Method for preparing vegetable freshness indicator based on nano material and application of indicator

Abstract

The invention relates to a method for preparing a vegetable freshness indicator based on a nano material. The method is simple in preparation process and low in cost. A vegetable freshness indicator label prepared after the vegetable freshness indicator prepared through the method forms a film overcomes the defects that an existing indicator label base material is poor in performance, low in response sensitivity, low in recognition precision, complex in preparation technology and the like, has excellent mechanical performance and heat stability and can accurately capture pH value changes. The invention also provides a method for judging the freshness of the vegetables, which comprises the following steps: preserving the prepared vegetable freshness indicator and / or the vegetable freshness indication label together with the vegetables, and judging the freshness of the vegetables on the basis of the pictures of the vegetable freshness indicator and / or the vegetable freshness indication label. And the vegetable freshness can be judged in combination with the trained machine learning model.

Description

Technical Field

[0001] This application relates to the field of intelligent food packaging and quality monitoring technology, and in particular to a method for preparing vegetable freshness indicators based on nanomaterials and its application. Background Technology

[0002] Vegetables play an important role in our daily diet, providing essential nutrients such as vitamins, dietary fiber, and minerals, as well as offering multiple health benefits. However, the nutritional value of fresh vegetables is not constant. The shelf life and rate of spoilage of vegetables are affected by their inherent characteristics and storage methods. After harvesting, the continuous production of acidic gases such as CO2 during metabolism leads to a gradual decline in freshness and eventual spoilage.

[0003] Existing technologies disclose a smart packaging hydrogel material with CO2-responsive properties to fruit and vegetable metabolic gas products and its applications. This material is prepared using photonic crystal template assembly and amino modification processes, but the preparation process is complex, requires large amounts of water, and is costly, hindering industrialization. Another existing technology discloses a color-sensitive indicator label for detecting fish freshness and its preparation method. This label is made by chemically grafting polylactic acid and a citrate derivative. The resulting polylactic acid-citrate derivative is used as a spinning solution for electrospinning to prepare nanofibers, which can be used for amine-based methods in fish preservation. The detection of substances utilizes electrospinning technology to prepare nanofiber membranes, which requires high preparation conditions and is suitable for detecting the freshness of fish. Existing technologies also disclose methods for detecting the freshness of fruits and vegetables, which use the RGB values ​​of indicator labels combined with machine learning models to determine the freshness of fruits and vegetables. However, the raw materials used in the indicator labels in this method are mostly conventional and ordinary materials, which are slow to respond to changes during food preservation and have low accuracy. This makes the changes in the RGB values ​​of the indicator labels inaccurate, and the accuracy of the changes in RGB values ​​directly affects the model's judgment results, leading to inaccurate judgment results.

[0004] Accurate detection of vegetable freshness and timely implementation of preventative and control measures are crucial for ensuring food safety and minimizing economic losses. Therefore, the rational use of smart food packaging is particularly important. However, existing smart packaging and labels have limited detection sensitivity, making it difficult to meet the stringent requirements of certain food preservation tests and real-time monitoring.

[0005] Therefore, there is an urgent need to provide a vegetable freshness indicator label that is fast-responding, highly sensitive, and highly accurate in monitoring. Summary of the Invention

[0006] The purpose of this invention is to overcome the above-mentioned shortcomings of the prior art and provide a method for preparing vegetable freshness indicators based on nanomaterials and its application.

[0007] The first objective of this invention is to provide a method for preparing vegetable freshness indicators based on nanomaterials.

[0008] The second objective of this invention is to provide a vegetable freshness indicator prepared by the above method.

[0009] A third objective of this invention is to provide a vegetable freshness indicator label based on nanomaterials.

[0010] A fourth objective of this invention is to provide the application of the above-mentioned vegetable freshness indicator and / or the above-mentioned vegetable freshness indicator label in determining the freshness of vegetables.

[0011] The fifth objective of this invention is to provide a method for determining the freshness of vegetables.

[0012] To achieve the above objectives, the present invention is implemented through the following solution: This invention claims protection for a method for preparing a vegetable freshness indicator based on nanomaterials, comprising the following steps: S1. Prepare nano-chitosan solution and nano-starch solution respectively; S2. Mix the nano starch solution and nano chitosan solution obtained in step S1 at a mass ratio of 6-7:4-3 until homogeneous. Then add glycerol with a final mass concentration of 2.0%-3.0% and nano silica with a mass concentration of 0.5%-1.0%. After mixing thoroughly, add an acid-base indicator and sonicate to obtain a vegetable freshness indicator.

[0013] Preferably, the preparation method of the nano-chitosan solution in step S1 is as follows: chitosan and acetic acid solution are mixed evenly and then Tween 80 is added. Then, the mixture is homogenized at 30 MPa for 5 to 8 times, at 50 MPa for 5 to 8 times, and at 80 MPa for 5 to 8 times. After homogenization, ultrasonic treatment is performed to obtain the nano-chitosan solution. The volume concentration of the acetic acid solution is 1.0–2.0%, and the homogenization time is 5–20 min each time.

[0014] Preferably, the homogenization is performed sequentially at 30 MPa for 5 times, at 50 MPa for 5 times, and at 80 MPa for 8 times.

[0015] More preferably, the homogenization is performed using a homogenizer (model: X-flow laboratory homogenizer 2000, manufacturer: Niru GEA Group).

[0016] Preferably, the homogenization time is 10 minutes each time.

[0017] Preferably, the ultrasonic treatment is performed by ultrasonication at a frequency of 40 kHz and a power of 300-400 W for 20-40 minutes.

[0018] More preferably, ultrasound is performed at a frequency of 40 kHz and a power of 400 W for 30 minutes.

[0019] Preferably, the chitosan and acetic acid solution are mixed evenly at a ratio of 1.0-2.0 g: 100 mL.

[0020] More preferably, the chitosan and acetic acid solution are mixed evenly at a ratio of 1.0 g: 100 mL.

[0021] More preferably, the volume concentration of the acetic acid solution is 1%.

[0022] More preferably, the final volume concentration of Tween 80 is 0.5%.

[0023] Preferably, the preparation method of the nano starch solution in step S1 is as follows: starch and water are mixed evenly and then Tween 80 is added. Then, the mixture is homogenized at 30 MPa for 5 to 8 times, at 60 MPa for 5 to 8 times, and at 120 MPa for 10 to 15 times. After homogenization, ultrasonic treatment is performed to obtain the nano starch solution. The homogenization time for each time is 5 to 20 minutes.

[0024] Preferably, the homogenization is performed sequentially at 30 MPa for 5 times, at 60 MPa for 5 times, and at 120 MPa for 10 cycles.

[0025] Preferably, the homogenization time is 10 minutes each time.

[0026] More preferably, the homogenization is performed using a homogenizer (model: X-flow laboratory homogenizer 2000, manufacturer: Niru GEA Group).

[0027] Preferably, the ultrasonic treatment is performed by ultrasonication at a frequency of 40 kHz and a power of 350-400 W for 20-40 minutes.

[0028] More preferably, the ultrasound is performed at a frequency of 40 kHz and a power of 350 W for 30 minutes.

[0029] Preferably, the starch and water are mixed evenly at a ratio of 4.0-5.0g:100mL.

[0030] More preferably, the starch and water are mixed evenly at a ratio of 5.0g:100mL.

[0031] Preferably, the final volume concentration of Tween 80 is 0.5%.

[0032] Preferably, in step S2, the nano starch solution and the nano chitosan solution are mixed evenly at a mass ratio of 3:2.

[0033] Preferably, step S2 involves adding glycerol with a final mass concentration of 2.0% and nano-silica of 0.5%.

[0034] Preferably, the thorough mixing in step S2 is achieved by magnetic stirring in a 60°C water bath for 2 hours.

[0035] More preferably, the magnetic stirring is performed at 800 r / min.

[0036] Preferably, the acid-base indicator in step S2 is m-cresol purple.

[0037] Preferably, before adding the acid-base indicator in step S2, the pH value of the thoroughly mixed liquid is adjusted to 11.

[0038] Preferably, the ultrasonic treatment in step S2 is: ultrasonic treatment at a frequency of 40kHz and a power of 350-400W for 30-50 minutes.

[0039] More preferably, the ultrasound is performed at a frequency of 40 kHz and a power of 350 W for 30 minutes.

[0040] The present invention also claims protection for vegetable freshness indicators prepared by any of the methods described above.

[0041] The present invention also claims protection for a vegetable freshness indicator label based on nanomaterials, wherein the vegetable freshness indicator label is prepared by forming the above-mentioned vegetable freshness indicator into a film.

[0042] Preferably, the film formation is achieved by pouring the above-mentioned vegetable freshness indicator into a mold and then placing it in a 50°C drying oven for thorough drying.

[0043] The method for preparing vegetable freshness indicators based on nanomaterials is simple and low-cost. The vegetable freshness indicators prepared by this method and the vegetable freshness indicator labels obtained by forming films with the vegetable freshness indicators overcome the defects of existing vegetable freshness indicator labels, such as poor substrate performance, low response sensitivity, low recognition accuracy, and complex processes. At the same time, it can also be combined with machine learning models to realize intelligent detection of vegetable freshness.

[0044] This invention also claims protection for the use of the above-mentioned vegetable freshness indicator and / or the above-mentioned vegetable freshness indicator label in determining vegetable freshness.

[0045] Preferably, the vegetable is broccoli.

[0046] Preferably, the freshness includes fresh, slightly fresh, and spoiled.

[0047] This invention also claims protection for a method for determining the freshness of vegetables, comprising the following steps: S11. Store the above-mentioned vegetable freshness indicator and / or the above-mentioned vegetable freshness indicator label in the same environment as the vegetables, and take photos of the vegetable freshness indicator and / or vegetable freshness indicator label during the storage process to obtain photos of the vegetable freshness indicator and / or vegetable freshness indicator label; S12. Input the photos of the vegetable freshness indicator and / or vegetable freshness indicator label obtained in step S11 into the trained machine learning model, and the trained machine learning model outputs the freshness of the vegetables; The trained machine learning model is a machine learning model trained using photos of vegetables with known freshness and their corresponding vegetable freshness indicators and / or vegetable freshness indicator labels. The freshness of the vegetables includes fresh, slightly fresh, and rotten.

[0048] Preferably, the photo taking in step S11 is based on a D65 standard light source.

[0049] Preferably, placing the food in the same environment in step S11 means placing it in the same transparent food preservation box.

[0050] Preferably, the trained machine learning model in step S12 is a trained deep convolutional neural network model.

[0051] Compared with the prior art, the present invention has the following beneficial effects: This invention provides a method for preparing vegetable freshness indicators based on nanomaterials. This method is simple and low-cost. Furthermore, the vegetable freshness indicator film prepared by this method, when used to create a vegetable freshness indicator label, overcomes the shortcomings of existing indicator labels, such as poor substrate performance, low response sensitivity, low recognition accuracy, and complex preparation processes. It exhibits excellent mechanical properties and thermal stability and can accurately capture pH changes. This invention also provides a method for determining vegetable freshness. By storing the prepared vegetable freshness indicator and / or vegetable freshness indicator label together with the vegetables, and using photographs of the vegetable freshness indicator and / or vegetable freshness indicator label, combined with a trained machine learning model, the freshness of the vegetables can be determined. Attached Figure Description

[0052] Figure 1 The results of particle size determination for each nano-chitosan solution and each nano-starch solution in Test Example 1 are shown in the figure. Figure 2 The mechanical strength results of vegetable freshness indicator labels 1 to 6 in test example 1 are shown in the figure. Figure 3 The graph shows the weight retention rate of vegetable freshness indicator labels 1 to 6 in test example 1. Figure 4 This is a photographic observation of the broccoli stored using various vegetable freshness indicator labels in Test Example 2. Figure 5 The following diagrams show the principal component analysis (PCA) results for the odor of broccoli stored with each vegetable freshness indicator label in Test Example 2: A) shows the PCA results when combined with vegetable freshness indicator label 1; B) shows the PCA results when combined with vegetable freshness indicator label 2; C) shows the PCA results when combined with vegetable freshness indicator label 3; D) shows the PCA results when combined with vegetable freshness indicator label 4; E) shows the PCA results when combined with vegetable freshness indicator label 5; and F) shows the PCA results when combined with vegetable freshness indicator label 6. Figure 6 The following graphs show the training accuracy results of the deep convolutional neural network models combining various vegetable freshness indicator labels in Test Example 2: A shows the training accuracy result of the deep convolutional neural network model combining vegetable freshness indicator label 1; B shows the training accuracy result of the deep convolutional neural network model combining vegetable freshness indicator label 2; C shows the training accuracy result of the deep convolutional neural network model combining vegetable freshness indicator label 3; D shows the training accuracy result of the deep convolutional neural network model combining vegetable freshness indicator label 4; E shows the training accuracy result of the deep convolutional neural network model combining vegetable freshness indicator label 5; and F shows the training accuracy result of the deep convolutional neural network model combining vegetable freshness indicator label 6. Figure 7 A represents the confusion matrix of the deep convolutional neural network model combining each vegetable freshness indicator label in Test Example 2; B represents the confusion matrix of the deep convolutional neural network model combining vegetable freshness indicator label 1; C represents the confusion matrix of the deep convolutional neural network model combining vegetable freshness indicator label 2; D represents the confusion matrix of the deep convolutional neural network model combining vegetable freshness indicator label 3; E represents the confusion matrix of the deep convolutional neural network model combining vegetable freshness indicator label 4; E represents the confusion matrix of the deep convolutional neural network model combining vegetable freshness indicator label 5; and F represents the confusion matrix of the deep convolutional neural network model combining vegetable freshness indicator label 6. Detailed Implementation

[0053] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. These embodiments are for illustrative purposes only and are not intended to limit the scope of the invention. Unless otherwise specified, the experimental methods used in the following embodiments are conventional methods; the materials and reagents used, unless otherwise specified, are commercially available.

[0054] In the embodiments and comparative examples of this invention, corn starch, chitosan, nano-silica, Tween 80, glycerol, acetic acid, anhydrous ethanol and m-cresol purple are all analytical grade materials.

[0055] The high-pressure homogenizer used in this embodiment of the invention is an X-ray laboratory homogenizer 2000, manufactured by Niru GEA Group.

[0056] Example 1: A vegetable freshness indicator label based on nanomaterials A vegetable freshness indicator label based on nanomaterials is prepared as follows: S1. Weigh 1.0 g of chitosan and add it to 100 mL of acetic acid solution (acetic acid volume concentration is 1.0%). Place it at 25℃ and stir magnetically at 800 r / min for 30 min until the solution is transparent and free of particles. Then add Tween 80 with a final volume concentration of 0.5%. After mixing evenly, use a high-pressure homogenizer to homogenize 5 times at 30 MPa pressure, then 5 times at 50 MPa pressure, and then 5 times at 80 MPa pressure. After homogenization, sonicate at 40 kHz frequency and 300 W power for 30 min to obtain nano-chitosan solution 1. Each homogenization time is 10 min. Weigh 5.0g of corn starch and add it to 100mL of deionized water. Stir magnetically at 800r / min for 30min to obtain a starch suspension. Place the starch suspension in a 90℃ water bath and heat with magnetic stirring at 800r / min for 30min. Then cool it to below 25℃ in an ice water bath. Add Tween 80 with a final volume concentration of 0.5%. Mix well and use a high-pressure homogenizer to homogenize 5 times at 30MPa, then 5 times at 60MPa, and then 10 times at 120MPa. After homogenization, sonicate at 40kHz and 350W for 30min to obtain nano starch solution 1. Each homogenization time is 10min.

[0057] S2. Mix the nano starch solution 1 and nano chitosan solution 1 prepared in step S1 at a mass ratio of 6:4 until homogeneous. Then add glycerol with a final mass concentration of 2.0% and nano silica with a mass concentration of 0.5%. Then stir magnetically at 800 r / min for 2 h in a 60°C water bath. Then cool the temperature to below 25°C in an ice water bath to obtain the solution to be used 1. Adjust the pH of the prepared solution 1 to pH=11. Then, weigh 0.005% of the mass of prepared solution 1 and dissolve it in 1 mL of anhydrous ethanol to obtain m-cresol purple solution 1. Add m-cresol purple solution 1 to prepared solution 1, stir magnetically at 300 r / min for 30 min, and then sonicate at 40 kHz power and 350 W power for 30 min to obtain vegetable freshness indicator 1.

[0058] S3. Pour the vegetable freshness indicator 1 obtained in step S2 into a mold with a diameter of 9cm (pour 30mL of vegetable freshness indicator 1 into each mold), then transfer the mold to a drying oven at 50℃ and dry for 12h. The film is then removed to obtain the vegetable freshness indicator label 1.

[0059] Example 2: A vegetable freshness indicator label prepared based on nanomaterials A vegetable freshness indicator label based on nanomaterials is prepared as follows: S1. Weigh 1.5g of chitosan and add it to 100mL of acetic acid solution (acetic acid volume concentration is 1.0%). Place it at 25℃ and stir magnetically at 800r / min for 40min until the solution is transparent and free of particles. Then add Tween 80 with a final volume concentration of 0.5%. After mixing evenly, use a high-pressure homogenizer to homogenize 8 times at 30MPa pressure, then 8 times at 50MPa pressure, and then 8 times at 80MPa pressure. After homogenization, sonicate at 40kHz frequency and 350W power for 40min to obtain nano-chitosan solution 2. Each homogenization time is 10min. Weigh 4.5g of corn starch and add it to 100mL of deionized water. Stir magnetically at 800r / min for 25min to obtain a starch suspension. Place the starch suspension in a 92℃ water bath and heat with magnetic stirring at 800r / min for 40min. Then cool it to below 25℃ in an ice water bath. Add Tween 80 with a final volume concentration of 0.5%. Mix well and use a high-pressure homogenizer to homogenize 8 times at 30MPa, then 8 times at 60MPa, and then 15 times at 120MPa. After homogenization, sonicate at 40kHz and 380W for 40min to obtain nano starch solution 2. Each homogenization time is 10min.

[0060] S2. Mix the nano starch solution 2 and nano chitosan solution 2 prepared in step S1 at a mass ratio of 6:4 until homogeneous. Then add glycerol with a final mass concentration of 2.0% and nano silica with a mass concentration of 0.5%. Then stir magnetically at 800 r / min for 130 min in a 60°C water bath. Then cool the temperature to below 25°C in an ice water bath to obtain the solution 2 to be used. Adjust the pH of the prepared solution 2 to pH=11. Then, weigh 0.005% of the mass of prepared solution 2 and dissolve it in 1 mL of anhydrous ethanol to obtain m-cresol purple solution 2. Add the m-cresol purple solution to prepared solution 2, stir magnetically at 300 r / min for 25 min, and then sonicate at 40 kHz power and 350 W power for 30 min to obtain vegetable freshness indicator 2.

[0061] S3. Pour the vegetable freshness indicator 2 obtained in step S2 into a mold with a diameter of 9cm (pour 30mL of vegetable freshness indicator 2 into each mold), then transfer the mold to a drying oven at 50℃ and dry for 12h. The film is then removed to obtain the vegetable freshness indicator label 2.

[0062] Example 3: A vegetable freshness indicator label based on nanomaterials A vegetable freshness indicator label based on nanomaterials is prepared as follows: S1. Weigh 1.8g of chitosan and add it to 100mL of acetic acid solution (acetic acid solution volume concentration is 1.0%). Place it at 25℃ and stir magnetically at 800r / min for 30min until the solution is transparent and free of particles. Then add Tween 80 with a final volume concentration of 0.5%. After mixing evenly, use a high-pressure homogenizer to homogenize 6 times at 30MPa pressure, then 6 times at 50MPa pressure, and then 7 times at 80MPa pressure. After homogenization, sonicate at 40kHz frequency and 300W power for 40min to obtain nano-chitosan solution 3. Each homogenization time is 10min. Weigh 4.0g of corn starch and add it to 100mL of deionized water. Stir magnetically at 800r / min for 20min to obtain a starch suspension. Place the starch suspension in a 90℃ water bath and heat with magnetic stirring at 800r / min for 50min. Then cool it to below 25℃ in an ice water bath. Add Tween 80 with a final volume concentration of 0.5%. Mix well and use a high-pressure homogenizer to homogenize 6 times at 30MPa, then 6 times at 60MPa, and then 12 times at 120MPa. After homogenization, sonicate at 40kHz and 400W for 35min to obtain nano starch solution 3. Each homogenization time is 10min.

[0063] S2. Mix the nano starch solution 3 and nano chitosan solution 3 prepared in step S1 at a mass ratio of 7:3 until homogeneous. Then add glycerol with a final mass concentration of 3.0% and nano silica with a mass concentration of 1.0%. Then stir magnetically at 800 r / min for 150 min in a 60°C water bath. Then cool the temperature to below 25°C in an ice water bath to obtain the solution 3 to be used. Adjust the pH of the prepared solution 3 to pH=11. Then, weigh 0.005% of the mass of prepared solution 3 and dissolve it in 1 mL of anhydrous ethanol to obtain m-cresol purple solution 3. Add m-cresol purple solution 3 to prepared solution 3, stir magnetically at 300 r / min for 20 min, and then sonicate at 40 kHz power and 350 W power for 50 min to obtain vegetable freshness indicator 3.

[0064] S3. Pour the vegetable freshness indicator 3 obtained in step S2 into a mold with a diameter of 9cm (pour 30mL of vegetable freshness indicator 3 into each mold), then transfer the mold to a drying oven at 50℃ and dry for 12h. The film is then removed to obtain the vegetable freshness indicator label 3.

[0065] Comparative Example 1: A label indicating the freshness of vegetables A vegetable freshness indicator label, prepared as follows: S1. Weigh 1.0 g of chitosan, add 100 mL of acetic acid solution (acetic acid concentration is 1.0%), place at 25℃ and magnetically stir at 800 r / min for 60 min until the solution is transparent and free of particles to obtain chitosan solution; Nano starch solution was prepared according to step S1 of Example 1.

[0066] S2. Mix the nano starch solution and chitosan solution obtained in step S1 at a mass ratio of 6:4, then add glycerol with a final mass concentration of 2.0%, and then stir magnetically at 800 r / min for 120 min in a 60℃ water bath. Then cool the temperature to below 25℃ in an ice water bath to obtain solution 4 for use. Adjust the pH of the prepared solution 4 to pH=11. Then, weigh 0.005% of the mass of prepared solution 4 and dissolve it in 1 mL of anhydrous ethanol to obtain m-cresol purple solution 4. Add m-cresol purple solution 4 to prepared solution 4, stir magnetically at 300 r / min for 30 min, and then sonicate at 40 kHz power and 350 W power for 30 min to obtain vegetable freshness indicator 4.

[0067] S3. Pour the vegetable freshness indicator 3 obtained in step S2 into a mold with a diameter of 9cm (pour 30mL of vegetable freshness indicator 4 into each mold), then transfer the mold to a drying oven at 50℃ and dry for 12h. The film is then removed to obtain the vegetable freshness indicator label 4.

[0068] Comparative Example 2: A label indicating the freshness of vegetables A vegetable freshness indicator label, prepared as follows: S1. Weigh 1.0 g of chitosan and add it to 100 mL of acetic acid solution (acetic acid concentration is 1.0%). Place it at 25℃ and stir magnetically at 800 r / min for 30 min until the solution is transparent and free of particles. Then add Tween 80 with a final volume concentration of 0.5%. After mixing evenly, use a high-pressure homogenizer to homogenize twice at 10 MPa pressure, then homogenize five times at 20 MPa pressure, and then homogenize five times at 30 MPa pressure. After homogenization, sonicate at 40 kHz frequency and 200 W power for 20 min to obtain nano-chitosan solution 4. The homogenization time for each step is 10 min. Weigh 5.0g of corn starch and add it to 100mL of deionized water. Stir magnetically at 800r / min for 30min to obtain a starch suspension. Place the starch suspension in a 90℃ water bath and heat with magnetic stirring at 800r / min for 30min. Then cool it to below 25℃ in an ice water bath. Add Tween 80 with a final volume concentration of 0.5%. Mix well and use a high-pressure homogenizer to homogenize twice at 10MPa pressure, then five times at 20MPa pressure, and then five times at 30MPa pressure. After homogenization, sonicate at 40kHz frequency and 200W power for 20min to obtain nano starch solution 4. Each homogenization time is 10min.

[0069] S2. Mix the nano starch solution 4 and nano chitosan solution 4 prepared in step S1 at a mass ratio of 6:4 until homogeneous. Then add glycerol with a final mass concentration of 2.0% and nano silica with a mass concentration of 0.5%. Then stir magnetically at 800 r / min for 2 h in a 60°C water bath. Then cool the temperature to below 25°C in an ice water bath to obtain the ready-to-use solution 5. Adjust the pH of the prepared solution 5 to pH=11. Then, weigh 0.005% of the mass of prepared solution 5 and dissolve it in 1 mL of anhydrous ethanol to obtain m-cresol purple solution 5. Add m-cresol purple solution 5 to prepared solution 5, stir magnetically at 300 r / min for 30 min, and then sonicate at 40 kHz power and 350 W power for 30 min to obtain vegetable freshness indicator 5.

[0070] S3. Pour the vegetable freshness indicator 5 obtained in step S2 into a mold with a diameter of 9cm (30mL of vegetable freshness indicator 5 is poured into each mold), then transfer the mold to a drying oven at 50℃ and dry for 12h. The film is then removed to obtain the vegetable freshness indicator label 5.

[0071] Comparative Example 3: A Vegetable Freshness Indication Label A vegetable freshness indicator label, prepared as follows: S1. Prepare a nano-chitosan solution as shown in step S1 of Example 1; Weigh 5.0g of corn starch and add it to 100mL of deionized water. Stir magnetically at 800r / min for 30min to obtain a starch suspension. Place the starch suspension in a 90℃ water bath and heat with magnetic stirring at 800r / min for 30min to obtain a starch solution.

[0072] S2. Mix the starch solution and nano-chitosan solution obtained in step S1 at a mass ratio of 6:4, then add glycerol with a final mass concentration of 2.0% and nano-silica of 0.5%, and then stir magnetically at 800 r / min for 2 h in a 60°C water bath. Then cool the temperature to below 25°C in an ice water bath to obtain solution 6 for use. Adjust the pH of the prepared solution 6 to pH=11. Then, weigh 0.005% of the mass of prepared solution 6 and dissolve it in 1 mL of anhydrous ethanol to obtain m-cresol purple solution 6. Add m-cresol purple solution 6 to prepared solution 6, stir magnetically at 300 r / min for 30 min, and then sonicate at 40 kHz power and 350 W power for 30 min to obtain vegetable freshness indicator 6.

[0073] S3. Pour the vegetable freshness indicator 6 obtained in step S2 into a mold with a diameter of 9cm (30mL of vegetable freshness indicator 6 is poured into each mold), then transfer the mold to a drying oven at 50℃ and dry for 12h. The film is then removed to obtain the vegetable freshness indicator label 6.

[0074] Test Example 1: Performance Testing of Nanosolutions and Indicator Labels I. Testing of Nano Solutions 1. Experimental Methods The nano-chitosan solution 1 prepared in Example 1, the nano-chitosan solution 2 prepared in Example 2, the nano-chitosan solution 3 prepared in Example 3, and the nano-chitosan solution 4 prepared in Comparative Example 2 were diluted in enzyme-free water to obtain nano-chitosan solutions with a volume concentration of 0.001%.

[0075] The nano starch solution 1 prepared in Example 1, the nano starch solution 2 prepared in Example 2, the nano starch solution 3 prepared in Example 3, and the nano starch solution 4 prepared in Comparative Example 2 were diluted in enzyme-free water to obtain nano starch solutions with a volume concentration of 0.005%.

[0076] Next, the particle size of the nano-chitosan solution with a volume concentration of 0.001% and the nano-starch solution with a volume concentration of 0.005% were measured using a particle size analyzer (model: NanoBrook omni multi-angle particle size analyzer; manufacturer: Brookhaven Instruments, USA). Each nano-chitosan solution and each nano-starch solution were measured in parallel three times, and the average value was taken as the final particle size.

[0077] 2. Experimental Results The particle size determination results of each nano-chitosan solution and each nano-starch solution are shown in the figure below. Figure 1As shown, the results indicate that the particle size of the nano-chitosan solution prepared in Examples 1 to 3 is about 300 nm, and the particle size of the nano-starch solution is less than 100 nm; the particle size of the nano-chitosan solution 4 prepared in Comparative Example 2 is about 800 nm, and the particle size of the nano-starch solution 4 is about 600 nm.

[0078] II. Performance Testing of Indicator Labels 1. Experimental Methods (1) Mechanical performance testing The vegetable freshness indicator labels 1 to 6 prepared in Examples 1 to 3 and Comparative Examples 1 to 3 were cut into rectangular strips of 10mm × 50mm and fixed with a clamping gap of 45mm. The tensile strength of each indicator label was measured at a speed of 1mm / min using a texture analyzer (model: TA-XT plus, manufacturer: Stable Micro Systems, UK).

[0079] The mechanical properties of the indicator labels were tested using a texture analyzer. The indicator labels were cut into rectangular strips of 10mm × 50mm and fixed with a clamping interval of 45mm. The tensile strength was measured at a speed of 1mm / min.

[0080] (2) Thermogravimetric analysis test 5 mg of each of the vegetable freshness indicator labels 1 to 6 prepared in Examples 1 to 3 and Comparative Examples 1 to 3 were weighed out. Each vegetable freshness indicator label was placed in the tray of a thermogravimetric analyzer (model: TGA2, manufacturer: Mettler Toledo AG, Switzerland). The initial temperature was 30°C, and the temperature was increased to 600°C at a rate of 20°C / min. Thermogravimetric analysis was performed on each vegetable freshness indicator label. The flow rate of N2 during the thermogravimetric analysis was 20 mm / min.

[0081] 2. Experimental Results The mechanical strength results of vegetable freshness indicator labels 1 to 6 are shown in the figure below. Figure 2 As shown, the results indicate that the mechanical strength of vegetable freshness indicator labels 1 to 3 prepared in Examples 1 to 3 is significantly better than that of vegetable freshness indicator labels 4 to 6 prepared in Comparative Examples 1 to 3, and there is no significant difference in mechanical strength among vegetable freshness indicator labels 1 and 3.

[0082] This demonstrates that only vegetable freshness indicator labels prepared according to Examples 1 to 3 can possess excellent mechanical properties.

[0083] The weight retention rate results of vegetable freshness indicator labels 1 to 6 are shown in the figure below. Figure 3 As shown, the results indicate that the weight retention rates of vegetable freshness indicator labels 1 to 3 prepared in Examples 1 to 3 are significantly higher than those of vegetable freshness indicator labels 4 to 6 under the same temperature conditions, indicating that vegetable freshness indicator labels 1 to 3 have better stability.

[0084] Test Example 2: Application of indicator labels in determining vegetable freshness I. Experimental Methods 1. Application of Vegetable Freshness Indication Labels Twenty-two portions of fresh broccoli (50g each) were selected. Each portion of fresh broccoli was placed in a transparent food storage container (12cm×12cm×9cm). The vegetable freshness indicator label 1 prepared in Example 1 was then cut into circular pieces with a diameter of 1.5cm and pasted onto filter paper strips (11cm×2cm, with 5 circular pieces pasted on each filter paper strip). The filter paper strips were then fixed to the inside of the food storage container (the vegetable freshness indicator label 1 was located between the filter paper strip and the container lid). Each food storage container was then placed in a refrigerator at 20°C for 6 days.

[0085] During the preservation process, each food preservation container was photographed and observed daily. The photography was conducted every hour in a light box with D65 illumination mode, for a total of 13 hours, resulting in 286 images per day. Simultaneously, during the preservation period, an electronic nose was used to measure the odor of each food preservation container at the same time each day, and principal component analysis of the odor was performed using Origin software plotting analysis methods. The preservation and processing conditions of each food preservation container were consistent.

[0086] Then, on the 1st, 2nd, 3rd, 4th, 5th and 6th days of storage, the vegetable freshness indicator label 1 in each food storage container was photographed every hour in D65 lighting mode, for a total of 13 hours, resulting in 1716 images of the vegetable freshness indicator label 1 in each food storage container during the storage process. Then, 1500 images of the vegetable freshness indicator label 1 (including 250 images per day) were randomly selected from the 1716 images of the vegetable freshness indicator label 1 as an image set.

[0087] The image set containing 1500 images of vegetable freshness indicator labels 1 was randomly divided into a training set (containing 1050 images) and a test set (containing 450 images). At the same time, the freshness of broccoli corresponding to each image in the training set and the test set was judged according to the freshness status judgment criteria of broccoli shown in Table 1, so as to obtain the actual freshness result of broccoli.

[0088] Table 1. Criteria for Judging the Freshness of Broccoli

[0089] The images of each vegetable freshness indicator label 1 in the training set are used as input to a deep convolutional neural network model (Xception). The freshness of broccoli (fresh, slightly fresh, and rotten) is used as the output of the deep convolutional neural network model. The deep convolutional neural network model is trained by 10-fold cross-validation by combining the actual freshness results of broccoli corresponding to each image in the training set, and a trained deep convolutional neural network model (the deep convolutional neural network model combined with vegetable freshness indicator label 1) is obtained. The deep convolutional neural network model includes an input layer, a convolutional layer, a pooling layer, a fully connected layer, and an output layer.

[0090] Next, the color value of the vegetable freshness indicator label 1 in the images of each vegetable freshness indicator label 1 in the test set is used as the input of the trained deep convolutional neural network model to obtain the broccoli freshness (prediction result) output by the model. The training accuracy of the trained deep convolutional neural network is judged by combining the actual broccoli freshness results corresponding to the images of each vegetable freshness indicator label 1 in the test set, and a confusion matrix is ​​drawn.

[0091] 2. Application of Vegetable Freshness Indication Labels 2 to 6 As shown in the application of vegetable freshness indicator label 1, vegetable freshness indicator label 1 was replaced with vegetable freshness indicator labels 2 to 6 respectively, and the same processing was performed. The results of photographic observation and odor principal component analysis were recorded for each day when broccoli was stored using each vegetable freshness indicator label. The deep convolutional neural network model combining vegetable freshness indicator labels 2 to 6 and the training accuracy and confusion matrix of each deep neural network model were obtained respectively.

[0092] II. Experimental Results When judging the freshness of broccoli according to the criteria shown in Table 1, broccoli is considered fresh if stored for 1-2 days, moderately fresh if stored for 3-4 days, and rotten if stored for 5-6 days.

[0093] The following are the results of photographic observations of broccoli stored using various vegetable freshness indicator labels: Figure 4As shown, the results indicate that the vegetable freshness indicator labels prepared in Examples 1 to 3 exhibited good color changes during broccoli storage. In the initial storage period (1-2 days), the broccoli was fresh, and the vegetable freshness indicator label was blue-purple. As the storage time increased, it gradually turned purplish-red, and in the later storage period (5-6 days), it turned yellow, corresponding to the broccoli's spoilage and inedibility. The color changes of the vegetable freshness indicator labels prepared in Examples 1 to 3 during broccoli storage were relatively consistent with the changes in the freshness of the broccoli. However, the vegetable freshness indicator labels prepared in Comparative Examples 1 to 3 remained red in the later storage period (6 days), while the broccoli had already spoiled, indicating that they could not effectively indicate the changes in freshness of broccoli during storage.

[0094] The principal component analysis results of the aroma of broccoli stored using various vegetable freshness indicator labels are shown in the figure below. Figure 5 As shown, Figure 5 In the figure, A represents the principal component analysis results of odor when combined with vegetable freshness indicator label 1. Figure 5 B in the figure represents the principal component analysis results of odor when combined with vegetable freshness indicator label 2. Figure 5 C in the figure represents the principal component analysis results of odor when combined with vegetable freshness indicator label 3. Figure 5 In the figure, D represents the principal component analysis results of odor when combined with the vegetable freshness indicator label 4. Figure 5 E in the figure represents the principal component analysis results of odor when combined with the vegetable freshness indicator label 5. Figure 5 F in the figure represents the principal component analysis results of odor when combined with the vegetable freshness indicator label 6.

[0095] The results showed that when broccoli was stored according to the freshness indicator labels of various vegetables, the odor of broccoli stored for 0-2 days was not significantly different, the odor of broccoli stored for 3-4 days was not significantly different but was significantly different from that of 0-2 days, and the odor of broccoli stored for 5-6 days was not significantly different. This indicates that the three time periods correspond to fresh broccoli, slightly fresh broccoli, and rotten (spoiled) broccoli, respectively.

[0096] The training accuracy results of the deep convolutional neural network model combining various vegetable freshness indicator labels are shown in the figure below. Figure 6 As shown, Figure 6 In the figure, A represents the training accuracy result of the deep convolutional neural network model that incorporates the vegetable freshness indicator label 1. Figure 6 In the figure, B represents the training accuracy result of the deep convolutional neural network model that incorporates the vegetable freshness indicator label 2. Figure 6 In the figure, C represents the training accuracy of the deep convolutional neural network model that incorporates the vegetable freshness indicator label 3. Figure 6In the figure, D represents the training accuracy of the deep convolutional neural network model that incorporates the vegetable freshness indicator label 4. Figure 6 E in the figure represents the training accuracy of the deep convolutional neural network model that incorporates the vegetable freshness indicator label 5. Figure 6 F in the figure represents the training accuracy of the deep convolutional neural network model that incorporates the vegetable freshness indicator label 6.

[0097] The results show that the training accuracy of the deep convolutional neural network models of vegetable freshness indicator labels 1 to 3 prepared in combination with Examples 1 to 3 is significantly higher, indicating that vegetable freshness indicator labels 1 to 3 are more suitable for assembling machine learning models to achieve more accurate monitoring of vegetable freshness.

[0098] The confusion matrix of the deep convolutional neural network model combining the freshness indicator labels of various vegetables is as follows: Figure 7 As shown, Figure 7 In this context, A represents the confusion matrix of a deep convolutional neural network model that incorporates the vegetable freshness indicator label 1. Figure 7 In the diagram, B represents the confusion matrix of a deep convolutional neural network model that incorporates the vegetable freshness indicator label 2. Figure 7 In the diagram, C represents the confusion matrix of a deep convolutional neural network model that incorporates the vegetable freshness indicator label 3. Figure 7 In this context, D represents the confusion matrix of a deep convolutional neural network model that incorporates the vegetable freshness indicator label 4. Figure 7 In this context, E represents the confusion matrix of a deep convolutional neural network model that incorporates the vegetable freshness indicator label 5. Figure 7 F in the equation is the confusion matrix of a deep convolutional neural network model that incorporates the vegetable freshness indicator label 6.

[0099] The results showed that when using a deep convolutional neural network model incorporating vegetable freshness indicator label 1 to predict the freshness of 450 broccoli samples during the preservation process, it was able to effectively distinguish between fresh, slightly rotten, and spoiled broccoli, with an accuracy rate of 98.88%. The accuracy rate was 98.25% when using a deep convolutional neural network model incorporating vegetable freshness indicator label 2, and 97.95% when using a deep convolutional neural network model incorporating vegetable freshness indicator label 3.

[0100] The accuracy rate of the deep convolutional neural network model combining vegetable freshness indicator label 4 for predicting the freshness of 450 broccoli samples during preservation was 85.33%, the accuracy rate of the deep convolutional neural network model combining vegetable freshness indicator label 5 was 85.98%, and the accuracy rate of the deep convolutional neural network model combining vegetable freshness indicator label 6 was 86.50%. The accuracy rates of these models were all much lower than those of the deep convolutional neural network models combining vegetable freshness indicator labels 1 to 3 for predicting the freshness of broccoli during preservation. Furthermore, the deep convolutional neural network models combining vegetable freshness indicator labels 4 to 6 could not effectively and accurately determine whether the broccoli had spoiled during preservation.

[0101] The results show that the vegetable freshness indicator labels prepared according to the methods shown in Examples 1 to 3, combined with a deep neural network model, have significantly better predictive and generalization abilities for broccoli freshness, and can more accurately predict broccoli freshness in actual prediction processes.

[0102] Test Example 3: Sensitivity Test of Indicator Labels to CO2 I. Experimental Methods Take a transparent vacuum packaging bag (20cm×30cm) and attach a silicone sheet with adhesive (1cm in diameter) to the outer wall of the bag, 10cm from the opening.

[0103] The vegetable freshness indicator labels 1 to 6 prepared in Examples 1 to 3 and Comparative Examples 1 to 3 were cut into 1cm×2cm pieces. The cut vegetable freshness indicator labels were then pasted into different vacuum packaging bags using double-sided tape. The pasting position was on the inner wall 5cm away from the bag opening.

[0104] Next, put 10ml of distilled water into the vacuum packaging bag with the vegetable freshness indicator label, then vacuum pack it, and then insert the CO2 inflation head into the vacuum packaging bag through the silicone sheet, and fill the vacuum packaging bag with CO2 through the CO2 inflation head until the volume concentration of CO2 in the vacuum packaging bag reaches 10%.

[0105] Next, the vacuum-packed bag was placed at 25°C, and the color change of the vegetable freshness indicator label was observed. The L, a, and b values ​​were then photographed and recorded under D65 lighting mode.

[0106] II. Experimental Results The results of the CO2 sensitivity test for each vegetable freshness indicator label are shown in Table 2.

[0107] Table 2. Sensitivity test results of various vegetable freshness indicator labels to CO2.

[0108] The results showed that, compared with the vegetable freshness indicator labels 4 to 6 prepared in Comparative Examples 1 to 3, the vegetable freshness indicator labels 1 to 3 prepared in Examples 1 to 3 were more sensitive to the indication of acidic gas (CO2). The vegetable freshness indicator labels 1 to 3 were more sensitive to the color change upon contact with CO2, and the color change of the indicator labels was faster (specifically, the a value decreased lower and the b value changed faster).

[0109] The vegetable freshness indicator labels 1 to 3 prepared in Examples 1 to 3 are more suitable for monitoring the freshness of vegetables (CO2 is generated during storage).

[0110] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the scope of protection of the present invention. For those skilled in the art, other variations or modifications can be made based on the above description and ideas, and it is neither necessary nor possible to exhaustively describe all implementation methods here. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the claims of the present invention.

Claims

1. A method for preparing a vegetable freshness indicator based on nanomaterials, characterized in that, Includes the following steps: S1. Prepare nano-chitosan solution and nano-starch solution respectively; S2. Mix the nano starch solution and nano chitosan solution obtained in step S1 at a mass ratio of 6-7:4-3 until homogeneous. Then add glycerol with a final mass concentration of 2.0%-3.0% and nano silica with a mass concentration of 0.5%-1.0%. After mixing thoroughly, add an acid-base indicator and sonicate to obtain a vegetable freshness indicator.

2. The method according to claim 1, characterized in that, The preparation method of the nano-chitosan solution in step S1 is as follows: chitosan and acetic acid solution are mixed evenly and then Tween 80 is added. Then, the mixture is homogenized at 30 MPa for 5 to 8 times, at 50 MPa for 5 to 8 times, and at 80 MPa for 5 to 8 times. After homogenization, ultrasonic treatment is performed to obtain the nano-chitosan solution. The volume concentration of the acetic acid solution is 1.0–2.0%, and the homogenization time is 5–20 min each time.

3. The method according to claim 1, characterized in that, The preparation method of the nano starch solution in step S1 is as follows: after mixing starch and water evenly, add Tween 80, then homogenize at 30MPa for 5 to 8 times, at 60MPa for 5 to 8 times, and at 120MPa for 10 to 15 times in sequence. After homogenization, perform ultrasonic treatment to obtain the nano starch solution. The homogenization time for each step is 5 to 20 minutes.

4. The method according to claim 1, characterized in that, The acid-base indicator mentioned in step S2 is m-cresol purple.

5. The vegetable freshness indicator prepared by the method according to any one of claims 1 to 4.

6. A vegetable freshness indicator label based on nanomaterials, characterized in that, The vegetable freshness indicator label is prepared by forming a film of the vegetable freshness indicator according to claim 5.

7. The application of the vegetable freshness indicator of claim 5 and / or the vegetable freshness indicator label of claim 6 in determining the freshness of vegetables.

8. A method for determining the freshness of vegetables, characterized in that, Includes the following steps: S11. The vegetable freshness indicator of claim 5 and / or the vegetable freshness indicator label of claim 6 are stored in the same environment as the vegetables, and the vegetable freshness indicator and / or vegetable freshness indicator label are photographed during the storage process to obtain photographs of the vegetable freshness indicator and / or vegetable freshness indicator label; S12. Input the photos of the vegetable freshness indicator and / or vegetable freshness indicator label obtained in step S11 into the trained machine learning model, and the trained machine learning model outputs the freshness of the vegetables; The trained machine learning model is a machine learning model trained using photos of vegetables with known freshness and their corresponding vegetable freshness indicators and / or vegetable freshness indicator labels. The freshness of the vegetables includes fresh, slightly fresh, and rotten.

9. The method according to claim 8, characterized in that, The photo taken in step S11 is taken using a D65 standard light source.

10. The method according to claim 8, characterized in that, The trained machine learning model mentioned in step S12 is a trained deep convolutional neural network model.

Images