A method and system for detecting iron in down based on spectral analysis

Abstract

The present application relates to the technical field of down detection, and particularly relates to a method and system for detecting iron in down based on spectral analysis; the present application obtains a down sample to be detected by pretreating the down to be detected; performs dual-mode spectral synergistic excitation on the down sample to be detected, collects signals, and obtains detection data of the down sample to be detected; performs background correction on the detection data of the down sample to be detected, and obtains corrected data of the down sample to be detected; obtains an iron detection result of the down sample to be detected according to the corrected data of the down sample to be detected; and thus accurate detection of iron in down is realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of down detection, and particularly relates to a method and system for detecting iron in down based on spectral analysis. BACKGROUND

[0002] Down is a natural thermal material, and its safety directly affects the health of consumers. In the production, storage or transportation process of down, iron impurities may be introduced due to contact with metal equipment (such as sorting machines and dryers), environmental corrosion or human intervention. Excessive iron can cause down to oxidize and discolor (such as yellowing), catalyze the growth of microorganisms, and even cause allergies or inflammation through skin contact. Therefore, domestic and foreign standards (such as GB / T 17685-2016 "Down Feathers" and FZ / T 81002-2016 "Washed Down Feathers") usually stipulate that the iron content in down should be lower than a certain limit value (such as 0.3 mg / g or lower), and if the threshold is exceeded, it is determined to be unqualified.

[0003] In the prior art, there is a scheme for detecting down by spectral analysis to determine whether the iron content exceeds the standard. However, the uneven dispersion of down suspension is one of the main interference factors affecting the accuracy of spectral detection. Due to the natural agglomeration of down fibers, there may be local differences in fiber density on the surface of the suspension (such as fiber agglomerate protrusions or sparse area depressions). If the laser focus is fixed at an initial position, when the light spot falls on the fiber agglomerate, the detection signal may be too high due to excessive accumulation of fibers; on the contrary, if the light spot falls in the sparse area, the signal may be too low due to insufficient fibers; this may eventually cause the detection result to deviate significantly from the actual value.

[0004] In addition, in the prior art, when the sampling image is used to locate the down in the down suspension, the micro features of the down suspension are significantly different due to differences in sample source (duck down / goose down) and storage conditions, which affects the accurate positioning of the down in the down suspension. SUMMARY

[0005] To solve the above technical problems, the present application provides a method and system for detecting iron in down based on spectral analysis to solve the problems in the prior art.

[0006] The present application provides a method for detecting iron in down based on spectral analysis, comprising the following steps:

[0007] S1: pretreating the down to be detected to obtain a down sample to be detected;

[0008] S2: performing dual-mode spectral cooperative excitation on the down sample to be detected and collecting signals to obtain detection data of the down sample to be detected;

[0009] The adaptive focusing step based on image feedback is used to compensate for the focusing deviation caused by uneven sample dispersion or microfluidic chip surface fluctuation in real time. Specifically, a CMOS camera is used to capture microscopic image data of the sample pool surface in real time, and the image data is transmitted to an image processing unit through a USB 3.0 interface. The image acquisition unit calculates the actual focusing position of the sample to be detected based on a mixed algorithm of edge detection algorithm and gray center method. The edge detection algorithm is an improved Canny operator algorithm, and the improved Canny operator algorithm uses an adaptive threshold to realize edge detection of the microscopic image data. Specifically: Sa: calculate the gradient amplitude of each pixel point of the microscopic image data and generate a gradient histogram; Sb: count the peak-valley structure and cumulative distribution of the gradient histogram; Sc: determine the threshold number and value of the Canny operator algorithm according to the peak-valley structure and cumulative distribution;

[0010] S3: Perform background correction operation on the detection data of the sample to be detected to obtain the corrected data of the sample to be detected.

[0011] S4: Obtain the iron detection result of the sample to be detected according to the corrected data of the sample to be detected.

[0012] Preferably, in the Sb, the peak-valley structure includes two indicators of main peak position and peak width. The main peak position μ is a concentrated interval of gradient amplitude in the microscopic image data, which is used to reflect the dominant contrast of the fiber edge in the microscopic image data. The peak width σ is the standard deviation of the main peak in the microscopic image data, which is used to reflect the dispersion degree of the gradient distribution.

[0013] The cumulative distribution is reflected by a background noise level N indicator. The background noise level N is the pixel ratio of the low gradient interval, which is used to reflect the purity of the sample to be detected.

[0014] Preferably, the Sc is specifically:

[0015] If the peak width σ of the gradient histogram of the microscopic image data is ≤15 and the background noise level N is ≤0.3, a double-threshold detection algorithm is used.

[0016] And the calculation formula of the double threshold is:

[0017] ;

[0018] ;

[0019] In the formula, T high is the high threshold of the improved Canny operator algorithm, and T low is the low threshold of the improved Canny operator algorithm.

[0020] If the peak width σ of the gradient histogram of the microscopic image data is greater than 15 and the background noise level N is less than or equal to 0.3, a three-threshold detection algorithm is adopted;

[0021] And the calculation formula of the three thresholds is:

[0022] ;

[0023] ;

[0024] ;

[0025] wherein, T high is the high threshold of the improved Canny operator method, T mid is the middle threshold of the improved Canny operator method, and the T low is the low threshold of the improved Canny operator method.

[0026] If the background noise level N of the gradient histogram of the microscopic image data is greater than 0.3, at this time, the suspension of the sample contains a large amount of bubbles or oil, the low gradient interval proportion of the gradient distribution is high, and the background noise interference is strong, a four-threshold detection algorithm is adopted;

[0027] And the calculation formula of the four thresholds is:

[0028] ;

[0029] ;

[0030] ;

[0031] ;

[0032] wherein, T high is the high threshold of the improved Canny operator method, T mid is the middle threshold of the improved Canny operator method, and the T low is the low threshold of the improved Canny operator method, and the Tv low is the very low threshold of the improved Canny operator method.

[0033] Preferably, the S1 is specifically:

[0034] S1.1: sample weighing and initial dispersion are performed on the down-filled down to obtain an initial dispersion suspension;

[0035] S1.2: a surfactant is added to the initial dispersion suspension for auxiliary dispersion to obtain an auxiliary dispersion suspension;

[0036] S1.3: transferring the auxiliary dispersion suspension into the sample pool of the microfluidic chip and performing magnetic stirring to obtain the eider sample to be detected.

[0037] Preferably, the eider sample to be detected is subjected to dual-mode spectral synergistic excitation, and signal acquisition is performed to obtain the detection data of the eider sample to be detected, which is specifically:

[0038] A dual-wavelength compound laser is used to simultaneously output a 488 nm laser beam and a 785 nm laser beam. After the two laser beams are combined by a dichroic mirror, they are focused to the sample pool of the microfluidic chip through the same optical fiber probe. Then, time-sharing detection is performed. When the 488 nm laser beam is turned on, the 785 nm laser beam is turned off at the same time. After the fluorescent signal is collected by the optical fiber probe, it is transmitted to a fluorescence spectrometer through a filter set, so as to acquire the fluorescence data of the eider sample to be detected. When the 488 nm laser beam is turned off, the 785 nm laser beam is turned on at the same time. After the Raman signal is collected by the same optical fiber probe, it is transmitted to a Raman spectrometer through a long-pass filter, so as to acquire the Raman spectrum data of the eider sample to be detected. The fluorescence acquisition and the Raman spectrum acquisition are alternately performed, and the alternation period is 1 s.

[0039] Preferably, the S2 further comprises: using a CMOS camera to capture microimage data of the surface of the sample pool in real time at a rate of 30 frames / s, and transmitting the image data to an image processing unit through a USB 3.0 interface.

[0040] The image processing unit calculates the actual focusing position of the eider sample to be detected based on a mixed algorithm of an edge detection algorithm and a gray gravity center method.

[0041] Preferably, the S3 is specifically:

[0042] S3.1: blank background spectrum acquisition and data preprocessing to obtain preprocessed blank background spectrum data;

[0043] S3.2: performing dynamic background deduction on the detection data of the eider sample to be detected based on the preprocessed blank background spectrum data to obtain corrected data of the eider sample to be detected.

[0044] Preferably, the dynamic background deduction on the fluorescence data of the eider sample to be detected based on the preprocessed blank background spectrum data is specifically:

[0045] A scattering light proportion factor k is introduced scat The scattering light proportion factor is the background intensity ratio below 400 nm of the preprocessed blank background spectrum data to the detection data of the eider sample to be detected. According to the scattering light proportion factor k scatThe detection data of the to-be-detected down sample is subjected to dynamic background deduction to obtain corrected fluorescence data of the to-be-detected down sample.

[0046] The dynamic background deduction of the Raman spectrum data of the to-be-detected down sample based on the preprocessed blank background spectrum data is specifically:

[0047] Outside the iron characteristic peak interval, a region with obvious matrix characteristic peaks is selected to fit polynomial baseline data B matrix λ The background is deducted in the Raman spectrum data of the to-be-detected down sample to obtain corrected Raman spectrum data of the to-be-detected down sample.

[0048] Preferably, S4 is specifically:

[0049] S4.1: Extracting fluorescence features and Raman features from the corrected data of the to-be-detected down sample.

[0050] S4.2: Establishing a deep learning model architecture.

[0051] S4.3: Training the deep learning model architecture.

[0052] S4.4: Inputting the fluorescence features and Raman features of the to-be-detected down sample into the deep learning model architecture to obtain an iron detection result of the to-be-detected down sample.

[0053] According to another aspect of the present application, a down iron detection system based on spectral analysis is provided, which adopts the down iron detection method based on spectral analysis of embodiment 1, and the system comprises:

[0054] A sample pretreatment module is configured to pretreat the to-be-detected down to obtain a to-be-detected down sample.

[0055] A detection data acquisition module is configured to perform dual-mode spectrum synergistic excitation on the to-be-detected down sample and collect signals to obtain detection data of the to-be-detected down sample.

[0056] A detection data correction module is configured to perform background correction on the detection data of the to-be-detected down sample to obtain corrected data of the to-be-detected down sample.

[0057] An iron detection module is configured to obtain an iron detection result of the to-be-detected down sample according to the corrected data of the to-be-detected down sample.

[0058] The embodiments of the present application have the following technical effects:

[0059] ​The present application obtains a sample of the eiderdown to be detected by pretreating the eiderdown to be detected; obtains detection data of the sample of the eiderdown to be detected by performing double-mode spectrum cooperative excitation on the sample of the eiderdown to be detected and performing signal collection; obtains an iron detection result of the sample of the eiderdown to be detected by performing background correction on the detection data of the sample of the eiderdown to be detected; and thereby realizes accurate detection of iron in eiderdown.

[0060] Meanwhile, according to the characteristics of the eiderdown suspension, the adaptive threshold detection scheme is used to automatically match the threshold value strategy through the gradient distribution characteristics, without manual intervention, covering all scenarios of high uniformity, high aggregation-sparse mixture and high noise, and the double threshold simplifies the calculation, avoids weak edge detection caused by excessive subdivision, the three thresholds retain the medium contrast edges, improve the fiber coverage, the four thresholds + strict connection rules, the noise false detection rate is less than 0.5% (traditional method > 5%), and the gradient distribution characteristics of the eiderdown suspension are analyzed dynamically to intelligently match the number and value of the threshold value, while ensuring the integrity of the real edge and significantly suppressing noise interference. This scheme solves the limitations of the traditional fixed threshold method, and provides more accurate and robust eiderdown suspension image positioning technology support for rapid detection of iron in eiderdown. BRIEF DESCRIPTION OF DRAWINGS

[0061] In order to more clearly illustrate the specific embodiments of the present application or the technical solutions in the prior art, the drawings needed in the specific embodiments or the prior art description will be briefly introduced below. Obviously, the drawings in the following description are some embodiments of the present application, and those skilled in the art can also obtain other drawings according to these drawings without creative labor.

[0062] Figure 1 is a flowchart of an iron detection method in eiderdown based on spectrum analysis provided by an embodiment of the present application;

[0063] Figure 2 is a flowchart of edge detection of the microscopic image data using adaptive threshold provided by an embodiment of the present application. DETAILED DESCRIPTION

[0064] In order to make the purpose, technical scheme and advantages of the present application more clear, the technical scheme of the present application will be described clearly and completely below. Obviously, the described embodiments are only some of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of the present application.

[0065] Example 1,Figure 1 A flow chart of a method for detecting iron in down based on spectral analysis is shown in FIG. 1. Figure 1 As shown in FIG. 1, a method for detecting iron in down based on spectral analysis includes the following steps:

[0066] S1: pretreating the down to be detected to obtain a down sample to be detected;

[0067] The traditional method of homogenizing the down sample (such as drying and acid digestion) can destroy the original morphology of the sample or introduce interference, and simple mechanical stirring or ultrasonic treatment can cause the aggregation of down fibers, affecting the uniformity of subsequent spectral detection. In view of the above situation, a rapid pretreatment method based on the synergistic effect of a microfluidic chip and a surfactant is proposed in this step, which realizes the uniform dispersion of the down sample under the condition of no damage and no chemical pollution, and ensures the repeatability and accuracy of the iron detection.

[0068] Specifically, S1 is specifically:

[0069] S1.1: sample weighing and initial dispersion of the down to be detected to obtain an initial dispersion suspension;

[0070] According to the fluffiness of the down sample, 0.1-0.5g of the down to be detected is weighed.

[0071] If the sample fluffiness is high (such as raw down material without compression), the upper limit of 0.5g is taken; if the sample has been compressed (such as down pillow core disassembled material), the lower limit of 0.1g is taken.

[0072] The weighed down to be detected is transferred to a sample tube of a micro vortex mixer, 5mL of deionized water (temperature 25±1℃, pH 6.5-7.5) is added, and vortex mixing is performed at a speed of 300rpm for 2 minutes.

[0073] In this step, the mechanical shear force is used to break the part of the aggregation of the down fibers, so that the sample is preliminarily dispersed into a suspension state.

[0074] The temperature of the deionized water is 25±1℃, and the pH value is 6.5-7.5.

[0075] S1.2: adding a surfactant to the initial dispersion suspension for auxiliary dispersion to obtain an auxiliary dispersion suspension;

[0076] Add 0.1% Triton X-100 surfactant to the initial dispersion suspension and continue vortex mixing at 300 rpm for 1 minute. Triton X-100 is a nonionic surfactant whose hydrophobic end adsorbs oils on the surface of down fibers, while its hydrophilic end reduces the surface tension of water, promoting fiber dispersion in water. Down fiber surfaces often have natural oils (such as waxes) and processing residues (such as silicone oil) adhering to them. These substances increase the friction between fibers, leading to aggregation. Triton X-100 weakens the adhesion between fibers by "encapsulating" oil molecules, while its hydrophilic groups form hydrogen bonds with water molecules, making the fibers easier to disperse in the aqueous phase.

[0077] S1.3: The auxiliary dispersion suspension is transferred to the sample cell of the microfluidic chip and magnetically stirred to obtain the down sample to be tested;

[0078] The vortex-mixed suspension is transferred to the sample cell of a microfluidic chip. Based on the principle of "confined flow," the microfluidic chip's narrow channels force the down fibers into a single layer, preventing aggregation. A 3mm diameter magnetic stir bar is placed in a micro-chamber beneath the microfluidic chip, and the mixture is stirred at 500 rpm for 1 minute using an external magnetic stirrer. The vortex generated by the magnetic stirring further breaks up the fiber aggregates while avoiding damage to the fibers from direct mechanical shearing.

[0079] The microfluidic chip has a channel size of 50μm×100μm and is made of PDMS.

[0080] In this step, the combination of micro-vortex and magnetic stirring avoids the damage to the morphology of iron caused by ultrasonic waves or high-speed shearing, ensuring the authenticity of the test results. 0.1% TritonX-100 reduces surface tension without introducing detectable chemical residues. The design of micron-level channels and transparent materials enables real-time monitoring and efficient homogenization of dispersion uniformity, solving the problem of "uneven dispersion → large detection error" in traditional methods.

[0081] S2: Perform dual-mode spectral co-excitation on the down sample to be tested and acquire the signal to obtain the detection data of the down sample to be tested.

[0082] In this step, a dual-mode spectral detection module is used, which integrates a fluorescence detection unit and a surface-enhanced Raman (SERS) detection unit. Through synchronous excitation-time division detection and adaptive focusing technology, efficient collaborative acquisition of fluorescence and Raman signals is achieved, thereby obtaining the detection data of the down sample to be tested.

[0083] Specifically, S2 is as follows:

[0084] The dual-wavelength composite laser outputs 488nm laser beam and 785nm (SERS excitation) laser beam simultaneously, the two laser beams are combined by a dichroic mirror and then focused to the sample cell of the microfluidic chip through the same fiber probe;

[0085] The 488nm laser beam is used for fluorescence excitation, the 785nm laser beam is used for SERS excitation, and the diameter of the fiber probe is 50μm.

[0086] Then, time-sharing detection is performed, when the 488nm laser beam is turned on, the 785nm laser beam is turned off synchronously, the fluorescence signal is collected by the fiber probe and then transmitted to a fluorescence spectrometer through a filter set, so as to collect the fluorescence data of the to-be-detected down sample.

[0087] The cut-off wavelength of the filter set is 500nm, which is used to block the interference of 785nm laser scattering, the detection range of the fluorescence spectrometer is 500-700nm, and the integration time is 100ms.

[0088] When the 488nm laser beam is turned off, the 785nm laser beam is turned on synchronously, the Raman signal is collected by the same fiber probe and then transmitted to a Raman spectrometer through a long-pass filter, so as to collect the Raman spectrum data of the to-be-detected down sample.

[0089] The cut-off wavelength of the long-pass filter is 750nm, which is used to block the interference of 488nm laser scattering; the detection range of the Raman spectrometer is 1000-1800cm -1 , and the integration time is 500ms.

[0090] The fluorescence collection and Raman spectrum collection are alternately performed, and the alternating period is 1s.

[0091] In this step, a synchronous excitation-time-sharing detection mode is adopted, the synchronous spatial collection of dual-mode signals is realized through the same fiber probe, and the error caused by sample displacement in the traditional discrete optical path is eliminated.

[0092] Further, the uneven dispersion of the down suspension is one of the main interference factors affecting the spectral detection accuracy. Due to the natural agglomeration of down fibers, local fiber density differences may occur on the surface of the suspension (e.g., fiber agglomerates protrude or sparse areas are depressed). If the laser focus is fixed at an initial position, when the light spot falls on the fiber agglomerates, the detection signal may be biased high due to excessive fiber accumulation; on the contrary, if the light spot falls in the sparse area, the signal may be biased low due to insufficient fibers. This dynamic shift between the light spot and the sample surface can cause significant fluctuations in repeated detection results of the same sample, seriously affecting the accuracy and reliability of the iron content detection. In view of the uneven dispersion of the down suspension, an adaptive focusing step based on image feedback is introduced during signal acquisition to compensate for the focusing deviation caused by uneven dispersion of the sample or fluctuations on the surface of the microfluidic chip in real time, ensuring the spatial consistency of the dual-mode spectral signals.

[0093] Specifically,

[0094] A CMOS camera is used to capture real-time microscopic image data of the sample pool surface at a rate of 30 frames / second, and the image data is transmitted to the image processing unit through a USB3.0 interface;

[0095] The CMOS camera is integrated directly below the microfluidic chip and is vertically 500μm away from the surface of the sample pool, and is fixed by a support to ensure that the optical axis is coaxial with the laser probe; the technical parameters of the CMOS camera are: resolution: 2 million pixels (1600x1200 pixels); frame rate: 30 frames / second to meet the real-time dynamic monitoring requirements; field of view: 200μm x 200μm to correspond to a pixel resolution of 0.125μm / pixel, which can identify the micron-level topographic changes of fiber agglomerates; exposure time: 1ms to avoid image blurring caused by the flow of down suspension.

[0096] The image processing unit calculates the actual focusing position of the down sample to be detected based on a hybrid algorithm of edge detection algorithm and gray center of gravity method;

[0097] The edge detection algorithm is an improved Canny operator algorithm, and the Gaussian filter parameters of the improved Canny operator algorithm are: σ=1.0, which is used to balance noise suppression and edge preservation; the edge width is set to a single pixel level to avoid interference from false edges.

[0098] More importantly, the improved Canny operator method uses an adaptive threshold to realize edge detection of the microscopic image data,

[0099] The micro features (such as fiber density, agglomeration degree, and oil content) of the down suspension are significantly different due to differences in sample sources (duck down / goose down) and storage conditions. The segmentation method of the traditional fixed threshold (such as double threshold or triple threshold) is difficult to adapt to this diversity: high agglomeration sample: the edge gradient amplitude of the fiber agglomerate is high and concentrated (a high threshold is needed to distinguish the strong edge), but the fiber gradient in the sparse area is low (a fixed high threshold is easy to cause the edge of the sparse area to be missed); the fat sample: the background of the suspension is clean, and the noise interference is small (the low threshold can be relaxed to retain more weak edges), and the high oil sample has a complex background (the low threshold needs to be tightened to suppress noise); therefore, the embodiment proposes an adaptive threshold detection algorithm, which dynamically determines the number and specific value of the threshold by analyzing the gradient distribution characteristics of the down suspension, and realizes the precise edge segmentation of the "one strategy for one sample".

[0100] As shown in the accompanying Figure 2 , the edge detection of the microscopic image data is implemented by using the adaptive threshold, and specifically:

[0101] Sa: calculating the gradient amplitude of each pixel point of the microscopic image data, and generating a gradient histogram;

[0102] Wherein, the horizontal axis of the gradient histogram is the gradient amplitude G , and the vertical axis is the pixel point number ratio.

[0103] Sb: statistics of the peak-valley structure and cumulative distribution of the gradient histogram;

[0104] The peak-valley structure includes two indexes of main peak position and peak width; the main peak position mu is the main concentration interval of the gradient amplitude in the microscopic image data, which is used to reflect the dominant contrast of the fiber edge in the microscopic image data; the peak width sigma is the standard deviation of the main peak in the microscopic image data, which is used to reflect the dispersion degree of the gradient distribution and is related to the uniformity of the fiber agglomeration;

[0105] The cumulative distribution is reflected by the background noise level N index, and the background noise level N is the pixel ratio of the low gradient interval (G<20), which is used to reflect the purity of the suspension.

[0106] Sc: determining the threshold number and value of the Canny operator method according to the peak-valley structure and cumulative distribution;

[0107] Wherein, the Sc is specifically:

[0108] If the peak width sigma of the gradient histogram of the microscopic image data is less than or equal to 15 and the background noise level N is less than or equal to 0.3, a double threshold detection algorithm is used;

[0109] And the calculation formula of the double threshold is:

[0110] ;

[0111] ;

[0112] wherein, T high is the high threshold value of the improved Canny operator method, T low is the low threshold value of the improved Canny operator method.

[0113] If the peak width σ of the gradient histogram of the microscopic image data is greater than 15 and the background noise level N is less than or equal to 0.3, a three-threshold detection algorithm is used;

[0114] And the calculation formula of the three thresholds is:

[0115] ;

[0116] ;

[0117] ;

[0118] wherein, T high is the high threshold value of the improved Canny operator method, T mid is the medium threshold value of the improved Canny operator method, and T low is the low threshold value of the improved Canny operator method.

[0119] If the background noise level N of the gradient histogram of the microscopic image data is greater than 0.3, at this time, the suspension of the sample contains a large amount of bubbles or oil, the proportion of the low gradient interval of the gradient distribution is high, and the background noise interference is strong, a four-threshold detection algorithm is used;

[0120] And the calculation formula of the four thresholds is:

[0121] ;

[0122] ;

[0123] ;

[0124] ;

[0125] wherein, T high is the high threshold value of the improved Canny operator method, T mid is the medium threshold value of the improved Canny operator method, T low is the low threshold value of the improved Canny operator method, and Tv low is the very low threshold value of the improved Canny operator method.

[0126] At the same time, the connection rule is: for the weak edge (T vlow <G≤T low ), it needs to be connected with the medium / high edge at the same time and form a chain of ≥5 pixels to further eliminate noise.

[0127] The present application covers all scene samples such as high uniformity, high aggregation-sparse mixture and high noise through the adaptive threshold detection scheme, automatic matching of threshold strategy through gradient distribution characteristics, without manual intervention, and double threshold simplifies calculation, avoids weak edge missing caused by excessive subdivision; three thresholds retain medium contrast edges, improve fiber coverage; four thresholds + strict connection rule, noise false detection rate <0.5% (traditional method >5%); through dynamic analysis of the gradient distribution characteristics of down feather suspension, intelligent matching of threshold number and value, while ensuring the integrity of the real edge, significantly suppresses noise interference. This scheme solves the limitations of the traditional fixed threshold method "one size fits all", and provides more accurate and robust image positioning technology support for rapid detection of iron in down.

[0128] After edge detection of the microscopic image data, the actual focusing position of the down sample to be detected is calculated by the gray center method, and then the Z-axis height of the optical fiber probe is dynamically adjusted (step motor precision ±1 μm) to ensure that the laser is always focused on the most dense area of the sample, i.e. the uniform distribution of down fibers.

[0129] S3: performing background correction operation on the detection data of the down sample to be detected to obtain corrected data of the down sample to be detected;

[0130] The S3 is specifically:

[0131] S3.1: blank background spectrum acquisition and data preprocessing to obtain preprocessed blank background spectrum data;

[0132] Before each detection, 50 μL of deionized water containing 0.1% Triton X-100 surfactant is injected into the microfluidic chip, and then spectrum acquisition is performed;

[0133] Wherein, the fluorescence channel is excited by 488nm laser, and the emission spectrum of 500-700nm is collected, the integration time is 100ms, and the average of continuous 3 times is taken; the Raman channel is excited by 785nm laser, and the Raman spectrum of 1000-1800cm -1 Raman spectrum, integration time 500ms, repeated 3 times to get the average of blank background spectrum.

[0134] Then the baseline correction is performed on the blank background spectrum;

[0135] The blank background spectrum is fitted by a polynomial to subtract background noise, wherein the polynomial fitting is a 3-order polynomial fitting.

[0136] S3.2: performing dynamic background deduction on the detection data of the to-be-detected down sample based on the preprocessed blank background spectrum data, to obtain corrected data of the to-be-detected down sample;

[0137] The dynamic background deduction on the fluorescence data of the to-be-detected down sample based on the preprocessed blank background spectrum data is specifically:

[0138] The fluorescence signal of the down suspension is easily interfered by particle scattering, therefore, a scattering light proportion factor k is introduced scat to perform fluorescence intensity correction, wherein the scattering light proportion factor is a ratio of background intensity below 400 nm of the preprocessed blank background spectrum data to the detection data of the to-be-detected down sample;

[0139] The calculation formula is:

[0140] ;

[0141] In the formula, ∑ 400nm represents integral intensity of the waveband below 400 nm, F sample is the fluorescence data of the to-be-detected down sample, and F blank is the preprocessed blank background fluorescence data.

[0142] According to the scattering light proportion factor k scat , dynamic background deduction is performed on the detection data of the to-be-detected down sample, to obtain corrected fluorescence data of the to-be-detected down sample.

[0143] The specific formula is:

[0144] ;

[0145] In the formula, F corrected is the corrected fluorescence data of the to-be-detected down sample.

[0146] The dynamic background deduction on the Raman spectrum data of the to-be-detected down sample based on the preprocessed blank background spectrum data is specifically:

[0147] Organic matters (such as proteins and oils) in the down matrix will produce wide peaks (such as protein amide I band ~1650 cm -1 , and oil C-H bond ~2900 cm -1 ) in the Raman spectrum, which overlap with the characteristic peaks of iron (such as Fe-O bond ~600 cm -1 ). Dynamic baseline fitting is adopted to eliminate background interference:

[0148] in the iron characteristic peak interval (such as 200-800 cm-1 In addition, select regions with obvious matrix characteristic peaks (such as 1500-1700 cm⁻¹). -1 Fitting polynomial baseline data B matrix ( λ );

[0149] The specific formula is as follows:

[0150] ;

[0151] In the formula, a 0, a 1, a 2, a 3 is a coefficient, and λ is the wavelength.

[0152] Background subtraction is performed on the Raman spectral data of the down sample to be tested to obtain the corrected Raman spectral data of the down sample to be tested;

[0153] The specific formula is as follows:

[0154] ;

[0155] In the formula, R baseline_corrected The corrected Raman spectral data of the down sample to be tested. R sample The Raman spectral data of the down sample to be tested are as follows: B matrix ( λ () represents the polynomial baseline data.

[0156] S4: Obtain the iron content test result of the down sample to be tested based on the calibration data of the down sample to be tested.

[0157] Specifically, S4 is:

[0158] S4.1: Extract fluorescence and Raman features from the calibration data of the down sample to be tested;

[0159] The fluorescence characteristic is the maximum intensity F of the fluorescence peak at 520 nm. peak and the full width at half maximum (FWHM) of the 500-700 nm fluorescence peak width The Raman feature is 600 cm⁻¹. -1 The peak area R of the Fe-O bond Raman peak area and 600cm -1 Peak offset R relative to standard value shift .

[0160] S4.2: Establish a deep learning model architecture;

[0161] The deep learning model architecture is a multi-layer perception (MLP) and attention mechanism fusion architecture, specifically:

[0162] An input layer, configured to receive a 4-dimensional input feature (F peak , F width , R area , R shift ).

[0163] A hidden layer (MLP part), including: a first hidden layer: 16 neurons, an activation function ReLU (Rectified Linear Unit), performing a preliminary nonlinear transformation on the input feature; a second hidden layer: 8 neurons, an activation function ReLU, further extracting high-order features; a Dropout layer: a dropout rate of 0.2, preventing overfitting.

[0164] An attention mechanism layer, configured to apply attention weights to the feature vectors extracted by the hidden layer, dynamically adjusting the contribution of different features to the prediction of iron content.

[0165] An output layer, configured to output the iron detection result.

[0166] S4.3: training the deep learning model architecture;

[0167] The loss function of the deep learning model architecture is a mean square error (MSE) loss function, the optimizer is an Adam (adaptive moment estimation) optimizer, the initial learning rate is 0.001, the batch size is 16, and the training epochs are 100.

[0168] S4.4: inputting the fluorescence features and Raman features of the to-be-detected down sample into the deep learning model architecture to obtain the iron detection result of the to-be-detected down sample;

[0169] The iron detection result includes: iron detection pass and iron detection fail.

[0170] Further, the output layer of the deep learning model architecture further includes a threshold comparison module, which compares the content of the iron in the to-be-detected down sample predicted by the deep learning model architecture with a preset threshold. When the content is greater than the preset threshold, the output layer outputs the result of iron detection fail, otherwise, the output layer outputs the result of iron detection pass.

[0171] Embodiment 2, the present application also provides a down feather iron detection system based on spectral analysis, which adopts the down feather iron detection method based on spectral analysis in embodiment 1, and the system includes:

[0172] A sample pretreatment module is configured to pretreat the down to be detected to obtain a down sample to be detected;

[0173] A detection data acquisition module is configured to perform dual-mode spectrum synergistic excitation on the down sample to be detected and perform signal acquisition to obtain detection data of the down sample to be detected.

[0174] A detection data correction module is configured to perform background correction on the detection data of the down sample to be detected to obtain corrected data of the down sample to be detected.

[0175] An iron detection module is configured to obtain an iron detection result of the down sample to be detected according to the corrected data of the down sample to be detected.

[0176] In some embodiments, the electronic device can further include one or more of the following components: a bus system, a power supply system, a power management system, a processor, a memory, an input device, an output device, a communication device, a user interface, and the like.

[0177] The processor can be a central processing unit (CPU) or other form of processing unit having data processing and / or instruction execution capabilities, and can control other components in the electronic device to perform desired functions.

[0178] The memory can include one or more computer program products, which can include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may, for example, include random access memory (RAM), cache memory, and the like. The non-volatile memory may, for example, include read-only memory (ROM), hard disk, flash memory, and the like. One or more computer program instructions can be stored on the computer-readable storage medium, and the processor can execute the program instructions to implement the method for detecting iron in down based on spectrum analysis according to any embodiment of the present application and / or other desired functions. Various contents such as initial external parameters, threshold values, and the like can also be stored in the computer-readable storage medium.

[0179] In one example, the electronic device can further include an input device and an output device, which are interconnected by a bus system and / or other forms of connection mechanism (not shown). The input device can include, for example, a keyboard, a mouse, and the like. The output device can output various information to the outside, including pre-warning prompt information, braking force, and the like. The output device can include, for example, a display, a speaker, a printer, a communication network and a remote output device connected thereto, and the like.

[0180] Of course, components such as buses, input / output interfaces, and the like are omitted for simplicity. In addition, the electronic device can include any other appropriate components according to specific application cases.

[0181] In addition to the above method and device, the embodiments of the present application can also be a computer program product, which comprises computer program instructions, and the computer program instructions enable the processor to realize the functions of the method for detecting iron in down provided by any of the embodiments of the present application based on spectral analysis when the processor runs.

[0182] In addition, the embodiments of the present application can also be a computer readable storage medium, which stores computer program instructions, and the computer program instructions enable the processor to realize the method for detecting iron in down provided by any of the embodiments of the present application based on spectral analysis when the processor runs.

[0183] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the technical solutions of the embodiments of the present application.

Claims

1. A method for detecting iron in down based on spectral analysis, characterized in that, The method comprises the following steps: S1: pretreating the to-be-tested down feather to obtain a to-be-tested down feather sample; S2: performing dual-mode spectrum synergistic excitation on the to-be-tested down feather sample and collecting signals to obtain detection data of the to-be-tested down feather sample; in this step, an adaptive focusing step based on image feedback is adopted to compensate for focusing deviation caused by uneven sample dispersion or surface fluctuation of the microfluidic chip in real time; Specifically, a CMOS camera is used to capture microscopic image data of the surface of the sample pool in real time, and the image data is transmitted to an image processing unit through a USB 3.0 interface; the image acquisition unit calculates the actual focusing position of the to-be-tested down feather sample based on a mixed algorithm of an edge detection algorithm and a gray gravity center method; the edge detection algorithm is an improved Canny operator algorithm, and the improved Canny operator method uses an adaptive threshold to realize edge detection of the microscopic image data; specifically, Sa: calculating the gradient amplitude of each pixel point of the microscopic image data and generating a gradient histogram; Sb: counting the peak-valley structure and cumulative distribution of the gradient histogram; in Sb, the peak-valley structure includes two indexes of a main peak position and a peak width; the main peak position μ is a central interval of the gradient amplitude in the microscopic image data, which is used to reflect the dominant contrast of the fiber edge in the microscopic image data; the peak width σ is the standard deviation of the main peak in the microscopic image data, which is used to reflect the dispersion degree of the gradient distribution; The cumulative distribution is reflected by a background noise level N index, and the background noise level N is the pixel ratio of the low gradient interval, which is used to reflect the purity of the to-be-tested down feather sample; Sc: determining the number and value of the threshold of the Canny operator method according to the peak-valley structure and the cumulative distribution; Sc specifically is: If the peak width σ of the gradient histogram of the microscopic image data is ≤15 and the background noise level N is ≤0.3, a double-threshold detection algorithm is used; The calculation formula of the double threshold is: ； ； In the formula, T high is the high threshold value of the improved Canny operator method, and the T low is the low threshold value of the improved Canny operator method; If the peak width σ of the gradient histogram of the microscopic image data is >15 and the background noise level N is ≤0.3, a three-threshold detection algorithm is used; The calculation formula of the three threshold is: ； ； ； wherein T high is a high threshold value for the improved Canny operator method, T mid is a middle threshold value for the improved Canny operator method, the T low is a low threshold value for the improved Canny operator method; If the background noise level N of the gradient histogram of the microscopic image data is >0.3, at this time, the suspension of the sample contains a large amount of bubbles or oil, the gradient distribution has a high proportion of low gradient interval, and the background noise interference is strong, a four-threshold detection algorithm is used; The calculation formula of the four threshold is: ； ； ； ； wherein T high is a high threshold value for the improved Canny operator method, T mid is a medium threshold value for the improved Canny operator method, the T low is a low threshold value for the improved Canny operator method, the Tv low is an ultra-low threshold value for the improved Canny operator method; S3: performing background correction on the detection data of the to-be-tested down feather sample to obtain corrected data of the to-be-tested down feather sample; S4: obtaining the iron detection result of the to-be-tested down feather sample according to the corrected data of the to-be-tested down feather sample.

2. The method according to claim 1, wherein S1 specifically comprises: S1.1: performing sample weighing and initial dispersion on the to-be-tested down feather to obtain an initial dispersion suspension; S1.2: adding a surfactant to the initial dispersion suspension for auxiliary dispersion to obtain an auxiliary dispersion suspension; ​ S1.3: transferring the auxiliary dispersion suspension into the sample pool of the microfluidic chip and performing magnetic stirring to obtain the eider sample to be detected. 3.The eider iron detection method based on spectral analysis according to claim 1, characterized in that: The eider sample to be detected is subjected to dual-mode spectral synergistic excitation, and signal acquisition is performed to obtain detection data of the eider sample to be detected, which is specifically: A dual-wavelength composite laser is used to simultaneously output a 488 nm laser beam and a 785 nm laser beam, which are combined by a dichroic mirror and then focused on the sample pool of the microfluidic chip through the same optical fiber probe; then time-sharing detection is performed, the 785 nm laser beam is turned off synchronously when the 488 nm laser beam is turned on, the fluorescence signal is collected by the optical fiber probe and then transmitted to a fluorescence spectrometer through a filter group, thereby acquiring fluorescence data of the eider sample to be detected; the 785 nm laser beam is turned on synchronously when the 488 nm laser beam is turned off, the Raman signal is collected by the same optical fiber probe and then transmitted to a Raman spectrometer through a long-pass filter, thereby acquiring Raman spectrum data of the eider sample to be detected, and the fluorescence acquisition and Raman spectrum acquisition are alternately performed with an alternating period of 1 s. 4.The eider iron detection method based on spectral analysis according to claim 1, characterized in that: The S2 further comprises: using a CMOS camera to capture microscopic image data of the surface of the sample pool in real time at a rate of 30 frames / s, and transmitting the image data to an image processing unit through a USB 3.0 interface; The image processing unit calculates the actual focusing position of the eider sample to be detected based on a mixed algorithm of edge detection algorithm and gray center of gravity method. 5.The eider iron detection method based on spectral analysis according to claim 1, characterized in that: The S3 is specifically: S3.1: blank background spectrum acquisition and data preprocessing to obtain preprocessed blank background spectrum data; S3.2: performing dynamic background subtraction on the detection data of the eider sample to be detected based on the preprocessed blank background spectrum data to obtain corrected data of the eider sample to be detected. 6.The eider iron detection method based on spectral analysis according to claim 5, characterized in that: The dynamic background subtraction on the fluorescence data of the eider sample to be detected based on the preprocessed blank background spectrum data is specifically: Introducing a scattering light proportion factor k scat The fluorescence intensity is corrected, and the scattering light proportion factor is the background intensity ratio of the pretreated blank background spectrum data and the detection data of the down sample to be detected below 400 nm. According to the scattered light proportion factor k scat The detection data of the to-be-detected down sample is subjected to dynamic background deduction to obtain corrected fluorescence data of the to-be-detected down sample. The dynamic background subtraction on the Raman spectrum data of the eider sample to be detected based on the preprocessed blank background spectrum data is specifically: In the interval outside the iron characteristic peak, the area with obvious matrix characteristic peak is selected to fit the polynomial baseline data B matrix ( λ ) background subtraction is performed in the Raman spectrum data of the to-be-detected down sample to obtain the corrected Raman spectrum data of the to-be-detected down sample. 7.The eider iron detection method based on spectral analysis according to claim 1, characterized in that: The S4 is specifically: S4.1: extracting fluorescence features and Raman features from the corrected data of the eider sample to be detected; S4.2: establishing a deep learning model architecture; S4.3: training the deep learning model architecture; S4.4: inputting the fluorescence features and Raman features of the eider sample to be detected into the deep learning model architecture to obtain an iron detection result of the eider sample to be detected.

8. A system for detecting iron in down based on spectral analysis, characterized by, The system adopts the method for detecting iron in down according to spectral analysis in any one of claims 1-7, and the system comprises: a sample pretreatment module, configured to pretreat the down to be detected to obtain a down sample to be detected; a detection data acquisition module, configured to perform dual-mode spectral synergistic excitation on the down sample to be detected and collect signals to obtain detection data of the down sample to be detected; a detection data correction module, configured to perform background correction on the detection data of the down sample to be detected to obtain corrected data of the down sample to be detected; and an iron detection module, configured to obtain an iron detection result of the down sample to be detected according to the corrected data of the down sample to be detected.

Images