A method for detecting salt additive content based on near-infrared spectroscopy imaging

Abstract

The invention discloses a method for detecting the content of salt additives based on near-infrared spectral imaging. The method comprises the following steps: firstly, salt samples before and after spraying additives at different depths on an assembly line are collected, and the salt samples at different depths are combined and spread on a platform to be tested; then, the salt on the platform is imaged and measured using a near-infrared spectrum; the average water content of the whole salt before and after spraying the additives is estimated by using the near-infrared spectrum; and finally, the content of the salt additives is estimated by using the average water content of mixed salt samples at different depths before and after spraying the additives. The method of the invention can well eliminate the interference of impurities in the salt on the additives, has the characteristics of accurate detection results, fast speed, and no damage to the detection samples, and is simple and convenient to operate, and is suitable for on-line detection of the assembly line.

Description

Technical Field

[0001] The invention belongs to the technical field of salt additive content detection, and in particular relates to a salt additive content detection method based on near-infrared spectral imaging. Background Art

[0002] In addition to the refined salt, crushed washing salt, sun-dried salt, low-sodium salt, and iodine-fortified salt specified in the "National Food Safety Standard for Edible Salt" (GB2721-2015), the salt currently circulating in the market also includes salt with added nutritional enhancers, seasoning additives, or processed through special processes, such as seasoning salt, seaweed salt, pickling salt, kimchi salt, etc. Among them, iodized salt is the most common salt on the market.

[0003] Iodine failure is indeed worthy of attention. The reason for failure is mostly low iodine content, and most of the unqualified types are salts containing reducing substances such as pickling salt, seasoning salt and seaweed salt. In the salt industry, the detection method of iodine in products stipulates that the redox titration method is used as the arbitration method. In particular, for iodized edible salt containing reducing substances, if the testing agency chooses a different detection method, it will interfere with the iodine determination results, and the measured iodine content will be low. Therefore, a method for detecting the content of additives in salt that is fast, accurate and suitable for the determination of large quantities of samples is needed.

[0004] Near infrared spectroscopy is a technique that absorbs electromagnetic radiation with wavelengths between 780 and 2500 nm. All organic bonds have absorption bands in the near infrared region, and water in particular shows a strong signal in this part of the spectrum. Samples with high water content are strongly dominated by the water signal.

[0005] In the current industrial production line, the manufacturing process of sea salt is stable, and various additives are mixed with water to form a solution, which is added to the salt as a spray solution on the production line. The input of the additive mixture is linearly related to the water content. Therefore, the amount of additive input can be determined by detecting the change in water content. Summary of the invention

[0006] The object of the present invention is to provide a method for detecting the content of salt additives based on near-infrared spectral imaging. The method is based on spectral imaging technology and can quickly judge the content of salt additives without destroying the object to be measured. The obtained results are stable and the operation is simple and convenient.

[0007] To achieve the above object, the technical solution adopted by the present invention is: a method for detecting the content of salt additives based on near-infrared spectral imaging, characterized in that it comprises the following steps:

[0008] A method for detecting salt additive content based on near infrared spectroscopy imaging comprises the following steps:

[0009] (1) Multi-level sampling: Collect salt samples at five depths on the production line before and after spraying additives, and spread the salt combinations at different depths on the test platform;

[0010] (2) Near-infrared spectroscopy imaging measurement: Use near-infrared spectroscopy to image and measure salt at different depths on the platform;

[0011] (3) Moisture estimation: The average moisture content of the salt before and after spraying the additive is estimated by near infrared spectroscopy;

[0012] (4) Estimation of additive content: The additive content of salt is estimated by measuring the average water content of salt samples mixed at different depths before and after spraying additives.

[0013] Furthermore, the step (1) is specifically as follows: salt samples at different depths are respectively on the upper surface of the assembly line, 1 / 4 from the surface, 2 / 4 from the surface, 3 / 4 from the surface and the bottom layer, the total weight of the salt sample must be ≥50g, 10g is taken at each depth, the platform to be tested is a black plastic bottom plate, an active light source is set under dark room conditions, and the sampling positions are respectively 10cm in front and 10cm behind the spray head of the assembly line.

[0014] Furthermore, the active light source is a 50W halogen lamp.

[0015] Furthermore, the step (2) is specifically as follows: the wavelength of the near-infrared spectrometer is 900-1700nm. During the measurement, the platform to be tested is in a dark room, the distance between the platform to be tested and the imaging lens is 25cm, and the salt samples of different depths are evenly stirred and spread into a circle with a thickness of ≥20mm and a diameter of ≥300mm.

[0016] Furthermore, the salt water content estimation method is as follows: correcting the spectral image of the salt sample, using the threshold method to delineate the region of interest (ROI) of the salt sample at different depths, extracting the average spectrum of all pixels in the ROI, and describing the extracted average spectrum data with a three-dimensional data block, wherein x and y represent two-dimensional plane pixel information, and the third dimension λ is wavelength information; the three-dimensional data block can display the difference information of the physical structure and chemical composition of the salt sample; using the continuous projection method SPA to extract 10 optimal characteristic wavelengths: 926nm, 951nm, 1003nm, 1343nm, 1360nm, 1385nm, 1392nm, 1434nm, 1441nm, 1700nm; substituting the spectral data corresponding to the characteristic wavelengths extracted by SPA into the least squares regression algorithm PLSR for modeling, and obtaining the calculated moisture content of the salt corresponding to each pixel in the hyperspectral image of the salt.

[0017] Specifically, the formula for the moisture content of table salt is:

[0018] Y=(6.28-1.94X 926nm +49.74X 951nm +18.23X 1003nm +32.97X 1343nm +85.34X 1360nm +44.61X 1385nm +87.39X 1392nm +32.10X 1434nm -28.87X 1441nm -61.43X 1700nm )×10 -5

[0019] Where Xi is the spectral value of a pixel in the spectral image at the characteristic wavelength of inm; Y is the salt water content value at the corresponding pixel.

[0020] Specifically, the salt additive content is calculated according to the following formula:

[0021] p=(YX)×k×10 6 ;

[0022] Wherein, Y is the water content of the salt sample after spraying the additive, X is the water content of the salt sample before spraying the additive, k is the content of each additive in the additive solution, the unit is g / 100g, the additives include but are not limited to iodine and ferrous oxide, which are obtained according to the ratio of the sprayed additives, and p is the content of the additive in the salt sample, the unit is mg / kg.

[0023] The beneficial effects of the present invention are:

[0024] (1) The present invention provides a method for detecting the content of additives in table salt, which can effectively eliminate the interference of impurities in table salt on the additives, has the characteristics of accurate detection results and fast speed, and the operation method is simple and convenient.

[0025] (2) Compared with the existing national standard titration method, the detection method of the present invention is faster and does not damage the salt sample, and is suitable for on-line detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0027] Figure 1 It is a flow chart of a method for detecting salt additive content based on near infrared spectroscopy imaging of the present invention;

[0028] Figure 2 It is the near infrared spectra of 5 kinds of mixed salt samples with different depths when the weight of salt is 50g. DETAILED DESCRIPTION

[0029] Here, exemplary embodiments are described in detail, and examples thereof are shown in the accompanying drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application.

[0030] The terms used in this application are for the purpose of describing specific embodiments only and are not intended to limit this application. The singular forms of "a", "said" and "the" used in this application and the appended claims are also intended to include plural forms unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more associated listed items.

[0031] The content of the present invention is further explained below in conjunction with specific examples, but the content protected by the present invention is not limited thereto.

[0032] like Figure 1 and Figure 2 As shown, a method for detecting salt additive content based on near infrared spectroscopy imaging comprises the following steps:

[0033] 1) Multi-stage sampling: before and after spraying the additive, the salt on the assembly line is sampled once, and the sampling positions are 10 cm in front and 10 cm behind the spray head of the assembly line. The specific process of each sampling is to collect 10g of salt samples at 5 depths. The 5 depths are the upper surface of the assembly line, 1 / 4 from the surface, 2 / 4 from the surface, 3 / 4 from the surface and the bottom layer. The samples are evenly stirred and spread on the test platform. The test platform is a darkroom environment, the platform surface is black, and the material is PVC plastic material. The only light source is a 50w halogen lamp.

[0034] The specific tiling method is: circular, with a thickness of ≥20mm and a diameter of ≥300mm. When the tiling thickness of the salt is less than 20mm, the spectrometer will obtain the spectral information of the vehicle on the salt bottom, affecting the analysis. If the tiling diameter is less than 300mm, the spectrometer will receive the spectral information of the salt and surrounding objects of no concern, affecting the analysis. A diameter greater than 300mm ensures that the spectrometer sampling field of view is full of salt.

[0035] 2) Near-infrared spectral imaging measurement: Use a near-infrared spectrometer to take white frames and black frames, and then use the near-infrared spectrum to image the salt sample on the test platform. The salt sample is 25 cm away from the imaging lens. The spectral image is exported and data processed. The processing method is to select the pseudo-color image taken, perform black and white correction on the image through the black and white frame, select the area where the 5 kinds of deep mixed salt samples are located as ROI, and count the average spectral curves in each ROI area within 900-1700nm, so as to obtain the average reflectivity of the 5 kinds of deep mixed salt samples before and after spraying at 900-1700nm. The imaging range of the near-infrared spectrometer is 900-1700nm. Exposure is automatic exposure;

[0036] 3) Moisture estimation: The data obtained in 2) is imported into the moisture-near infrared spectrum model calculation formula obtained by prior training. The formula is as follows:

[0037] Y=(6.28-1.94X 926nm +49.74X 951nm +18.23X 1003nm +32.97X 1343nm +85.34X 1360nm +44.61X 1385nm +87.39X 1392nm +32.10X 1434nm -28.87X 1441nm -61.43X 1700nm )×10 -5

[0038] Where Xi is the spectral value of a pixel in the spectral image at the characteristic wavelength of inm; Y is the salt water content value at the corresponding pixel.

[0039] 4) Estimation of additive content: Substitute the water content of the five kinds of deeply mixed salt samples before and after spraying the additive obtained in 3) into the following formula for calculation:

[0040] p=(YX)×k×10 6

[0041] In the formula: Y is the water content of the salt sample after spraying the additive, X is the water content of the salt sample before spraying the additive, k is the content of each additive in the additive solution, the unit is g / 100g, the additives include but are not limited to iodine and ferrous oxide, which are obtained according to the ratio of the sprayed additives, p: the content of additives in the salt sample, the unit is mg / kg.

[0042] Thus, the average additive content p of salt in the running water after the additive is sprayed is obtained.

[0043] Example 1 (50g salt)

[0044] The method is as described above, taking iodine as an example of an additive. A 50g salt sample is collected, and the model calculates that the water content difference is 0.43%, so the iodine content of the salt sample is determined to be 22.98mg / kg. The actual iodine content of the salt is 24mg / kg, with an error of 4.25%;

[0045] Example 2 (80g salt)

[0046] The method is as described above, using iodine and ferrous oxide as examples of additives. 80g of salt sample was collected, and the model calculated that the water content difference was 0.52%, so the iodine content of the salt sample was determined to be 27.8mg / kg. The actual iodine content of the salt was 27mg / kg, with an error of 2.96%;

[0047] Example 3 (100g salt)

[0048] The method is as described above, taking iodine and ferrous oxide as examples of additives. A 100g salt sample is collected, and the model calculates that the water content difference is 0.47%, so the iodine content of the salt sample is determined to be 25.13mg / kg. The actual iodine content of the salt is 26mg / kg, with an error of 3.34%;

[0049] In Examples 1-3 of the present invention, the results of determination of iodine content in each salt sample are shown in Table 1:

[0050] Table 1

[0051]

[0052] As can be seen from Table 1, the results in Examples 1-3 of the present invention are not much different, and the detection accuracy is within 5%.

[0053] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the contents disclosed herein. The present application is intended to cover any variations, uses or adaptations of the present application, which follow the general principles of the present application and include common knowledge or customary technical means in the art that are not disclosed in the present application.

[0054] It should be understood that the present application is not limited to the exact construction that has been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof.

Claims

1. A method for detecting salt additive content based on near infrared spectroscopy imaging, characterized in that: The following steps are involved: (1) Multi-level sampling: Collect salt samples before and after spraying additives at five depths on the production line, and spread the salt samples at different depths on the test platform; specifically, the salt samples at different depths are the salt samples on the surface of the production line, 1 / 4 of the surface, 2 / 4 of the surface, 3 / 4 of the surface and the bottom layer. The total weight of the salt sample must be ≥50g, and 10g is taken at each depth. The test platform is a black plastic bottom plate, and an active light source is set under darkroom conditions. The sampling positions are 10cm in front of and 10cm behind the spray head of the production line; (2) Near-infrared spectroscopy imaging measurement: Use near-infrared spectroscopy to image and measure salt at different depths on the platform; (3) Moisture estimation: The average moisture content of the whole salt before and after spraying the additive is estimated by near infrared spectroscopy; the method for estimating the overall moisture content of the salt is as follows: calibrate the spectral image of the salt sample, use the threshold method to delineate the region of interest (ROI) of the salt sample at different depths, extract the average spectrum of all pixels in the ROI, and use a three-dimensional data block to describe the extracted average spectrum data, wherein x and y represent the two-dimensional plane pixel information, and the third dimension λ is the wavelength information; the three-dimensional data block can display the difference information of the physical structure and chemical composition of the salt sample; use the continuous projection method SPA to extract 10 optimal characteristic wavelengths: 926nm, 951nm, 1003nm, 1343nm, 1360nm, 1385nm, 1392nm, 1434nm, 1441nm, 1700nm; substitute the spectral data corresponding to the characteristic wavelengths extracted by SPA into the least squares regression algorithm PLSR for modeling, and obtain the moisture content of the salt corresponding to each pixel in the hyperspectral image of the salt; the formula for the moisture content of the salt is: <h2 style=";text-align:left;direction:ltr">Y = (6.28-1.94X<h2 style=";text-align:left;direction:ltr"> 926nm <h2 style=";text-align:left;direction:ltr"> +49.74X<h2 style=";text-align:left;direction:ltr"> 951nm <h2 style=";text-align:left;direction:ltr"> +18.23X<h2 style=";text-align:left;direction:ltr"> 1003nm <h2 style=";text-align:left;direction:ltr"> +32.97X<h2 style=";text-align:left;direction:ltr"> 1343nm <h2 style=";text-align:left;direction:ltr"> +85.34X<h2 style=";text-align:left;direction:ltr"> 1360nm +44.61X 1385nm +87.39X 1392nm +32.10X 1434nm -28.87X 1441nm -61.43X 1700nm )×10 -5 Where Xi is the spectral value of a certain pixel in the spectral image at the characteristic wavelength of inm; Y is the salt water content value at the corresponding pixel; (4) Estimation of additive content: The additive content of salt is estimated by the average water content of salt samples mixed at different depths before and after spraying additives; The salt additive content is calculated according to the following formula: p=(Y-X)×k×10 6 ; Wherein, Y is the water content of the salt sample after spraying the additive, X is the water content of the salt sample before spraying the additive, k is the content of each additive in the additive solution, the unit is g / 100g, the additives include but are not limited to iodine and ferrous oxide, which are obtained according to the ratio of the sprayed additives, and p is the content of the additive in the salt sample, the unit is mg / kg.

2. The method for detecting salt additive content based on near infrared spectroscopy imaging according to claim 1, characterized in that: The active light source is a 50w halogen lamp.

3. The method for detecting salt additive content based on near infrared spectroscopy imaging according to claim 1, characterized in that: The step (2) is specifically as follows: the wavelength of the near-infrared spectrometer is 900-1700nm. During the measurement, the platform to be tested is in a dark room, the distance between the platform to be tested and the imaging lens is 25cm, and the salt samples of different depths are evenly stirred and spread into a circle with a thickness of ≥20mm and a diameter of ≥300mm.

Images