a molar mass ratio of 2:1; analog 2: choline chloride and 1,4-butanediol, with a molar mass ratio of 1:4).

[0019] In one embodiment of the present invention, the peeling off of the film after drying in step (2) refers to drying in an oven at a temperature of 30 to 60° C. for 6 to 16 hours.

[0020] In one embodiment of the present invention, the hydrophobic coating in step (2) is one or more of a fluorine-containing coating, a silicon-containing coating and a paraffin coating.

[0021] In one embodiment of the present invention, the fluorine-containing coating is one or two of fluorinated polyethylene and fluorocarbon wax; the silicon-containing coating is ethanol-dispersed nano-silicon dioxide.

[0022] In one embodiment of the present invention, the mass fraction of the hydrophobic coating in step (2) is 0.5-4wt% (based on the mass of the solvent); preferably 1-2wt%.

[0023] In one embodiment of the present invention, the drying condition in step (2) is 20-50° C. and the drying time is 10-40 min.

[0024] In one embodiment of the present invention, the hydrophobic coating in step (2) is specifically performed by immersing the membrane in a 2 wt % ethanol-dispersed nano-silica solution for 1 min, and then drying it in a 40° C. oven for 30 min.

[0025] The present invention provides a colorimetric sensor array prepared by the above-mentioned preparation method.

[0026] The third object of the present invention is to provide an application of the above-mentioned colorimetric sensing array in detecting food freshness.

[0027] The fourth object of the present invention is to provide a method for visually monitoring the freshness of meat products. The method is to put the above-mentioned colorimetric sensing array into fresh meat products and observe the freshness of the meat products according to the color change of the colorimetric sensing array.

[0028] The fifth object of the present invention is to provide a method for intelligently detecting the freshness of meat. The method comprises the following steps:

[0029] (1) Fix the above-mentioned colorimetric sensing array in a fresh-keeping box containing fresh meat and seal it with plastic wrap. The colorimetric sensing array will respond and change color with the generation of volatile amines during the meat spoilage process. Use a camera to take pictures and record the color change images of the colorimetric sensing array. At the same time, measure the content of total volatile basic nitrogen (TVB-N) in the meat and establish the relationship between the color of the colorimetric sensing array and the freshness of the meat.

[0030] (2) Use a deep learning algorithm to train and test the color change images of the colorimetric sensing array collected in step (1) and construct a meat freshness classification model.

[0031] (3) According to the meat freshness classification model constructed in step (2), it is used for the detection of meat freshness.

[0032] In an embodiment of the present invention, the taking pictures and recording the color change images of the colorimetric sensing array and the measurement of TVB-N value in step (1) are carried out every 2 hours; the freshness of the meat is classified according to the measured TVB-N value. When TVB-N < 15 mg / 100 g, it is classified as fresh; when 15 mg / 100 g < TVB-N < 20 mg / 100 g, it is sub-fresh; when TVB-N > 20 mg / 100 g, it is spoiled; according to the classification standard of TVB-N value, the corresponding colorimetric sensing array images are also classified into three categories: fresh, sub-fresh and spoiled.

[0033] In an embodiment of the present invention, the deep learning algorithm in step (2) is one or more of deep convolutional neural network (DCNN), recurrent neural network (RNN), and multi-layer perceptron (MLP).

[0034] In an embodiment of the present invention, the classification model in step (2) selects any one or more of VGG, ResNet, Mobilenetv, hufflenetv, and Inception.

[0035] In one embodiment of the present invention, the deep learning algorithm in step (2) is a deep convolutional neural network (DCNN) algorithm; the classification model is VGG16, ResNet18, Mobilenetv2 and Shufflenetv2.

[0036] Beneficial effects:

[0037] The colorimetric sensor array prepared by the present invention has the advantages of portability, visualization, non-destructiveness, stability, biodegradability, simple preparation and operation, and reusability. It can monitor and predict the freshness of meat in real time and can solve the problems and shortcomings of traditional detection methods and existing detection methods.

[0038] Compared with the prior art, the preparation method provided by the present invention has the following advantages:

[0039] (1) Compared with the existing spectral technology for detecting meat freshness, the colorimetric sensor array prepared by the present invention has the advantages of being portable, non-destructive, simple to operate, and visual;

[0040] (2) The colorimetric sensor array prepared by the present invention is biodegradable, reusable and has good hydrophobic stability in a high humidity environment, and has broad application prospects in the field of meat freshness detection;

[0041] (3) The colorimetric sensor array prepared by the present invention combined with a deep learning algorithm can realize real-time, precise and intelligent monitoring and prediction of the freshness of pork (with an accuracy of up to 99.16%). BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 Schematic diagram of the hydrophobic stability of the colorimetric sensor array prepared in Example 1 and Comparative Example 1 under high humidity conditions; (a) water contact angle diagram; (b) hydrophobic schematic diagram;

[0043] Figure 2 This is a diagram showing the responsiveness of the colorimetric sensor array prepared in Example 1 to five common amines produced by meat spoilage;

[0044] Figure 3 This is a rendering of the reusability of the colorimetric sensor array prepared in Example 1;

[0045] Figure 4 The colorimetric sensor array prepared in Example 1 is used to monitor the freshness of meat in real time; (a) the color change image of the colorimetric sensor array during the process of monitoring the freshness of meat; (b) the TVB-N value change image of meat;

[0046] Figure 5The colorimetric sensor array prepared in Example 1 combined with deep learning for meat freshness detection and classification data diagram; (a) model training schematic diagram; (b) training loss and training accuracy diagram of four models; (c) confusion matrix corresponding to the four models. DETAILED DESCRIPTION

[0047] The above contents of the present invention are further described in detail below through examples, but this should not be understood as the scope of the above subject matter of the present invention being limited to the following examples, and all technologies implemented based on the above contents of the present invention belong to the scope of the present invention.

[0048] The test method involved in the present invention is:

[0049] 1. Hydrophobic stability test of colorimetric sensor array under high humidity conditions

[0050] The contact angle of the array was measured using an optical contact angle meter (OCA15EC; Dataphysics, Germany). Each sample was measured at least three times and the average value was taken.

[0051] 2. Responsiveness test of colorimetric sensor array to five common amines produced by meat spoilage

[0052] In order to evaluate the effectiveness of the colorimetric sensor array in meat freshness detection, the responsiveness of the colorimetric sensor array to five common amine gases [ammonia (NH3), dimethylamine (DMA), trimethylamine (TMA), putrescine (PUT), and cadaverine (CAD)] produced by meat spoilage was tested. The colorimetric sensor array was placed in a sealed bottle, and a certain volume of amine solution was injected into the sealed bottle to achieve the required gas concentration. The amine gas concentration (C ppm ) is calculated as follows:

[0053]

[0054] Where V μL is the volume of the analytical solution, D mg / L is the liquid density, W is the liquid mass fraction, M g / mol is the molecular weight of the analytical solution, V L The volume of the sealed bottle.

[0055] 3. Determination of the reusability of the colorimetric sensor array

[0056] The colorimetric sensor array was exposed to 5 amine gases at a concentration of 20% for 10 minutes. Then, the colorimetric sensor array was exposed to 10% acetic acid for 10 minutes, and the images of the colorimetric sensor array during these two processes were recorded using a camera. In addition, the number of cycles that the colorimetric sensor array can be reused under ammonia conditions was also determined.

[0057] 4. Colorimetric sensor array for real-time monitoring of meat freshness

[0058] (1) Color response

[0059] 200 g of fresh pork and the colorimetric sensor array were placed in a sealed fresh-keeping box and then stored in an incubator at 25° C. for 28 h. The color changes of the colorimetric sensor array were recorded with a camera every 4 h.

[0060] (2) Determination of total volatile basic nitrogen (TVB-N)

[0061] Without interfering with the colorimetric sensor array image acquisition, pork TVB-N was determined according to the national standard method (GB5009.228-2016).

[0062] Example 1

[0063] A method for preparing a colorimetric sensor array capable of intelligently detecting the freshness of meat, the method specifically comprising the following steps:

[0064] (1) Accurately weigh 2 g of carboxymethyl cellulose and dissolve it in 100 ml of deionized water. Stir in a magnetic stirrer at 60°C until the carboxymethyl cellulose is completely dissolved. After cooling to room temperature, add 1 wt% of blueberry anthocyanin (based on the mass of carboxymethyl cellulose). Then divide the above solution into 4 equal portions (20 ml each), one portion as a blank, and add 24 wt% (based on the mass of carboxymethyl cellulose in each solution) of three different ionic liquids (1-butyl-3-methylimidazolium chloride, two ionic liquid analogs (analog 1: choline chloride and anhydrous glucose, molar mass ratio of 2:1; analog 2: choline chloride and 1,4-butanediol, molar mass ratio of 1:4)) to the remaining 3 portions, forming anthocyanin solutions with three different color ranges;

[0065] (2) The above four solutions were cast into a culture dish with a diameter of 60 mm, and placed in an oven at 40°C for drying for 12 h, then taken out and peeled off, and the four films were hydrophobically coated with nano-silica dispersed in 2 wt% ethanol. The specific process was to immerse the four films in the nano-silica solution dispersed in 2 wt% ethanol for 1 min, and then place them in an oven at 40°C for drying for 30 min; finally, a puncher with an aperture of 7 mm was used to cut the four dried films into 7 mm discs and fixed them in a polytetrafluoroethylene plate with an aperture of 7 mm to prepare a colorimetric sensor array.

[0066] Example 2

[0067] A method for preparing a colorimetric sensor array capable of intelligently detecting the freshness of meat, the method specifically comprising the following steps:

[0068] (1) Accurately weigh 2 g of carboxymethyl cellulose and dissolve it in 100 ml of deionized water. Stir in a magnetic stirrer at 60°C until the carboxymethyl cellulose is completely dissolved. After cooling to room temperature, add 1 wt% of blueberry anthocyanin (based on the mass of carboxymethyl cellulose). Then divide the above solution into 4 equal portions (20 ml each), one portion as a blank, and add 8 wt% (based on the mass of carboxymethyl cellulose in each solution) of three different ionic liquids (1-butyl-3-methylimidazolium chloride, two ionic liquid analogs (analog 1: choline chloride and anhydrous glucose, molar mass ratio of 2:1; analog 2: choline chloride and 1,4-butanediol, molar mass ratio of 1:4)) to the remaining 3 portions, forming three anthocyanin solutions with different color ranges;

[0069] (2) The above four solutions were cast into a culture dish with a diameter of 60 mm, and placed in an oven at 40°C for drying for 12 h, then taken out and peeled off, and the four films were hydrophobically coated with nano-silica dispersed in 2wt% ethanol. The specific process was to immerse the four films in the nano-silica solution dispersed in 2wt% ethanol for 1 min, and then place them in an oven at 40°C for drying for 30 min; finally, a puncher with an aperture of 7 mm was used to cut the above four films into 7 mm discs and fixed them in a polytetrafluoroethylene plate with an aperture of 7 mm to prepare a colorimetric sensor array.

[0070] Example 3

[0071] A method for preparing a colorimetric sensor array capable of intelligently detecting the freshness of meat, the method specifically comprising the following steps:

[0072] (1) Accurately weigh 2g of carboxymethyl cellulose and dissolve it in 100ml of deionized water. Stir in a magnetic stirrer at 60°C until the carboxymethyl cellulose is completely dissolved. After the above solution is cooled to room temperature, 1w%t of blueberry anthocyanin (based on the mass of carboxymethyl cellulose) is added, and then the above solution is divided into 4 equal parts (20ml each), 1 part is used as a blank, and the remaining 3 parts are added with 16wt% (based on the mass of carboxymethyl cellulose in each solution) of 3 different ionic liquids (1-butyl-3-methylimidazolium chloride, two ionic liquid analogs (analog 1: choline chloride and anhydrous glucose, molar mass ratio of 2:1; analog 2: choline chloride and 1,4-butanediol, molar mass ratio of 1:4)) to form 3 anthocyanin solutions with different color ranges;

[0073] (2) The above four solutions were cast into a culture dish with a diameter of 60 mm, and placed in an oven at 40°C for drying for 12 h, then taken out and peeled off, and the four films were hydrophobically coated with nano-silica dispersed in 2wt% ethanol. The specific process was to immerse the four films in the nano-silica solution dispersed in 2wt% ethanol for 1 min, and then place them in an oven at 40°C for drying for 30 min; finally, a puncher with an aperture of 7 mm was used to cut the above four films into 7 mm discs and fixed them in a polytetrafluoroethylene plate with an aperture of 7 mm to prepare a colorimetric sensor array.

[0074] Example 4

[0075] A method for preparing a colorimetric sensor array capable of intelligently detecting the freshness of meat, the method specifically comprising the following steps:

[0076] (1) Accurately weigh 2 g of carboxymethyl cellulose and dissolve it in 100 ml of deionized water. Stir in a magnetic stirrer at 60°C until the carboxymethyl cellulose is completely dissolved. After cooling to room temperature, add 1 wt% t of blueberry anthocyanin (based on the mass of carboxymethyl cellulose). Then divide the above solution into 4 equal portions (20 ml each), one portion as a blank, and add 32 wt% (based on the mass of carboxymethyl cellulose in each solution) of three different ionic liquids (1-butyl-3-methylimidazolium chloride, two ionic liquid analogs (analog 1: choline chloride and anhydrous glucose, molar mass ratio of 2:1; analog 2: choline chloride and 1,4-butanediol, molar mass ratio of 1:4)) to the remaining three portions, forming three anthocyanin solutions with different color ranges;

[0077] (2) The above four solutions were cast into a culture dish with a diameter of 60 mm, and placed in an oven at 40°C for drying for 12 h, then taken out and peeled off, and the four films were hydrophobically coated with nano-silica dispersed in 2wt% ethanol. The specific process was to immerse the four films in the nano-silica solution dispersed in 2wt% ethanol for 1 min, and then place them in an oven at 40°C for drying for 30 min; finally, a puncher with an aperture of 7 mm was used to cut the above four films into 7 mm discs and fixed them in a polytetrafluoroethylene plate with an aperture of 7 mm to prepare a colorimetric sensor array.

[0078] Comparative Example 1

[0079] A method for preparing a colorimetric sensor array capable of intelligently detecting the freshness of meat, the method specifically comprising the following steps:

[0080] (1) Accurately weigh 2 g of carboxymethyl cellulose and dissolve it in 100 ml of deionized water. Stir in a magnetic stirrer at 60°C until the carboxymethyl cellulose is completely dissolved. After cooling to room temperature, add 1 wt% of blueberry anthocyanin (based on the mass of carboxymethyl cellulose). Then divide the above solution into 4 equal portions (20 ml each), one portion as a blank, and add 24 wt% (based on the mass of carboxymethyl cellulose in each solution) of three different ionic liquids (1-butyl-3-methylimidazolium chloride, two ionic liquid analogs (analog 1: choline chloride and anhydrous glucose, molar mass ratio of 2:1; analog 2: choline chloride and 1,4-butanediol, molar mass ratio of 1:4)) to the remaining 3 portions, forming anthocyanin solutions with three different color ranges;

[0081] (2) The above four solutions were cast into a culture dish with a diameter of 60 mm, and placed in an oven at 40°C for drying for 12 h before being taken out and peeled off; then, a hole puncher with a pore size of 7 mm was used to cut the above four films into 7 mm discs and fixed them in a polytetrafluoroethylene plate with a pore size of 7 mm to prepare a colorimetric sensor array.

[0082] Comparative Example 2

[0083] The difference from Example 1 is that the three ionic liquids in step (1) are 1-butyl-3-methylimidazolium hydrogen sulfate, 1-butyl-3-methylimidazolium trifluoromethanesulfonate, and an ionic liquid analog (choline chloride and propylene glycol, with a molar mass ratio of 1:2); the results show that the addition of these three ionic liquids has no regulatory effect on the color gamut of anthocyanins, and therefore, it is impossible to construct a colorimetric sensing array with cross-response.

[0084] Comparative Example 3

[0085] The only difference from Example 1 is that the three ionic liquids in step (1) are 1-butyl-3-methylimidazolium acetate and two ionic liquid analogs (analog 1: choline chloride and lactic acid, with a molar mass ratio of 1:1, and analog 2: proline and propylene glycol, with a molar mass ratio of 1:2). The results showed that the addition of these three ionic liquids also had no regulatory effect on the color gamut of anthocyanins, and therefore, it was impossible to construct a colorimetric sensing array with cross-response.

[0086] Results Analysis

[0087] 1. Hydrophobic properties of two colorimetric sensor arrays prepared in Example 1 and Comparative Example 1

[0088] Depend on Figure 1It can be seen that the colorimetric sensor array prepared in Example 1 has excellent hydrophobic properties, and the contact angle is about 130°, which is significantly better than that of Comparative Example 1. The excellent hydrophobic properties ensure the integrity and sensitivity of the colorimetric sensor array under high humidity conditions. Similarly, the colorimetric sensor arrays prepared in Examples 2-4 also have excellent hydrophobic properties.

[0089] 2. Responsiveness of the colorimetric sensor array to five common amines produced by meat spoilage

[0090] Depend on Figure 2 It can be seen that the colorimetric sensor array prepared in Example 1 has good responsiveness to the five amines. Therefore, the prepared colorimetric sensor array has the ability to detect the freshness of pork. The colorimetric sensor arrays prepared in Examples 2-4 all have good responsiveness to the five amines.

[0091] 3. Reusability of colorimetric sensing arrays

[0092] Depend on Figure 3 It can be seen that the colorimetric sensor array prepared in Example 1 changes color significantly when exposed to different amine gases. After the colorimetric sensor array with color change is exposed to 10% acetic acid, the color of the colorimetric sensor array returns to the original color, indicating that the colorimetric sensor array has good color reversibility and can be reused. And under the condition of alternating ammonia and acetic acid, the colorimetric sensor array can be reused 11 times. The colorimetric sensor arrays prepared in Examples 2-4 all have good reusability.

[0093] 4. Colorimetric sensor array for real-time monitoring of meat freshness

[0094] Depend on Figure 4 It can be seen that when the TVB-N value changes from 6.16 mg / 100 g at 0 h to 14.85 mg / 100 g at 16 h, and then to 25.08 mg / 100 g at 28 h, the color of the colorimetric sensor array prepared in Example 1 also changes accordingly. The results show that the color change of the colorimetric sensor array can reflect the freshness of meat to different degrees, that is, it can realize visual monitoring of the freshness of meat. The colorimetric sensor arrays prepared in Examples 2-4 can all realize visual monitoring of the freshness of meat.

[0095] Example 5

[0096] The colorimetric sensor array prepared in Example 1 was combined with deep learning for meat freshness detection and classification

[0097] Test method:

[0098] (1) Fix the colorimetric sensing array prepared in Example 1 in a fresh-keeping box containing fresh meat and seal it with plastic wrap. The colorimetric sensing array will respond and change color with the generation of volatile amines during the meat spoilage process. During the meat spoilage process, use a camera to record the color change images of the colorimetric sensing array. At the same time, measure the content of total volatile basic nitrogen (TVB-N) in the meat and establish the relationship between the color of the colorimetric sensing array and the freshness of the meat.

[0099] The shooting and recording of the color change images of the colorimetric sensing array and the measurement of TVB-N values are carried out every 2 h. Classify the freshness of the meat according to the measured TVB-N values. When TVB-N < 15 mg / 100 g, it is classified as fresh; when 15 mg / 100 g < TVB-N < 20 mg / 100 g, it is sub-fresh; when TVB-N > 20 mg / 100 g, it is spoiled. According to the classification criteria based on TVB-N values, the corresponding colorimetric sensing array images are also classified into three categories: fresh, sub-fresh, and spoiled.

[0100] (2) Use a deep learning algorithm to train and test the color features of the images in step (1) and construct a meat freshness classification model.

[0101] Among them, the deep learning algorithm selects the deep convolutional neural network (DCNN) algorithm; the classification models select VGG16, ResNet18, Mobilenetv2, and Shufflenetv2.

[0102] Four different DCNN freshness classification models (Mobilenetv2, Shufflenetv2, ResNet18, and VGG16) are established. These DCNN models classify the freshness of the meat by extracting the color features of the colorimetric sensing array images ( Figure 5 a).

[0103] As Figure 5 can be seen, the training processes of the four DCNN models are as Figure 5 shown in b. As the number of epochs increases, the training accuracy of the model improves and the training loss decreases, which means that all four models can accurately classify the freshness of pork. Among them, the VGG16 model has the highest classification accuracy (99.16%), and the classification accuracies of Mobilenetv2, Shufflenetv2, and ResNet18 are 98.67%, 97.96%, and 98.43% respectively.

[0104] To further compare the classification accuracies of these DCNN models, confusion matrices corresponding to the 4 DCNN models are created ( Figure 5c). The diagonal of the confusion matrix represents the number of correctly classified samples, and the other areas represent the number of incorrectly classified samples. The four models are basically conservative in freshness classification, which shows the effectiveness of the four models in predicting meat freshness. Therefore, the colorimetric sensor array prepared in Example 1 combined with the DCNN model can be used for accurate detection and classification of meat freshness. This detection method is fast, accurate, and non-destructive, and has great potential in monitoring pork freshness in intelligent packaging. Similarly, the colorimetric sensor arrays prepared in Examples 2-4 can be combined with deep learning for meat freshness detection and classification.

[0105] The present invention provides a method for preparing a colorimetric sensor array with multifunctional characteristics for real-time and accurate detection of the freshness of fresh meat, belonging to the technical field of intelligent detection system for meat freshness. Experimental results show that the comprehensive performance of the prepared colorimetric sensor array is better than that of similar products. At the same time, the present invention can realize real-time, accurate and intelligent monitoring and prediction of pork freshness (with an accuracy of up to 99.16%) in combination with a deep learning algorithm.

[0106] The embodiments provided above are not intended to limit the scope of the present invention, and the steps described are not intended to limit the execution order thereof. Those skilled in the art may make obvious improvements to the present invention in combination with existing common knowledge, which also fall within the scope of protection defined by the claims of the present invention.

Claims

1. A method for preparing a colorimetric sensor array capable of intelligently detecting the freshness of meat, characterized in that: The preparation method includes the following steps: (1) Dissolve carboxymethyl cellulose in water to prepare an aqueous solution, then add anthocyanin, and then divide the solution into N equal parts. One part is used as a blank, and the remaining N - 1 parts are respectively added with corresponding ionic liquids to form anthocyanin solutions with different color gamuts; The N is a natural number, selected from 2 to 10; The ionic liquid is one or more of 1-butyl-3-methylimidazolium bromide, 1-butyl-3-methylimidazolium chloride, 1-hexyl-3-methylimidazolium chloride and ionic liquid analogs; (2) Cast the anthocyanin solutions with different color gamuts in step (1) into petri dishes respectively. After drying, remove the film. Then, use a hydrophobic coating to coat the film hydrophobically. After drying, use a punch to cut out round pieces with the same size and fix them in a well plate to obtain a colorimetric sensing array.

2. The preparation method according to claim 1, characterized in that: In the aqueous solution in step (1), the mass fraction of carboxymethyl cellulose is 1 - 6 wt%.

3. The preparation method according to claim 1, characterized in that: In step (1), the mass fraction of anthocyanin is 0.1 - 2 wt%, based on the mass of carboxymethyl cellulose.

4. The preparation method according to claim 1, characterized in that: In step (1), the mass fraction of the ionic liquid is 8 - 40 wt%, based on the mass of carboxymethyl cellulose in each portion of the solution.

5. A colorimetric sensing array prepared by the preparation method according to any one of claims 1 to 4.

6. An application of the colorimetric sensing array according to claim 5 in detecting the freshness of food.

7. A method for visually monitoring the freshness of meat products, characterized in that: The method is to put the colorimetric sensing array according to claim 5 into fresh meat products, and observe the freshness of the meat products according to the color change of the colorimetric sensing array.

8. A method for intelligently detecting the freshness of meat, characterized in that: The method includes the following steps: (1) Fix the above-mentioned colorimetric sensing array in a fresh-keeping box containing fresh meat and seal it with a fresh-keeping film. The colorimetric sensing array will respond and change color with the generation of volatile amines during the meat spoilage process. Use a camera to take pictures and record the color change images of the colorimetric sensing array. At the same time, measure the content of volatile basic nitrogen in the meat and establish the relationship between the color of the colorimetric sensing array and the freshness of the meat; (2) Use a deep learning algorithm to train and test the color change images of the colorimetric sensing array collected in step (1) to construct a meat freshness classification model; (3) According to the meat freshness classification model constructed in step (2), it is used for the detection of meat freshness.

9. The method according to claim 8, characterized in that In step (1), the taking pictures and recording the color change images of the colorimetric sensing array and the measurement of TVB-N value are carried out every 2 h; classify the freshness of the meat according to the measured TVB-N value. When TVB-N < 15 mg / 100 g, it is classified as fresh; when 15 mg / 100 g < TVB-N < 20 mg / 100 g, it is sub-fresh; when TVB-N > 20 mg / 100 g, it is spoiled; according to the classification standard of TVB-N value, the corresponding colorimetric sensing array images are also classified into three categories: fresh, sub-fresh and spoiled.

10. The method according to claim 8, characterized in that In step (2), the deep learning algorithm selects the deep convolutional neural network (DCNN) algorithm; the classification models are selected from VGG16, ResNet18, Mobilenetv2 and Shufflenetv2.

Images