A time-series detection method for tea oil adulteration based on colorimetric sensor array

Abstract

The invention relates to the technical field of new material detection, and specifically discloses a tea oil adulteration timing detection method based on a colorimetric sensor array. The method obtains performance information of each timing detection device through a tea oil adulteration timing detection platform, thereby determining a performance stability index of each timing detection device, and comparing it with a performance stability threshold of the timing detection device, thereby determining whether to adjust the performance of each timing detection device, performing image shooting detection on an image acquisition device, analyzing an image quality index of the image acquisition device, and comparing it with an image quality threshold, thereby determining whether to optimize the image quality of the image acquisition device, and the tea oil adulteration timing detection platform performs timing detection on a sensor array image, analyzes to obtain a tea oil timing detection index, and compares it with a tea oil timing detection threshold, thereby determining whether the tea oil is adulterated, thereby improving the accuracy and reliability of tea oil adulteration detection, and realizing efficient and stable detection of tea oil adulteration.

Description

Technical Field

[0001] The invention relates to the technical field of new material detection, and in particular to a tea oil adulteration time sequence detection method based on a colorimetric sensor array. Background Art

[0002] Tea oil is a highly nutritious and high-value vegetable oil, which is favored by consumers for its unique taste and rich nutrients. However, among the traditional tea oil adulteration detection methods, most detection methods are complicated and time-consuming. The effectiveness of spectral technology in detecting tea oil adulteration is limited by the high equipment cost and professional operation skills required. Therefore, it is particularly important to develop a fast, accurate and simple tea oil adulteration detection method.

[0003] For example, the invention patent with publication number CN114563387A discloses a method for identifying tea oil adulteration based on Raman spectroscopy technology, using a quantitative identification method, and its specific steps include (1) collecting a calibration set, a validation set, and Raman spectroscopy data processing; (2) establishing a determination model PLS quantitative determination model; and (3) measuring the sample to be tested.

[0004] For example, the invention patent with announcement number CN103217393B announces a method for detecting adulteration of tea oil. First, adulteration detection models of refractive index, iodine value, saponification value and oleic acid content are established, and the type of adulteration is determined by sample detection. Then, the adulteration amount model is determined by full-wave scanning of the ultraviolet spectrum. By comparing the data, the determined adulteration amount is obtained.

[0005] Combined with the above-mentioned technical solutions, it is found that in the current process of tea oil adulteration detection, tea oil samples are usually tested directly through the test data, which will lead to inaccurate testing of tea oil samples and difficulty in accurately evaluating the true quality of tea oil, which greatly reduces the detection level and accuracy of tea oil samples. Summary of the invention

[0006] In view of the deficiencies in the prior art, the present invention provides a tea oil adulteration time series detection method based on a colorimetric sensor array, which can effectively solve the problems involved in the above-mentioned background technology.

[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions: a tea oil adulteration timing detection method based on a colorimetric sensor array, comprising: S1. a tea oil adulteration timing detection platform obtains the performance information of each timing detection device, thereby determining the performance stability index of each timing detection device, and comparing it with the performance stability threshold of the timing detection device preset in the tea oil detection library, thereby determining whether to adjust the performance of each timing detection device; S2. performing image shooting detection on the image acquisition device, thereby obtaining the image data of the image acquisition device, analyzing the image quality index of the image acquisition device, and comparing it with the image quality threshold preset in the tea oil detection library. If the image quality index is higher than or equal to the image quality threshold, the tea oil adulteration timing detection platform directly publishes the tea oil adulteration timing detection. Instruction, whereby each timing detection device and image acquisition device are used for timing detection of tea oil adulteration, if the image quality index is lower than the image quality threshold, the image quality of the image acquisition device is optimized, so that the image quality index corresponding to the image data after the image quality optimization is higher than or equal to the image quality threshold; S3. Based on the timing detection process of each timing detection device and image acquisition device for tea oil adulteration, the sensor array image taken by the image acquisition device is obtained, and the tea oil adulteration timing detection platform performs timing detection on the sensor array image to obtain timing detection data, and at the same time obtains the environmental data in the timing detection process, and combines the image quality index for comprehensive analysis to obtain the tea oil timing detection index, which is compared with the tea oil timing detection threshold preset in the tea oil detection library to determine whether the tea oil is adulterated.

[0008] As a further method, the performance information of each timing detection device specifically includes the average operating temperature of each temperature measurement point of each timing detection device during the equipment performance detection period, the maximum transmission power of each timing detection device during the equipment performance detection period, the minimum transmission power of each timing detection device during the equipment performance detection period, the maximum noise intensity of each timing detection device during the equipment performance detection period, and the energy consumption ratio of each timing detection device during the equipment performance detection period.

[0009] As a further method, the performance stability index of each timing detection device is specifically analyzed as follows:

[0010] The maximum value and the minimum value of the transmission power of each timing detection device in the device performance detection period are subjected to difference processing to obtain the power fluctuation index of each timing detection device in the device performance detection period.

[0011] The performance stability index of each timing detection device is obtained by comprehensive analysis of the temperature fluctuation, power fluctuation index, maximum noise intensity and energy consumption ratio of each timing detection device during the equipment performance detection period.

[0012] As a further method, the determination of whether to adjust the performance of each timing detection device is carried out in the following specific determination process:

[0013] If the performance stability index of a certain timing detection device is less than the performance stability threshold of the timing detection device, the performance of the timing detection device is adjusted.

[0014] If the performance stability index of a timing detection device is greater than or equal to the performance stability threshold of the timing detection device, there is no need to adjust the performance of the timing detection device.

[0015] As a further method, the image quality index of the image acquisition device is specifically analyzed as follows:

[0016] The pixel value of each sample pixel point of the image acquisition device within the image detection period is subjected to variance processing, thereby obtaining the average image sharpness of the image acquisition device within the image detection period.

[0017] The maximum image brightness and the minimum image brightness of the image acquisition device within the image detection cycle are subjected to difference processing, and the result is marked as the image brightness contrast of the image acquisition device within the image detection cycle.

[0018] Obtain the maximum noise intensity and power fluctuation index of the image acquisition device during the image detection cycle.

[0019] The image quality index of the image acquisition device is comprehensively analyzed through the image resolution, image average sharpness, image grayscale value, image brightness contrast, image signal-to-noise ratio, maximum noise intensity and power fluctuation index of the image acquisition device during the image detection cycle.

[0020] As a further method, the image quality of the image acquisition device is optimized, and the specific optimization process may be denoising the captured image, adjusting the image grayscale value, and enhancing the image edge information.

[0021] The denoising process is performed on the captured image, specifically, denoising process is performed on the captured image of the image acquisition device.

[0022] The image grayscale value adjustment specifically adjusts the image grayscale value of the captured image to increase the difference between different grayscale levels in the image.

[0023] The image edge information enhancement specifically performs image edge information enhancement on the captured image to highlight edge features in the image.

[0024] As a further method, the tea oil time series detection index, the specific analysis process is:

[0025] The operation influence set corresponding to each timing detection device is obtained, specifically, the maximum operating temperature of each timing detection device within the tea oil adulteration detection cycle.

[0026] The sequence numbers of the time series characteristics of the tea oil samples are compared and verified one by one with the initial sequence numbers of the time series characteristics preset in the tea oil detection library, thereby obtaining the deviation rate of the sequence numbers of the time series characteristics of the tea oil samples.

[0027] The tea oil time series detection index is obtained by comprehensive analysis through the RGB color feature difference of tea oil samples, the deviation rate of the sequence number of the time series feature of tea oil samples, the average humidity of the detection room to which the tea oil samples belong during the tea oil adulteration detection period, the maximum concentration of carbon dioxide, the maximum concentration of sulfur dioxide, the maximum light intensity, the maximum concentration of floating dust, the image quality index, and the maximum operating temperature of each time series detection equipment during the tea oil adulteration detection period. The specific analysis method is as follows:

[0028] ;

[0029] In the formula, Indicates tea oil time series detection index, Indicates the RGB color feature difference of tea oil sample, Indicates the deviation rate of the sequence number to which the time series characteristics of tea oil samples belong, It indicates the average humidity of the testing room where the tea oil sample belongs during the tea oil adulteration testing period. Indicates the average humidity reference value preset in the tea oil detection library. It indicates the maximum concentration of carbon dioxide in the testing room where the tea oil sample belongs during the tea oil adulteration testing cycle. It indicates the maximum concentration of sulfur dioxide in the testing room where the tea oil sample belongs during the tea oil adulteration testing period. It indicates the maximum light intensity of the testing room where the tea oil sample belongs during the tea oil adulteration testing cycle. Indicates the maximum light intensity reference value preset in the tea oil detection library. It indicates the maximum concentration of floating dust in the testing room where the tea oil sample belongs during the tea oil adulteration testing period. Represents the image quality index of the image acquisition device, It represents the maximum operating temperature of the Nth timing detection device during the tea oil adulteration detection cycle. Represents the temporal detection weight index of the RGB color feature difference preset in the tea oil detection library, The time series detection weight index representing the deviation rate of the sequence number of the preset time series feature in the tea oil detection library, Indicates the time series detection weight index of the average humidity preset in the tea oil detection library, Indicates the time series detection weight index of the maximum concentration of carbon dioxide preset in the tea oil detection library, Indicates the time series detection weight index of the maximum sulfur dioxide concentration preset in the tea oil detection library, Represents the time series detection weight index of the maximum light intensity preset in the tea oil detection library, Indicates the time series detection weight index of the maximum concentration of floating dust preset in the tea oil detection library, represents the temporal detection weight index of the image quality index preset in the tea oil detection library, Indicates the timing detection weight index of the maximum operating temperature preset in the tea oil detection library, M, M is the total number of timing detection devices.

[0030] As a further method, the specific judgment process of judging whether the tea oil is adulterated is as follows:

[0031] If the tea oil timing detection index is greater than or equal to the tea oil timing detection threshold preset in the tea oil detection library, it means that the tea oil is not adulterated.

[0032] If the tea oil timing detection index is less than the tea oil timing detection threshold preset in the tea oil detection library, it indicates that the tea oil is adulterated. The tea oil timing detection index is subtracted from the tea oil timing detection threshold to obtain the tea oil timing detection deviation value, and the value is matched with the tea oil adulteration data set corresponding to each tea oil timing detection deviation value interval preset in the tea oil detection library to obtain the tea oil adulteration data set of the tea oil sample. Finally, the type and concentration of tea oil adulteration are obtained through the tea oil adulteration data set of the tea oil sample.

[0033] Compared with the prior art, the embodiments of the present invention have at least the following advantages or beneficial effects:

[0034] (1) The present invention provides a tea oil adulteration timing detection method based on a colorimetric sensor array. First, a tea oil adulteration timing detection platform obtains performance information of each timing detection device, thereby determining the performance stability index of each timing detection device, and finally determining whether to adjust the performance of each timing detection device. At the same time, an image capture device is subjected to image shooting detection, so that the sensor array image can be accurately captured and has high definition before the tea oil adulteration timing detection platform directly issues an instruction for tea oil adulteration timing detection. Based on the timing detection process of each timing detection device and the image acquisition device for tea oil adulteration, the sensor array image captured by the image acquisition device is obtained, and the tea oil timing detection index is obtained by analysis. The tea oil timing detection index is compared with the tea oil timing detection threshold preset in the tea oil detection library to determine whether the tea oil is adulterated. Whether the tea oil is adulterated can be predicted, thereby improving the accuracy and reliability of tea oil adulteration detection.

[0035] (2) The tea oil adulteration timing detection platform obtains the performance information of each timing detection device, thereby determining the performance stability index of each timing detection device, and compares it with the performance stability threshold of the timing detection device preset in the tea oil detection library, thereby determining whether to adjust the performance of each timing detection device. This analysis process ensures that the timing detection equipment can operate under good conditions of stable operation, thereby improving the accuracy of tea oil adulteration detection.

[0036] (3) Perform image capture detection on the image acquisition device to obtain image data of the image acquisition device, analyze the image quality index of the image acquisition device, and compare it with the image quality threshold preset in the tea oil detection library. If the image quality index is higher than or equal to the image quality threshold, the tea oil adulteration timing detection platform directly issues an instruction for tea oil adulteration timing detection, thereby each timing detection device and image acquisition device are used for tea oil adulteration timing detection. If the image quality index is lower than the image quality threshold, the image quality of the image acquisition device is optimized so that the image quality index corresponding to the image data after image quality optimization is higher than or equal to the image quality threshold, which can ensure the accurate acquisition of subsequent sensor array images and provide an accurate image quality basis for tea oil adulteration detection.

[0037] (4) Based on the timing detection process of tea oil adulteration using various timing detection devices and image acquisition devices, the sensor array image taken by the image acquisition device is obtained. The tea oil adulteration timing detection platform performs timing detection on the sensor array image to obtain timing detection data. At the same time, the environmental data in the timing detection process is obtained, and the tea oil timing detection index is obtained by comprehensive analysis combined with the image quality index. The tea oil timing detection index is compared with the tea oil timing detection threshold preset in the tea oil detection library to determine whether the tea oil is adulterated, thereby improving the accuracy of tea oil adulteration detection and achieving efficient and stable detection of tea oil adulteration. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] The present invention is further described using the accompanying drawings, but the embodiments in the accompanying drawings do not constitute any limitation to the present invention. A person skilled in the art can obtain other drawings based on the following drawings without creative work.

[0039] Figure 1 It is a schematic diagram of the method flow of the present invention.

[0040] Figure 2 It is a schematic diagram of the tea oil adulteration detection structure of the present invention.

[0041] Figure 3 It is a structural schematic diagram of the box body of the present invention.

[0042] Figure 4 It is a schematic structural diagram of the heat dissipation device of the present invention.

[0043] Figure 5 It is a structural schematic diagram of the tea oil adulteration time series detection platform of the present invention.

[0044] Figure 6 It is a structural schematic diagram of the image acquisition device of the present invention.

[0045] Figure 7 It is a schematic diagram of the structure of the detection room to which the tea oil sample belongs according to the present invention.

[0046] Figure numerals: 1. heat dissipation device; 2. tea oil adulteration timing detection platform; 3. image acquisition equipment; 4. tea oil sample detection room; 5. box; 6. heating platform; 7. side cover; 8. upper box top cover; 9. upper box; 10. positioning partition; 11. back plate one; 12. lower box two top cover; 13. back plate two; 14. lower box two; 15. lower box one; 16. cooling fan; 17. dust net; 18. Raspberry Pi (hardware); 19. display screen; 20. camera mounting plate; 21. camera; 22. bolts; 23. LED ring light source; 24. colorimetric sensor array; 25. colorimetric sensor array bracket; 26. square glass cover; 27. sample chamber. DETAILED DESCRIPTION

[0047] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0048] Reference Figure 1 As shown, the present invention provides a tea oil adulteration timing detection method based on a colorimetric sensor array, comprising: S1. The tea oil adulteration timing detection platform obtains the performance information of each timing detection device, thereby determining the performance stability index of each timing detection device, and compares it with the timing detection device performance stability threshold preset in the tea oil detection library, thereby determining whether to adjust the performance of each timing detection device.

[0049] In one embodiment, the above-mentioned tea oil adulteration timing detection platform refers to a platform for performing timing detection of adulteration of tea oil samples. This tea oil adulteration timing detection platform can be connected through the Internet of Things, thereby receiving instructions for tea oil adulteration timing detection through the Internet of Things, and finally performing adulteration timing detection operations on tea oil samples; the tea oil adulteration timing detection platform is internally provided with a SIEM (security information and event management) system and image processing technology. Through the built-in image processing technology, the image data of the image acquisition device within the image detection cycle can be obtained, and the feature difference image data can be extracted, and the extracted feature difference image data can be reduced in dimension, and then imported into the linear SVM classification model for classification and recognition, and the RGB color features of 16 color blocks in each minute of the image can be automatically extracted. The RGB difference feature vector generated by processing and analyzing the RGB color feature difference of the tea oil sample can also be used to obtain the sequence number to which the time series feature of the tea oil sample belongs; the above-mentioned timing detection devices specifically refer to Figure 2 The heat dissipation fan 16 in the heat dissipation device 1 and the electric heating plate in the heating platform 6 are shown.

[0050] Specifically, the performance information of each timing detection device includes the average operating temperature of each temperature measurement point of each timing detection device during the equipment performance detection period, the maximum transmission power of each timing detection device during the equipment performance detection period, the minimum transmission power of each timing detection device during the equipment performance detection period, the maximum noise intensity of each timing detection device during the equipment performance detection period, and the energy consumption ratio of each timing detection device during the equipment performance detection period.

[0051] It should be explained that the above-mentioned equipment performance detection period can be obtained through the SIEM system built into the tea oil adulteration timing detection platform. Specifically, it refers to the time period from the start of the performance detection of each timing detection device to the end of the detection, which is recorded as the equipment performance detection period. The setting of the specific detection duration will present different equipment performance detection periods based on multiple factors such as the differences in timing detection equipment and their respective unique usage scenarios.

[0052] It should be explained that the temperature measurement points belonging to the above-mentioned timing detection devices refer to several points for measuring the temperature of the timing detection devices, and one temperature measurement point belonging to a timing detection device corresponds to a temperature sensor; the average operating temperature of each temperature measurement point belonging to the above-mentioned timing detection devices during the equipment performance detection period specifically refers to the result obtained by monitoring the temperature of each temperature measurement point belonging to each timing detection device during the equipment performance detection period and averaging these temperature values, which can be obtained by using several high-sensitivity temperature sensors built into each timing detection device, wherein the temperature sensor can monitor multiple operating temperature values ​​of each temperature measurement point during the equipment performance detection period in real time, and obtain the average operating temperature through averaging processing; the maximum transmission power and minimum transmission power of each timing detection device during the equipment performance detection period specifically refer to the highest transmission power and the lowest transmission power that each timing detection device can achieve during the equipment performance detection period, which can be obtained by using the high-sensitivity power meter built into each timing detection device. It is obtained that the power meter can monitor multiple transmission power values ​​during the equipment performance detection period in real time, and through value size comparison, select the maximum value as the maximum transmission power value, and select the minimum value as the minimum transmission power value; the maximum noise intensity of the above-mentioned timing detection devices during the equipment performance detection period, specifically refers to the highest intensity reached by the noise generated by each timing detection device during the equipment performance detection period, which can be obtained by using the high-sensitivity noise tester built into each timing detection device, wherein the noise tester can monitor multiple noise intensity values ​​during the equipment performance detection period in real time, and through value size comparison, select the maximum value as the maximum noise intensity value; the energy consumption ratio of the above-mentioned timing detection devices during the equipment performance detection period, specifically refers to the ratio of the energy consumed by each timing detection device during operation to its output power during the equipment performance detection period, which can be obtained by using the high-sensitivity power consumption analyzer built into each timing detection device to obtain the total output power and total energy consumption of each timing detection device during the equipment performance detection period, and the energy consumption ratio is obtained by comparing the two.

[0053] Furthermore, the specific analysis process of the performance stability index of each timing detection device is as follows:

[0054] The average operating temperature of each temperature measurement point of each timing detection device during the equipment performance detection period is processed with standard deviation, thereby obtaining the temperature fluctuation of each timing detection device during the equipment performance detection period.

[0055] The maximum value and the minimum value of the transmission power of each timing detection device in the device performance detection period are subjected to difference processing to obtain the power fluctuation index of each timing detection device in the device performance detection period.

[0056] The performance stability index of each timing detection device is obtained by comprehensive analysis through the temperature fluctuation, power fluctuation index, maximum noise intensity and energy consumption ratio of each timing detection device during the device performance detection period. The specific acquisition method is as follows:

[0057]

[0058] In the formula, Indicates the temperature fluctuation of the Nth timing detection device during the device performance detection period. Indicates the power fluctuation index of the Nth timing detection device during the device performance detection period. Indicates the maximum noise intensity of the Nth timing detection device during the device performance detection period. Indicates the energy consumption ratio of the Nth timing detection device during the device performance detection period. It represents the performance stability index of the Nth timing detection device. If the temperature fluctuation of the timing detection device is large, it will affect the electrical and mechanical performance of the device, thereby causing the power fluctuation index to increase. This means that the power output of the device is unstable, which will cause increased wear or vibration of mechanical parts, thereby generating greater noise and increasing the maximum noise intensity. When the temperature fluctuation, power fluctuation index and maximum noise intensity of the device are all high, it means that the energy efficiency level of the device is low, that is, the energy consumption is relatively low, indicating that the device performance is unstable and the performance stability indexes of each timing detection device are low.

[0059] It represents the device performance impact weight of the temperature fluctuation preset in the tea oil detection library, and represents the proportion of the temperature fluctuation of each timing detection device in the device performance detection period to the performance stability index of each timing detection device. The tea oil detection library stores the corresponding relationship between the device performance impact weights. For example, the real-time temperature fluctuation of each timing detection device in the device performance detection period and the performance stability index of each timing detection device can be input into the tea oil detection library. The tea oil detection library automatically matches the device performance impact weight corresponding to the temperature fluctuation according to the preset corresponding relationship between the device performance impact weights.

[0060] It represents the device performance impact weight of the power fluctuation index preset in the tea oil detection library, and represents the proportion of the power fluctuation index of each timing detection device in the device performance detection period to the performance stability index of each timing detection device. The tea oil detection library stores the corresponding relationship between the device performance impact weights. For example, the real-time power fluctuation index of each timing detection device in the device performance detection period and the performance stability index of each timing detection device can be input into the tea oil detection library. The tea oil detection library automatically matches the device performance impact weight corresponding to the power fluctuation index according to the preset corresponding relationship between the device performance impact weights.

[0061] The device performance impact weight of the preset maximum noise intensity value in the tea oil detection library is represented, which represents the proportion of the maximum noise intensity of each timing detection device in the device performance detection period to the performance stability index of each timing detection device. The tea oil detection library stores the corresponding relationship between device performance impact weights. For example, the real-time maximum noise intensity of each timing detection device in the device performance detection period and the performance stability index of each timing detection device can be input into the tea oil detection library. The tea oil detection library automatically matches the device performance impact weight corresponding to the maximum noise intensity value according to the preset corresponding relationship between device performance impact weights.

[0062] The equipment performance impact weight of the energy consumption ratio preset in the tea oil detection library is represented, which represents the proportion of the energy consumption ratio of each timing detection device in the equipment performance detection period to the performance stability index of each timing detection device. The tea oil detection library stores the corresponding relationship between the equipment performance impact weights. For example, the real-time energy consumption ratio of each timing detection device in the equipment performance detection period and the performance stability index of each timing detection device can be input into the tea oil detection library. The tea oil detection library automatically matches the equipment performance impact weight corresponding to the energy consumption ratio according to the preset corresponding relationship between the equipment performance impact weights.

[0063] N is the number of each timing detection device, M, M is the total number of timing detection devices.

[0064] Specifically, the determination of whether to adjust the performance of each timing detection device is carried out in the following specific determination process:

[0065] If the performance stability index of a timing detection device is less than the performance stability threshold of the timing detection device, the performance of the timing detection device is adjusted; the performance stability threshold of the timing detection device represents the critical value used to determine whether the performance stability index of the timing detection device reaches the performance stability standard during the device performance detection process, and is used to measure the performance stability of the timing detection device during operation, and help determine whether the timing detection device needs to be adjusted; if the performance stability index of the above-mentioned timing detection device is less than the performance stability threshold of the timing detection device, it means that the performance of the timing detection device is unstable, and the timing detection device corresponding to the performance stability index of the timing detection device needs to be adjusted.

[0066] If the performance stability index of a timing detection device is greater than or equal to the performance stability threshold of the timing detection device, there is no need to adjust the performance of the timing detection device; if the performance stability index of a timing detection device is greater than or equal to the performance stability threshold of the timing detection device, it means that the performance of the timing detection device is stable, and there is no need to adjust the timing detection device corresponding to the performance stability index of the timing detection device.

[0067] It should be explained that the above-mentioned adjustment of the timing detection equipment corresponding to the performance stability index of the timing detection equipment specifically refers to cleaning the fan blades and radiator of the cooling fan, increasing the speed of the cooling fan, increasing the heating power of the electric heating plate, and cleaning the surface of the electric heating plate.

[0068] S2. Perform image shooting detection on the image acquisition device to obtain image data of the image acquisition device, analyze the image quality index of the image acquisition device, and compare it with the image quality threshold preset in the tea oil detection library. If the image quality index is higher than or equal to the image quality threshold, the tea oil adulteration timing detection platform directly issues an instruction for tea oil adulteration timing detection, thereby each timing detection device and image acquisition device are used for tea oil adulteration timing detection. If the image quality index is lower than the image quality threshold, the image acquisition device is optimized for image quality, so that the image quality index corresponding to the image data after image quality optimization is higher than or equal to the image quality threshold.

[0069] Specifically, the image data of the image acquisition device specifically includes the image resolution of the image acquisition device within the image detection period, the pixel value of each sample pixel point, the image gray value, the maximum brightness of the image, and the minimum brightness of the image;

[0070] The pixel value of each sample pixel is processed with variance to obtain the image signal variance.

[0071] The pixel value of each sample pixel is processed by standard deviation to obtain the image noise standard deviation.

[0072] The image signal-to-noise ratio is obtained by ratioing the image signal variance to the image noise standard deviation.

[0073] It should be explained that the above-mentioned image detection cycle can be obtained through the SIEM system built into the tea oil adulteration timing detection platform. Specifically, it refers to the time period from the start of image detection of the image acquisition device to the end of detection, which is recorded as the image detection cycle. The setting of the specific detection cycle length will present different image detection cycles based on multiple factors such as the differences in image acquisition devices and their respective unique usage scenarios.

[0074] It should be explained that the image resolution of the above-mentioned image acquisition device within the image detection period, the pixel value of each sample pixel point, the image grayscale value, the maximum image brightness and the minimum image brightness can all be obtained by reading the image taken by the image acquisition device within the image detection period through image editing software; the image signal-to-noise ratio of the above-mentioned image acquisition device within the image detection period can be obtained through the built-in image processing technology of the tea oil adulteration timing detection platform.

[0075] Among them, the image resolution of the image acquisition device within the image detection period specifically refers to the number of pixels in the length and width directions of the imaging area when the image acquisition device captures the image within the image detection period, which is usually expressed in the form of "horizontal pixel number × vertical pixel number"; the pixel value of each sample pixel point of the image acquisition device within the image detection period specifically refers to the color or brightness information of each minimum unit (i.e., pixel point) in the image taken by the image acquisition device within the image detection period, wherein the sample pixel points are randomly selected from different positions of the image, and the number of sample pixel points will also be different depending on the detail information in the image; the image grayscale value of the image acquisition device within the image detection period specifically refers to the grayscale value of the image taken by the image acquisition device within the image detection period. The grayscale level of each pixel in the captured image represents the brightness information of the pixel; the image signal-to-noise ratio of the image acquisition device within the image detection period specifically refers to the ratio of the average value of the pixel value of each sample pixel point in the image captured by the image acquisition device within the image detection period (image signal variance) to the standard deviation of the pixel value of each sample pixel point (image noise standard deviation); the maximum image brightness of the image acquisition device within the image detection period specifically refers to the highest brightness value that can be achieved by all pixels in the image captured by the image acquisition device within the image detection period; the minimum image brightness of the image acquisition device within the image detection period specifically refers to the lowest brightness value that can be achieved by all pixels in the image captured by the image acquisition device within the image detection period.

[0076] Furthermore, the image quality index of the image acquisition device is specifically analyzed as follows:

[0077] The pixel value of each sample pixel point of the image acquisition device within the image detection period is subjected to variance processing, thereby obtaining the average image sharpness of the image acquisition device within the image detection period.

[0078] The maximum image brightness and the minimum image brightness of the image acquisition device within the image detection cycle are subjected to difference processing, and the result is marked as the image brightness contrast of the image acquisition device within the image detection cycle.

[0079] Obtain the maximum noise intensity and power fluctuation index of the image acquisition device during the image detection cycle.

[0080] It should be explained that the maximum noise intensity and power fluctuation index of the above-mentioned image acquisition device during the image detection period are obtained in the same way as the maximum noise intensity of the above-mentioned timing detection devices during the device performance detection period and the power fluctuation index of each timing detection device during the device performance detection period.

[0081] The image quality index of the image acquisition device is comprehensively analyzed through the image resolution, image average sharpness, image gray value, image brightness contrast, image signal-to-noise ratio, maximum noise intensity and power fluctuation index of the image acquisition device during the image detection cycle. The specific analysis method is as follows:

[0082]

[0083] In the formula, It indicates the image resolution of the image acquisition device during the image detection cycle. AIS indicates the average image sharpness of the image acquisition device during the image detection cycle. Represents the image grayscale value of the image acquisition device during the image detection cycle, Represents the image grayscale reference value preset in the tea oil detection library, It represents the image brightness contrast of the image acquisition device during the image detection cycle, and SNR represents the image signal-to-noise ratio of the image acquisition device during the image detection cycle. Indicates the maximum noise intensity of the image acquisition device during the image detection cycle. It represents the power fluctuation index of the image acquisition device during the image detection period, and e is a natural constant.

[0084] The image grayscale reference value specifically refers to the optimal adaptation value of the image grayscale value set in the tea oil detection library, which aims to ensure that the image grayscale level reaches the optimal state, thereby presenting the optimal image quality, and can more effectively capture and present the detailed information in the tea oil adulteration timing detection image, thereby improving the accuracy and reliability of the detection.

[0085] IQIA represents the image quality index of the image acquisition device. When the power fluctuation index is small, the performance of the image acquisition device is more stable, which will result in a higher image resolution of the image captured within the image detection cycle, which means that there are more pixels in the image, the average image sharpness is greater, the image detail expression is stronger, and the image quality index of the image acquisition device is higher; when the image grayscale value is small, the overall brightness of the image is dim, because the brightness of most pixels is low, resulting in a lower brightness contrast of the image, that is, the brightness difference between the brightest and darkest parts of the image is not large, the image is not clear, and the image quality index of the image acquisition device is reduced; when the image grayscale value is large, the overall brightness of the image is too bright and the picture is overexposed, resulting in a lower brightness contrast of the image, that is, the brightness difference between the brightest and darkest parts of the image is not large, the image is not clear, and the image quality index of the image acquisition device is reduced; when the image signal-to-noise ratio is large, it means that the noise interference in the image is small, the maximum noise intensity is smaller, the image is clearer, and the image quality index of the image acquisition device is higher.

[0086] The image quality impact factor representing the image resolution preset in the tea oil detection library represents the proportion of the image resolution of the image acquisition device within the image detection cycle to the image quality index of the image acquisition device. The tea oil detection library stores the corresponding relationship between the image quality impact factors. For example, the real-time image resolution of the image acquisition device within the image detection cycle and the image quality index of the image acquisition device can be input into the tea oil detection library. The tea oil detection library automatically matches the image quality impact factor corresponding to the image resolution according to the preset corresponding relationship between the image quality impact factors.

[0087] The image quality impact factor representing the average image sharpness preset in the tea oil detection library represents the proportion of the average image sharpness of the image acquisition device within the image detection cycle to the image quality index of the image acquisition device. The tea oil detection library stores the corresponding relationship between the image quality impact factors. For example, the real-time average image sharpness of the image acquisition device within the image detection cycle and the image quality index of the image acquisition device can be input into the tea oil detection library. The tea oil detection library automatically matches the image quality impact factor corresponding to the average image sharpness according to the preset corresponding relationship between the image quality impact factors.

[0088] The image quality correction factor representing the image grayscale value preset in the tea oil detection library represents the proportion of the image grayscale value of the image acquisition device within the image detection cycle to the image quality index of the image acquisition device. The tea oil detection library stores the corresponding relationship of the image quality correction factor. For example, the real-time image grayscale value of the image acquisition device within the image detection cycle and the image quality index of the image acquisition device can be input into the tea oil detection library. The tea oil detection library automatically matches the image quality correction factor corresponding to the image grayscale value according to the preset corresponding relationship of the image quality correction factor.

[0089] The image quality influencing factor representing the image brightness contrast preset in the tea oil detection library represents the proportion of the image brightness contrast of the image acquisition device within the image detection cycle to the image quality index of the image acquisition device. The tea oil detection library stores the corresponding relationship between the image quality influencing factors. For example, the image brightness contrast of the real-time image acquisition device within the image detection cycle and the image quality index of the image acquisition device can be input into the tea oil detection library. The tea oil detection library automatically matches the image quality influencing factor corresponding to the image brightness contrast according to the preset corresponding relationship between the image quality influencing factors.

[0090] The image quality influencing factor representing the image signal-to-noise ratio preset in the tea oil detection library represents the proportion of the image signal-to-noise ratio of the image acquisition device within the image detection cycle to the image quality index of the image acquisition device. The tea oil detection library stores the corresponding relationship between the image quality influencing factors. For example, the real-time image signal-to-noise ratio of the image acquisition device within the image detection cycle and the image quality index of the image acquisition device can be input into the tea oil detection library. The tea oil detection library automatically matches the image quality influencing factor corresponding to the image signal-to-noise ratio according to the preset corresponding relationship between the image quality influencing factors.

[0091] The image quality impact factor representing the maximum noise intensity value preset in the tea oil detection library represents the proportion of the maximum noise intensity value of the image acquisition device during the image detection cycle to the image quality index of the image acquisition device. The tea oil detection library stores the corresponding relationship between the image quality impact factors. For example, the maximum noise intensity value of the real-time image acquisition device during the image detection cycle and the image quality index of the image acquisition device can be input into the tea oil detection library. The tea oil detection library automatically matches the image quality impact factor corresponding to the maximum noise intensity value according to the preset corresponding relationship between the image quality impact factors.

[0092] The image quality impact factor representing the power fluctuation index preset in the tea oil detection library represents the proportion of the power fluctuation index of the image acquisition device during the image detection cycle to the image quality index of the image acquisition device. The tea oil detection library stores the corresponding relationship between the image quality impact factors. For example, the power fluctuation index of the real-time image acquisition device during the image detection cycle and the image quality index of the image acquisition device can be input into the tea oil detection library. The tea oil detection library automatically matches the image quality impact factor corresponding to the power fluctuation index according to the preset corresponding relationship between the image quality impact factors.

[0093] Specifically, the image quality of the image acquisition device is optimized, and the specific optimization process may be denoising the captured image, adjusting the image grayscale value, and enhancing the image edge information.

[0094] The denoising process is performed on the captured image, specifically, denoising process is performed on the captured image of the image acquisition device.

[0095] The image grayscale value adjustment specifically adjusts the image grayscale value of the captured image to increase the difference between different grayscale levels in the image.

[0096] The image edge information enhancement specifically performs image edge information enhancement on the captured image to highlight edge features in the image.

[0097] It needs to be explained that the above-mentioned denoising processing of the captured image can effectively remove random noise and interference in the image, improve the clarity of the image, thereby increasing the image signal-to-noise ratio and improving the image quality; adjusting the image grayscale value of the captured image can increase the difference between different grayscale levels in the image, enhance the contrast of the image, increase the image brightness contrast, and greatly improve the image quality; enhancing the image edge information of the captured image can highlight the edge features in the image, make the outline of the object clearer, and increase the average sharpness of the image, thereby improving the image quality.

[0098] S3. Based on the timing detection process of tea oil adulteration using various timing detection devices and image acquisition devices, the sensor array image taken by the image acquisition device is obtained. The tea oil adulteration timing detection platform then performs timing detection on the sensor array image to obtain timing detection data. At the same time, the environmental data during the timing detection process is obtained, and the tea oil timing detection index is obtained by comprehensive analysis combined with the image quality index. The index is then compared with the tea oil timing detection threshold preset in the tea oil detection library to determine whether the tea oil is adulterated.

[0099] like Figure 2-Figure 7 As shown, the tea oil adulteration detection structure includes a heat dissipation device 1, a tea oil adulteration timing detection platform 2, an image acquisition device 3, a detection chamber 4 to which a tea oil sample belongs, a box 5 and a heating platform 6; the box 5 includes a side cover 7, an upper box top cover 8, an upper box 9, a positioning partition 10, a rear plate 11, a lower box second top cover 12, a rear plate 2 13, a lower box second 14 and a lower box 15; the heat dissipation device 1 includes a cooling fan 16 and a dustproof net 17; the tea oil adulteration timing detection platform 2 includes a Raspberry Pi (hardware) 18 and a display screen 19; the image acquisition device 3 includes a camera mounting plate 20, a camera 21, a bolt 22 and an LED ring light source 23 for acquiring images of a colorimetric sensor array; the detection chamber 4 to which a tea oil sample belongs includes a colorimetric sensor array 24, a colorimetric sensor array bracket 25, a square glass cover plate 26 and a sample chamber 27; the heating platform 6 uses an electric heating plate to provide a heat source.

[0100] It should be explained that the above-mentioned timing detection equipment and image acquisition equipment are used for the timing detection process of tea oil adulteration. Specifically, a color-sensitive material with high sensitivity to volatile gases in tea oil is first selected, mainly including porphyrin compounds and pH indicators, the porphyrin compounds are dissolved in dichloromethane, and the pH indicator is dissolved in anhydrous ethanol to prepare a solution with a concentration of 2 mg / mL; then, a microcapillary is used to titrate the color-sensitive material onto a 3 cm × 3 cm polyvinylidene fluoride (PVDF) membrane to construct a 4 × 4 colorimetric sensor array 24. After the preparation is completed, the colorimetric sensor array is placed in a fume hood for 20 minutes to stabilize its performance, and is stored in a sealed bag; then, a magnetic stirrer is used to fully stir the mixed oil sample, and after ensuring that the oil sample is uniform, it is placed in the sample chamber 27 in the detection chamber 4 to which the tea oil sample belongs; in the Raspberry Pi (hardware) 18 control program, the cooling fan is first started. 16 to keep the device cool, then activate the heating platform 6 and set its temperature to 100°C, and finally start the LED ring light source 23 to provide necessary lighting for the experiment; after the temperature of the heating platform reaches 100°C, open the square glass cover 26 of the detection chamber 4 to which the tea oil sample belongs, and accurately place the sensor array in the center of the sensor array bracket, ensuring that the color-sensitive material faces upward and the back faces the oil sample, and then close the cover, seal the culture dish containing the oil sample with the detection chamber to which the tea oil sample belongs, and place them in the positioning partition above the heating platform; ensure that the detection chamber to which the tea oil sample belongs is in direct contact with the heating platform and is located directly below the camera 21, set the shooting area (ROI) in the Raspberry Pi (hardware) control program, and use a computer-controlled camera to perform timed imaging of the sensor array, capture an image every one minute, and continuously record data at 20 time points to obtain a complete time series of sensor response;After the volatile gas of the oil sample reacts with the sensor array until the color is stable, the final stable image is input into the computer, and the image of each minute is processed by using visual technology to identify the RGB values ​​of the 16 color-sensitive points in the image of each minute. Then, the RGB color feature difference between the image of each minute and the initial image is calculated to generate the RGB difference feature vector of each minute. Finally, a 960-dimensional time series feature (16 color-sensitive points × 3 color features × 20 minutes) can be obtained. The generated 960-dimensional time series feature vector is used as input and input into the linear SVM classification model. The generated feature vector is combined with the corresponding label (adulterated or not adulterated) into a training data set, and the selected Select appropriate kernel functions (such as linear kernel, RBF kernel, etc.), and adjust parameters such as penalty coefficient C to optimize the performance of the linear SVM classification model. Use the training data set to train the linear SVM classification model so that it can learn the characteristic differences between adulterated oil samples and non-adulterated oil samples. Use the trained linear SVM classification model to predict the new oil sample feature vector, and the output result is 0 (indicating that the oil sample is not adulterated) or 1 (indicating that the oil sample is adulterated). According to the output result, judge whether the oil sample is adulterated; construct the feature vector, design the structure of the MLP model, including the input layer, hidden layer and output layer. The number of neurons in the input layer should match the dimension of the feature vector (i.e., 960), and the number of neurons in the hidden layer should match the dimension of the feature vector (i.e., 960). The number of neurons in each layer can be adjusted according to actual conditions to optimize the performance of the model. The output layer uses one neuron, and the output value is 0 (indicating that the oil sample is not adulterated) or 1 (indicating that the oil sample is adulterated). The MLP model is trained using the above training data set, and the weights and bias parameters of the model are randomly initialized. The feature vectors in the training data set are input into the MLP model. The output of each layer and the final prediction result are calculated through forward propagation. The loss value is calculated based on the prediction result and the actual label. The cross entropy loss function is used to calculate the loss. The gradient of the loss function with respect to the model parameters is calculated by the chain rule, and these gradients are back-propagated to each layer. The batch gradient descent method is used according to Gradient update model weights and bias parameters. During the training process, K-fold cross validation can be used to randomly divide the training data set into K equal (or almost equal) parts or "folds". For each of the K folds, use it as a validation set, and the remaining K-1 folds as training sets. Train and validate each fold, record the accuracy, and calculate the average accuracy of all folds as the final evaluation of the model to evaluate the performance of the model, and adjust the model structure (adjust the number of hidden layers and the number of neurons in each layer) or hyperparameters (learning rate) to optimize the results. Use the trained MLP model to predict the new oil sample feature vector, and judge whether the oil sample is adulterated based on the value of the output layer (0 or 1);

[0101] The collection and processing of time series data of color-sensitive sensors enables efficient and stable detection of adulterated tea oil based on a visual colorimetric sensor array. With the assistance of Raspberry Pi (hardware), its advanced image processing program automatically records and analyzes the color changes of the sensor array over time under the action of oil sample gas. This integrated control solution simplifies the operating process, improves the accuracy and reproducibility of detection, and can perform complex detection tasks more easily.

[0102] It should be explained that the above-mentioned color-sensitive dyes mainly include 9 porphyrin compounds (tetrakis-p-chlorophenylporphyrin copper, tetraphenylporphyrin, tetraphenylporphyrin iron (III) chloride, tetraphenyl-21H, 23H-porphyrin cobalt (II), 5,10,15,20-tetrakis(4-methoxyphenyl)-21H, 23H-porphyrin iron (III) chloride, tetraphenyl-porphyrin zinc, 2,3,7,8,12,13,17,18-octaethyl-21H, 23H-porphyrin manganese (III) chloride, tetraphenylporphyrin copper, 5,10,15,20-tetrakis(4-methoxyphenyl)porphyrin cobalt (II) methylphenylporphyrin cobalt) and 7 pH indicators (chlorophenol red, bromocresol green, methyl red, cresol red, phenol red, bromophenol blue, neutral red).

[0103] Specifically, the time series detection data specifically includes the RGB color feature difference of the tea oil sample and the sequence number to which the time series feature of the tea oil sample belongs.

[0104] It should be explained that the RGB color feature difference of the above-mentioned tea oil sample specifically refers to the difference between the color values ​​of the red (R), green (G), and blue (B) three color channels of the sensor array image of the tea oil sample after the oil adulteration time series detection taken by the image acquisition device and the RGB color feature reference values ​​preset in the tea oil detection library. The RGB color features of the 16 color blocks in the image every minute can be automatically extracted by the built-in camellia oil olfactory visualization detection system of the tea oil adulteration time series detection platform, and the difference is calculated with the RGB color feature reference values ​​preset in the tea oil detection library to obtain the result; the sequence number to which the time series feature of the above-mentioned tea oil sample belongs specifically refers to a 960-dimensional time series feature. The sequence number corresponding to the sequence feature (16 color sensitive points × 3 color features × 20 minutes), for example, this example is 6-dimensional (2 color sensitive points × 3 color features × 1 minute), 0 means no, 1 means yes, the first color sensitive point has red (R), green (G), and no blue (B) within 1 minute, and the second color sensitive point has no red (R), green (G), and blue (B) within 1 minute, then the time series feature of its tea oil sample belongs to the sequence number 110011, which can be obtained by processing and analyzing the RGB color feature difference of the tea oil sample through the built-in camellia oil olfactory visualization detection system of the tea oil adulteration time series detection platform. The RGB difference feature vector generated by the difference.

[0105] The environmental data during the time series detection process specifically include the average humidity, maximum carbon dioxide concentration, maximum sulfur dioxide concentration, maximum light intensity and maximum dust concentration of the testing room to which the tea oil sample belongs during the tea oil adulteration detection cycle.

[0106] It should be explained that the above-mentioned tea oil adulteration detection cycle can be obtained through the SIEM system built into the tea oil adulteration timing detection platform. Specifically, it refers to the time period from the start of adulteration detection of tea oil samples to the end of detection, which is recorded as the tea oil adulteration detection cycle. The setting of the specific detection cycle duration will present different tea oil adulteration detection cycles based on multiple factors such as the differences in tea oil samples and the differences in detection reagents.

[0107] It needs to be explained that the average humidity of the testing room to which the above-mentioned tea oil samples belong during the tea oil adulteration detection cycle, specifically refers to the average value of the humidity in the testing room to which the tea oil samples belong when the tea oil adulteration detection is carried out during the tea oil adulteration detection cycle, which can be obtained by a high-sensitivity humidity sensor installed in the testing room to which the tea oil samples belong, wherein the humidity sensor can monitor in real time multiple humidity values ​​of the testing room to which the tea oil samples belong during the tea oil adulteration detection cycle, and obtain the average humidity through mean processing; the maximum carbon dioxide concentration in the testing room to which the above-mentioned tea oil samples belong during the tea oil adulteration detection cycle, specifically refers to the maximum carbon dioxide concentration in the testing room to which the tea oil samples belong when the tea oil adulteration detection is carried out during the tea oil adulteration detection cycle. The maximum concentration can be obtained by a high-sensitivity carbon dioxide detector installed in the testing room to which the tea oil sample belongs, wherein the carbon dioxide detector can monitor in real time multiple carbon dioxide concentration values ​​of the testing room to which the tea oil sample belongs during the tea oil adulteration detection period, and screen out the maximum value as the maximum carbon dioxide concentration by comparing the values; the maximum sulfur dioxide concentration of the testing room to which the tea oil sample belongs during the tea oil adulteration detection period specifically refers to the maximum sulfur dioxide concentration of the testing room to which the tea oil sample belongs when the tea oil adulteration detection is carried out during the tea oil adulteration detection period, which can be obtained by a high-sensitivity sulfur dioxide detector installed in the testing room to which the tea oil sample belongs, wherein the sulfur dioxide detector The instrument can monitor in real time multiple sulfur dioxide concentration values ​​of the testing room to which the tea oil sample belongs within the tea oil adulteration detection cycle, and screen out the maximum value as the maximum sulfur dioxide concentration by comparing the values. The maximum light intensity of the testing room to which the tea oil sample belongs within the tea oil adulteration detection cycle specifically refers to the maximum value of the light intensity of the testing room to which the tea oil sample belongs when the tea oil adulteration detection is carried out within the tea oil adulteration detection cycle, which can be obtained by a high-sensitivity illuminance meter installed in the testing room to which the tea oil sample belongs. The illuminance meter can monitor in real time multiple light intensity values ​​of the testing room to which the tea oil sample belongs within the tea oil adulteration detection cycle, and screen out the maximum value by comparing the values. Select the maximum value as the maximum light intensity; the maximum concentration of suspended dust in the testing room to which the above-mentioned tea oil sample belongs during the tea oil adulteration detection period specifically refers to the highest concentration value of suspended particulate matter (dust) in the air of the testing room to which the tea oil sample belongs when the tea oil adulteration detection is carried out during the tea oil adulteration detection period. The optical scatterometer installed in the testing room to which the tea oil sample belongs can be used to emit a laser beam into the air. The laser beam is scattered by the particulate matter in the air, and the concentration of the particulate matter is calculated based on the signal of the scattered light. Multiple dust concentration values ​​in the testing room to which the tea oil sample belongs during the tea oil adulteration detection period can be monitored in real time, and the maximum value can be screened out as the maximum dust concentration by comparing the values.

[0108] Furthermore, the tea oil time series detection index has the following specific analysis process:

[0109] The operation influence set corresponding to each timing detection device is obtained, specifically, the maximum operating temperature of each timing detection device within the tea oil adulteration detection cycle.

[0110] It should be explained that the method of obtaining the maximum operating temperature of the above-mentioned timing detection devices within the tea oil adulteration detection cycle is the same as the average operating temperature of each temperature measurement point belonging to the above-mentioned timing detection devices during the equipment performance detection period.

[0111] The sequence numbers of the time series features of the tea oil samples are compared one by one with the initial sequence numbers of the time series features preset in the tea oil detection library, thereby obtaining the deviation rate of the sequence numbers of the time series features of the tea oil samples. For example, the example here is 6-dimensional (2 color sensitive points × 3 color features × 1 minute), 0 represents none, 1 represents yes, the first color sensitive point has red (R), green (G), and no blue (B) within 1 minute, and the second color sensitive point has no red (R), green (G), and blue (B) within 1 minute, then the sequence number of the time series features of its tea oil samples is 110011. If the initial sequence number of the time series features preset in the tea oil detection library is 101001, then the deviation rate of the sequence number of the time series features of its tea oil samples is 50%.

[0112] Through the RGB color feature difference of tea oil samples, the sequence number deviation rate of the time series characteristics of tea oil samples, the average humidity of the testing room to which the tea oil samples belong during the tea oil adulteration detection cycle, the maximum concentration of carbon dioxide, the maximum concentration of sulfur dioxide, the maximum light intensity, the maximum concentration of floating dust, the image quality index and the maximum operating temperature of each timing detection equipment during the tea oil adulteration detection cycle.

[0113] The image quality index specifically refers to the image quality index of the image acquisition device obtained through a comprehensive analysis of the image resolution, average image sharpness, image grayscale value, image brightness contrast, image signal-to-noise ratio, maximum noise intensity and power fluctuation index of the image acquisition device within the image detection cycle during the image detection process.

[0114] The tea oil time series detection index is obtained through comprehensive analysis. The specific analysis method is as follows:

[0115] ;

[0116] In the formula, Indicates the RGB color feature difference of tea oil sample, Indicates the deviation rate of the sequence number to which the time series characteristics of tea oil samples belong, It indicates the average humidity of the testing room where the tea oil sample belongs during the tea oil adulteration testing period. Indicates the average humidity reference value preset in the tea oil detection library. It indicates the maximum concentration of carbon dioxide in the testing room where the tea oil sample belongs during the tea oil adulteration testing cycle. It indicates the maximum concentration of sulfur dioxide in the testing room where the tea oil sample belongs during the tea oil adulteration testing period. It indicates the maximum light intensity of the testing room where the tea oil sample belongs during the tea oil adulteration testing cycle. Indicates the maximum light intensity reference value preset in the tea oil detection library. It indicates the maximum concentration of floating dust in the testing room where the tea oil sample belongs during the tea oil adulteration testing period. Represents the image quality index of the image acquisition device, It indicates the maximum operating temperature of the Nth timing detection device during the tea oil adulteration detection cycle.

[0117] The average humidity reference value specifically refers to the optimal adaptation value of the average humidity set in the tea oil detection library, which aims to ensure that the tea oil adulteration detection reaches the optimal state and improve the accuracy and reliability of the detection.

[0118] The maximum light intensity reference value specifically refers to the optimal adaptation value of light intensity set in the tea oil detection library, which aims to ensure that tea oil adulteration detection reaches the optimal state and improve the accuracy and reliability of detection.

[0119] It represents the tea oil timing detection index. If the maximum light intensity is too high or too low, it will affect the shooting process of the image acquisition device, resulting in a decrease in the image quality index and a large difference in the RGB color features of the tea oil sample. This will cause the deviation rate of the sequence number of the time series features of the tea oil sample to increase, indicating that the oil sample is adulterated, which will lead to a decrease in the tea oil timing detection index. Excessive concentrations of carbon dioxide and sulfur dioxide will accelerate the oxidation of the oil sample. The higher the temperature, the faster the oxidation. If the humidity is too high, it will also accelerate the dissolution of carbon dioxide and sulfur dioxide in the detection reagent, affecting the pH (acid-base) value of the detection reagent, resulting in a decrease in the accuracy of tea oil timing detection and a decrease in the tea oil timing detection index. If the humidity is too low, the maximum concentration of floating dust will increase, polluting the detection reagent, resulting in a decrease in the accuracy of tea oil timing detection, and ultimately a decrease in the tea oil timing detection index.

[0120] The timing detection weight index representing the preset RGB color feature difference in the tea oil detection library represents the proportion of the RGB color feature difference of the tea oil sample to the tea oil timing detection index. The tea oil detection library stores the corresponding relationship between the timing detection weight index. For example, the real-time RGB color feature difference of the tea oil sample and the tea oil timing detection index can be input into the tea oil detection library. The tea oil detection library automatically matches the timing detection weight index corresponding to the RGB color feature difference according to the preset corresponding relationship between the timing detection weight index.

[0121] The time series detection weight index representing the deviation rate of the sequence number to which the time series feature preset in the tea oil detection library belongs, represents the proportion of the deviation rate of the sequence number to which the time series feature of the tea oil sample belongs to the tea oil time series detection index. The tea oil detection library stores the corresponding relationship between the time series detection weight index. For example, the deviation rate of the sequence number to which the time series feature of the real-time tea oil sample belongs and the tea oil time series detection index can be input into the tea oil detection library. The tea oil detection library automatically matches the time series detection weight index corresponding to the deviation rate of the sequence number to which the time series feature belongs according to the preset corresponding relationship between the time series detection weight index.

[0122] The time series detection weight index representing the average humidity preset in the tea oil detection library represents the proportion of the average humidity of the detection room to which the tea oil sample belongs during the tea oil adulteration detection cycle to the tea oil time series detection index. The tea oil detection library stores the corresponding relationship between the time series detection weight index. For example, the real-time average humidity of the detection room to which the tea oil sample belongs during the tea oil adulteration detection cycle and the tea oil time series detection index can be input into the tea oil detection library. The tea oil detection library automatically matches the time series detection weight index corresponding to the average humidity according to the preset corresponding relationship between the time series detection weight index.

[0123] The time series detection weight index representing the preset maximum carbon dioxide concentration in the tea oil detection library represents the proportion of the maximum carbon dioxide concentration of the detection room to which the tea oil sample belongs during the tea oil adulteration detection cycle to the tea oil time series detection index. The tea oil detection library stores the corresponding relationship between the time series detection weight indices. For example, the real-time maximum carbon dioxide concentration of the detection room to which the tea oil sample belongs during the tea oil adulteration detection cycle and the tea oil time series detection index can be input into the tea oil detection library. The tea oil detection library automatically matches the time series detection weight index corresponding to the maximum carbon dioxide concentration according to the preset corresponding relationship between the time series detection weight indices.

[0124] The time series detection weight index representing the preset maximum sulfur dioxide concentration in the tea oil detection library represents the proportion of the maximum sulfur dioxide concentration of the detection room to which the tea oil sample belongs during the tea oil adulteration detection cycle to the tea oil time series detection index. The tea oil detection library stores the corresponding relationship between the time series detection weight indices. For example, the real-time maximum sulfur dioxide concentration of the detection room to which the tea oil sample belongs during the tea oil adulteration detection cycle and the tea oil time series detection index can be input into the tea oil detection library. The tea oil detection library automatically matches the time series detection weight index corresponding to the maximum sulfur dioxide concentration according to the preset corresponding relationship between the time series detection weight indices.

[0125] The timing detection weight index representing the preset maximum light intensity in the tea oil detection library represents the proportion of the maximum light intensity of the detection room to which the tea oil sample belongs during the tea oil adulteration detection cycle to the tea oil timing detection index. The tea oil detection library stores the corresponding relationship between the timing detection weight indices. For example, the real-time maximum light intensity of the detection room to which the tea oil sample belongs during the tea oil adulteration detection cycle and the tea oil timing detection index can be input into the tea oil detection library. The tea oil detection library automatically matches the timing detection weight index corresponding to the maximum light intensity according to the preset timing detection weight index corresponding relationship.

[0126] The time series detection weight index representing the preset maximum concentration of floating dust in the tea oil detection library represents the proportion of the maximum concentration of floating dust in the detection room to which the tea oil sample belongs during the tea oil adulteration detection cycle to the tea oil time series detection index. The tea oil detection library stores the corresponding relationship between the time series detection weight indices. For example, the real-time maximum concentration of floating dust in the detection room to which the tea oil sample belongs during the tea oil adulteration detection cycle and the tea oil time series detection index can be input into the tea oil detection library. The tea oil detection library automatically matches the time series detection weight index corresponding to the maximum concentration of floating dust according to the preset corresponding relationship between the time series detection weight indices.

[0127] The timing detection weight index representing the image quality index preset in the tea oil detection library represents the proportion of the image quality index of the image acquisition device to the tea oil timing detection index. The tea oil detection library stores the corresponding relationship between the timing detection weight indices. For example, the image quality index of the real-time image acquisition device and the tea oil timing detection index can be input into the tea oil detection library. The tea oil detection library automatically matches the timing detection weight index corresponding to the image quality index according to the preset corresponding relationship between the timing detection weight indices.

[0128] The timing detection weight index representing the maximum operating temperature value preset in the tea oil detection library represents the proportion of the maximum operating temperature of each timing detection device within the tea oil adulteration detection cycle to the tea oil timing detection index. The tea oil detection library stores the corresponding relationship between the timing detection weight indices. For example, the real-time maximum operating temperature of each timing detection device within the tea oil adulteration detection cycle and the tea oil timing detection index can be input into the tea oil detection library. The tea oil detection library automatically matches the timing detection weight index corresponding to the maximum operating temperature according to the preset corresponding relationship between the timing detection weight indices.

[0129] M, M is the total number of timing detection devices.

[0130] Specifically, the specific process of judging whether the tea oil is adulterated is as follows:

[0131] If the tea oil timing detection index is greater than or equal to the tea oil timing detection threshold preset in the tea oil detection library, it means that the tea oil is not adulterated; the tea oil timing detection threshold represents the critical value used to judge whether the tea oil timing detection index reaches the tea oil purity standard during the tea oil timing detection process, and is used to measure whether the tea oil is adulterated and the adulteration concentration; the above-mentioned tea oil timing detection index is greater than or equal to the tea oil timing detection threshold preset in the tea oil detection library, indicating that the tea oil is not adulterated.

[0132] If the tea oil timing detection index is less than the tea oil timing detection threshold preset in the tea oil detection library, it indicates that the tea oil is adulterated. The tea oil timing detection index is subtracted from the tea oil timing detection threshold to obtain the tea oil timing detection deviation value, and the value is matched with the tea oil adulteration data set corresponding to each tea oil timing detection deviation value interval preset in the tea oil detection library to obtain the tea oil adulteration data set of the tea oil sample. Finally, the type and concentration of tea oil adulteration are obtained through the tea oil adulteration data set of the tea oil sample.

[0133] The tea oil adulteration data set corresponding to the above tea oil time series detection deviation value intervals has the following specific matching process:

[0134] The tea oil adulteration data set corresponding to the tea oil time series detection deviation value mentioned above is constructed through scientific collection, preprocessing and analysis. Each tea oil time series detection deviation value is carefully divided, and the corresponding tea oil adulteration data set is designed based on the key factors such as tea oil purity standard and common adulteration types for each tea oil time series detection deviation value range.

[0135] For example, if the tea oil timing detection deviation value in this example is 3, it belongs to the tea oil timing detection deviation value interval corresponding to [3, 5]. At the same time, the tea oil adulteration data sets corresponding to the tea oil timing detection deviation value intervals [3, 5] preset in the tea oil detection library are adulteration type: rapeseed oil and adulteration concentration: 20%.

[0136] The above contents are merely examples and explanations of the structure of the present invention. The technicians in this technical field may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the structure of the invention or exceed the scope defined by the present invention, they should all fall within the protection scope of the present invention.

Claims

1. A tea oil adulteration time series detection method based on a colorimetric sensor array, characterized in that: include: S1. The tea oil adulteration timing detection platform obtains the performance information of each timing detection device, thereby determining the performance stability index of each timing detection device, and compares it with the timing detection device performance stability threshold preset in the tea oil detection library, thereby determining whether to adjust the performance of each timing detection device; S2. Perform image capture detection on the image acquisition device, thereby obtaining image data of the image acquisition device, analyzing the image quality index of the image acquisition device, and comparing it with the image quality threshold preset in the tea oil detection library. If the image quality index is higher than or equal to the image quality threshold, the tea oil adulteration timing detection platform directly issues an instruction for tea oil adulteration timing detection, thereby each timing detection device and the image acquisition device are used for tea oil adulteration timing detection. If the image quality index is lower than the image quality threshold, the image acquisition device is optimized for image quality, so that the image quality index corresponding to the image data after image quality optimization is higher than or equal to the image quality threshold; S3. Based on the timing detection process of tea oil adulteration by each timing detection device and image acquisition device, the sensor array image captured by the image acquisition device is obtained, and the tea oil adulteration timing detection platform performs timing detection on the sensor array image to obtain timing detection data, and simultaneously obtains environmental data in the timing detection process, and obtains the tea oil timing detection index by combining the image quality index for comprehensive analysis, and compares it with the tea oil timing detection threshold preset in the tea oil detection library to determine whether the tea oil is adulterated; The specific analysis process of the tea oil time series detection index is as follows: Obtaining the operation influence set corresponding to each timing detection device, specifically the maximum operating temperature of each timing detection device within the tea oil adulteration detection cycle; The sequence numbers of the time series characteristics of the tea oil samples are compared and verified one by one with the initial sequence numbers of the time series characteristics preset in the tea oil detection library, thereby obtaining the deviation rate of the sequence numbers of the time series characteristics of the tea oil samples; The tea oil time series detection index is obtained by comprehensive analysis through the RGB color feature difference of tea oil samples, the deviation rate of the sequence number of the time series feature of tea oil samples, the average humidity of the detection room to which the tea oil samples belong during the tea oil adulteration detection period, the maximum concentration of carbon dioxide, the maximum concentration of sulfur dioxide, the maximum light intensity, the maximum concentration of floating dust, the image quality index, and the maximum operating temperature of each time series detection equipment during the tea oil adulteration detection period. The specific analysis method is as follows: In the formula, Indicates the tea oil time series detection index, Indicates the RGB color feature difference of tea oil sample, Indicates the deviation rate of the sequence number to which the time series characteristics of tea oil samples belong, It indicates the average humidity of the testing room where the tea oil sample belongs during the tea oil adulteration testing period. Indicates the average humidity reference value preset in the tea oil detection library. It indicates the maximum concentration of carbon dioxide in the testing room where the tea oil sample belongs during the tea oil adulteration testing cycle. It indicates the maximum concentration of sulfur dioxide in the testing room where the tea oil sample belongs during the tea oil adulteration testing period. It indicates the maximum light intensity of the testing room where the tea oil sample belongs during the tea oil adulteration testing cycle. Indicates the maximum light intensity reference value preset in the tea oil detection library. It indicates the maximum concentration of floating dust in the testing room where the tea oil sample belongs during the tea oil adulteration testing period. Represents the image quality index of the image acquisition device, It represents the maximum operating temperature of the Nth timing detection device during the tea oil adulteration detection cycle. Represents the temporal detection weight index of the RGB color feature difference preset in the tea oil detection library, The time series detection weight index representing the deviation rate of the sequence number of the preset time series feature in the tea oil detection library, Indicates the time series detection weight index of the average humidity preset in the tea oil detection library, Indicates the time series detection weight index of the maximum concentration of carbon dioxide preset in the tea oil detection library, Indicates the time series detection weight index of the maximum sulfur dioxide concentration preset in the tea oil detection library, Represents the time series detection weight index of the maximum light intensity preset in the tea oil detection library, Indicates the time series detection weight index of the maximum concentration of floating dust preset in the tea oil detection library, represents the temporal detection weight index of the image quality index preset in the tea oil detection library, Indicates the timing detection weight index of the maximum operating temperature preset in the tea oil detection library, M, M is the total number of timing detection devices.

2. The method for detecting tea oil adulteration based on a colorimetric sensor array according to claim 1, characterized in that: The performance information of each timing detection device specifically includes the average operating temperature of each temperature measurement point of each timing detection device during the equipment performance detection period, the maximum transmission power of each timing detection device during the equipment performance detection period, the minimum transmission power of each timing detection device during the equipment performance detection period, the maximum noise intensity of each timing detection device during the equipment performance detection period, and the energy consumption ratio of each timing detection device during the equipment performance detection period.

3. A tea oil adulteration timing detection method based on a colorimetric sensor array according to claim 2, characterized in that: The specific analysis process of the performance stability index of each timing detection device is as follows: Performing difference processing on the maximum transmission power and the minimum transmission power of each timing detection device during the device performance detection period to obtain the power fluctuation index of each timing detection device during the device performance detection period; The performance stability index of each timing detection device is obtained by comprehensive analysis of the temperature fluctuation, power fluctuation index, maximum noise intensity and energy consumption ratio of each timing detection device during the equipment performance detection period.

4. The method for detecting tea oil adulteration based on a colorimetric sensor array according to claim 1, characterized in that: The specific process of determining whether to adjust the performance of each timing detection device is as follows: If the performance stability index of a certain timing detection device is less than the performance stability threshold of the timing detection device, the performance of the timing detection device is adjusted; If the performance stability index of a timing detection device is greater than or equal to the performance stability threshold of the timing detection device, there is no need to adjust the performance of the timing detection device.

5. The method for detecting tea oil adulteration based on a colorimetric sensor array according to claim 1, characterized in that: The image data of the image acquisition device specifically includes the image resolution of the image acquisition device within an image detection period, the pixel value of each sample pixel point, the image grayscale value, the maximum brightness of the image, and the minimum brightness of the image; Perform variance processing on the pixel value of each sample pixel point to obtain the variance of the image signal; Perform standard deviation processing on the pixel value of each sample pixel point to obtain the image noise standard deviation; The image signal-to-noise ratio is obtained by ratioing the image signal variance to the image noise standard deviation.

6. A tea oil adulteration time sequence detection method based on a colorimetric sensor array according to claim 5, characterized in that: The image quality index of the image acquisition device, the specific analysis process is: The pixel values ​​of each sample pixel point of the image acquisition device within the image detection period are processed by variance, thereby obtaining the average sharpness of the image of the image acquisition device within the image detection period; Perform difference processing on the maximum brightness and the minimum brightness of the image of the image acquisition device within the image detection period, and mark the result as the image brightness contrast of the image acquisition device within the image detection period; Obtain the maximum noise intensity and power fluctuation index of the image acquisition device during the image detection cycle; The image quality index of the image acquisition device is comprehensively analyzed through the image resolution, image average sharpness, image grayscale value, image brightness contrast, image signal-to-noise ratio, maximum noise intensity and power fluctuation index of the image acquisition device during the image detection cycle.

7. The method for detecting tea oil adulteration based on a colorimetric sensor array according to claim 1, characterized in that: The image quality of the image acquisition device is optimized, and the specific optimization process is to perform denoising on the captured image, adjust the image grayscale value, and enhance the image edge information; The denoising of the captured image is specifically denoising of the captured image by the image acquisition device; The image grayscale value adjustment specifically adjusts the image grayscale value of the captured image to increase the difference between different grayscale levels in the image; The image edge information enhancement specifically performs image edge information enhancement on the captured image to highlight edge features in the image.

8. The method for detecting tea oil adulteration based on a colorimetric sensor array according to claim 1, characterized in that: The time series detection data specifically includes the RGB color feature difference of the tea oil sample and the sequence number to which the time series feature of the tea oil sample belongs; The environmental data during the time series detection process specifically include the average humidity, maximum carbon dioxide concentration, maximum sulfur dioxide concentration, maximum light intensity and maximum dust concentration of the testing room to which the tea oil sample belongs during the tea oil adulteration detection cycle.

9. The method for detecting tea oil adulteration based on a colorimetric sensor array according to claim 1, characterized in that: The specific process of judging whether the tea oil is adulterated is as follows: If the tea oil time series detection index is greater than or equal to the tea oil time series detection threshold preset in the tea oil detection library, it means that the tea oil is not adulterated; If the tea oil timing detection index is less than the tea oil timing detection threshold preset in the tea oil detection library, it indicates that the tea oil is adulterated. The tea oil timing detection index is subtracted from the tea oil timing detection threshold to obtain the tea oil timing detection deviation value, and the value is matched with the tea oil adulteration data set corresponding to each tea oil timing detection deviation value interval preset in the tea oil detection library to obtain the tea oil adulteration data set of the tea oil sample. Finally, the type and concentration of tea oil adulteration are obtained through the tea oil adulteration data set of the tea oil sample.

Images