A method and device for detecting the quality index of rice throughout the whole process

Abstract

The present application relates to the technical field of fine rice rate detection, in particular to a rice quality index whole-process detection method and device, comprising the following steps: according to the rice quality index detection whole process, image recognition is carried out on each detection link, and according to the images collected in each link, the effective feed particle index, the imperfect grain proportion index, the rice quality classification result, the head rice rate index and the yellow rice proportion are calculated respectively. The present application identifies the position and shape of each rice kernel through high-precision image processing, screens the effective feed particle index input into the hulling chamber, the infrared transmission intensity detection of brown rice directly reflects the structural characteristics of rice kernels, the real-time quality evaluation method optimizes the conversion process from brown rice to fine rice, ensures the optimization of rice quality and yield, the real-time monitoring of surface brightness and texture changes during the milling process helps to adjust the milling process, and the dynamic data analysis optimizes the product quality and improves the appearance uniformity.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of milled rice rate detection, in particular to a rice quality index full-process detection method and device. BACKGROUND

[0002] The rice milled rice rate detection technical field includes related technologies for detecting and evaluating various index parameters involved in the rice milling process. The core content of this technical field mainly focuses on the quality changes of rice from raw grain state to finished product rice at each stage, including the acquisition and calculation of key quality indicators such as rice moisture content determination, impurity proportion detection, rice breakage rate analysis, finished product milled rice rate calculation, etc. The entire technical system covers the extraction process of multiple physical quantities such as sample collection, image recognition, weight and volume determination, and particle counting, and is widely used in agricultural grain storage, processing enterprise quality control, and grain detection agencies, etc. It has important significance for ensuring the consistency and standardization of rice processing products.

[0003] Among them, the rice quality index full-process detection method refers to a method for continuously measuring and recording multiple key quality parameters of rice from entering the factory to leaving the factory. This patent subject mainly aims at the problems of single-point detection, discontinuous sampling, incomplete index coverage, etc. in rice quality management, and detects specific index items such as particle number statistics, moisture weight ratio determination, whole milled rice rate calculation, and chaff residue rate identification by setting fixed points. It uses image recognition technology for particle screening, measures precision through a weighing device, and combines with a rice grain shape database for shape comparison, so as to systematically obtain the required quality information, and realize full-link quality detection from the stages of entering the factory inspection, pre-processing screening, and post-milling finished product evaluation.

[0004] The existing technology still has deficiencies in automation and real-time monitoring, especially in the limited ability to accurately measure rice grain spacing and instantly evaluate structural characteristics. The degree of automation and data processing capability is limited by the limitations of equipment and algorithms, which cannot perform fine quality sorting, resulting in hindered improvement of processing efficiency and product quality. Data processing relies on batch processing, lacking support for dynamic adjustment of production lines, which is particularly evident in large-scale production environments, unable to adjust processing parameters according to real-time data, thereby increasing production costs. SUMMARY

[0005] The present application provides a rice quality index full-process detection method and device to solve the technical problems in the prior art.

[0006] The technical solution of the present application to solve the above technical problems is as follows: a rice quality index full-process detection method, comprising the following steps:

[0007] In one aspect, S1: collecting the images of the input particles by the infrared camera, screening the rice particles according to the images of the input particles, extracting the rice texture and contour coordinates according to the surface images of the rice particles, identifying the position and shape of the rice and comparing the sample rice texture to obtain the effective input particle index of the input husking chamber;

[0008] S2: performing image recognition on the brown rice particles output by the husking chamber, detecting the infrared transmission intensity of the brown rice, calculating the average transmission intensity, comparing the transmission intensity of the sample brown rice with the average value, analyzing the structural characteristics of the brown rice, generating a structural integrity rating, calculating the perfect particle ratio index and the imperfect particle ratio index according to the proportion of the transmission intensity meeting the standard particles, and generating the husked rate index;

[0009] S3: grinding the brown rice and performing image recognition on the milled rice, identifying the contour image of the single rice, calculating the contour defect ratio, and classifying the rice into broken rice and milled rice to obtain the rice quality classification result;

[0010] S4: based on the rice quality classification result, introducing the rice into the corresponding weighing channel, recording the cumulative weight of the channel, comparing the total weight of the rice, calculating the weight proportion of the milled rice, and generating the whole milled rice rate index;

[0011] S5: performing image recognition on the milled rice obtained by grinding, identifying the proportion of yellow rice according to the image color, calculating the proportion of yellow rice in the total milled rice, combining the husked rate and whole milled rice rate indexes to calculate the final qualified rate, and outputting the proportion of yellow rice and the comprehensive quality rating.

[0012] In one aspect, the effective input particle index includes particle size distribution, arrangement density, and input uniformity, the perfect particle ratio index specifically includes structural continuity ratio, low transmission defect rate, and complete particle proportion, the rice quality classification result includes contour integrity level, surface damage degree, and grain type consistency index, the whole milled rice rate index specifically includes weight proportion, milled rice stable output ratio, and unit raw material milled rice output ratio, and the comprehensive quality rating includes color uniformity, milled rice purity, and yellow rice proportion level.

[0013] In one aspect, the step of the effective input particle index of the input husking chamber is specifically:

[0014] S101: collecting the images of the input particles by the infrared camera, screening the rice particles according to the images of the input particles, detecting the edge contour of the rice image according to the surface images of the rice particles, calculating the pixel gradient change of the continuous area of the image, positioning the edge contour coordinates of the rice, measuring the texture interval distance of each rice, analyzing the texture interval change trend, and obtaining the edge parameters of the rice;

[0015] S102: Based on the set of rice edge parameters, the morphological characteristics of single rice grains are identified, the length-width ratio and surface area of standard rice grains are compared, the contour closure degree of each rice grain is judged, morphologically abnormal rice grains are identified, the rice grain morphology distribution interval is analyzed, and the rice grain morphology parameters are generated.

[0016] S103: Based on the rice grain morphology parameters, the number of rice grains that meet the standard is screened, and the effective feed grain index of the input hulling chamber is obtained.

[0017] In one aspect, the step of structure integrity rating is specifically:

[0018] S201: Based on the effective feed grain, the infrared transmission device is used to perform transmission detection on the milled rice after hulling, the transmission data of each sample of milled rice is collected, the energy change of the transmission light passing through the milled rice is analyzed, the transmission intensity value of single milled rice is calculated, and the milled rice transmission intensity feature is generated.

[0019] S202: Based on the milled rice transmission intensity feature, the average value of the transmission intensity of the milled rice is calculated, compared with the transmission data of single milled rice, the deviation degree of the transmission value is analyzed, the milled rice data with abnormal transmission value is screened, the numerical range of the transmission intensity abnormality is judged, and the transmission intensity deviation value is generated.

[0020] S203: Based on the milled rice transmission deviation value, the transmission uniformity of the milled rice is analyzed, the distribution of the transmission deviation is calculated, the structural characteristics of the light intensity abnormal milled rice are judged, the milled rice meeting the standard structure is screened, the structure integrity rating is divided, and the structure integrity rating is generated.

[0021] In one aspect, the transmission intensity value of single milled rice is calculated by the formula:

[0022] ;

[0023] The milled rice transmission intensity feature is generated.

[0024] Among them, represents the transmission intensity value of single milled rice, represents the incident light intensity, represents the optical absorption coefficient of the milled rice, represents the thickness of single milled rice, represents the transmission and scattering light intensity, represents the light scattering correction parameter, represents the reference thickness.

[0025] In one aspect, the S3 further comprises:

[0026] Based on the structural integrity rating, the milled rice image is denoised, the surface characteristics of the brown rice are analyzed, the brightness fluctuation and the texture interval are calculated, the milling sample is called, the data deviating from the milling range is identified, the uniformity of the milling is judged, and the milling quality analysis result is obtained. Based on the milling quality analysis result, the sample is transmitted to the image acquisition area.

[0027] In one aspect, the step of the rice grain quality classification result is specifically:

[0028] S301: Based on the structural integrity rating, the image data of the brown rice after milling is called, and the image is denoised, the pixel information of the rice surface area is extracted, the spatial distribution of the pixel gray value is analyzed, the area with high surface pixel brightness is screened, the brightness change value of the area is calculated, and the rice surface brightness fluctuation data is generated;

[0029] S302: Based on the rice surface brightness fluctuation data, the distribution of the pixel area with high brightness change is analyzed, the interval between adjacent brightness extreme points is measured, the offset amplitude of the brightness mutation interval is calculated, and the texture difference of the rice surface is judged, and the rice texture interval difference is generated;

[0030] S303: Based on the rice surface brightness fluctuation data and the rice texture interval difference, the rice milling sample data is called, the data points exceeding the conventional milling interval are screened, the surface brightness of each rice is compared with the texture change range, the smoothness of the rice surface is analyzed, the milling uniformity offset is calculated, and the milling quality analysis result is obtained;

[0031] S304: Based on the milling quality analysis result, the rice sample is transmitted to the image acquisition area by using the conveying device, the contour boundary of the single rice grain is detected, the contour edge pixel points are screened and connected to form a closed contour, the contour overlap area is calculated for the overlapping area, the overlapping part is morphologically segmented, and the rice grain contour feature is obtained;

[0032] S305: Based on the rice grain contour feature, the pixel distribution of the rice surface is analyzed, the surface spot area is screened according to the pixel brightness change, and the pixel fracture along the main axis direction is detected, the number and length of the cracks are calculated, the structural integrity of the rice surface is analyzed, and the surface damage data is obtained;

[0033] S306: Based on the surface damage data, the surface area of the rice grain is measured and compared with the spot and crack area, the total area ratio of the damaged area is calculated, the rice grain is classified according to the contour damage ratio, the surface smooth and complete is classified as milled rice, and the damaged rice grain is classified as broken rice, and the rice grain quality classification result is obtained.

[0034] In one aspect, the offset amplitude of the brightness mutation interval is calculated by using the formula:

[0035] ;

[0036] It also determines the differences in surface texture of rice grains and generates rice grain texture interval differences;

[0037] in, The offset magnitude representing the range of brightness abrupt changes. Representing the The brightness of a point with a maximum brightness. Representing the The brightness of the minimum brightness point This represents the pixel interval between adjacent extreme brightness points. This represents the total number of detected extreme brightness point pairs.

[0038] On the one hand, the steps for determining the head rice yield are as follows:

[0039] S401: Based on the rice grain quality classification results, automatically allocate the rice grain conveying path, and import broken rice and refined rice into the corresponding weighing channels respectively, record the cumulative weight data of each channel in real time, and obtain the total amount of rice grains of each type;

[0040] S402: Call the total amount of each type of rice grain, call the total weight of the rice grain, compare the cumulative weight of each type of rice grain with the total weight, calculate the weight ratio of broken rice and polished rice respectively, combine with the total amount of paddy rice, determine the average proportion of polished rice in the total amount of paddy rice, and generate the whole rice rate index.

[0041] On the other hand, a device for the whole-process detection of rice quality indicators is provided. This device is used to execute the above-mentioned whole-process detection method for rice quality indicators. The device includes:

[0042] The infrared image acquisition module acquires images of the feed particles through an infrared camera, filters rice particles based on the feed particle images, extracts rice texture and contour coordinates based on the surface images of the rice particles, identifies the position and shape of the rice particles, compares the texture of sample rice particles, and obtains the effective feed particle index.

[0043] The transmission intensity detection module detects the infrared transmission intensity of brown rice based on the effective feed particle index, calculates the average transmission intensity, compares the transmission intensity of the sample brown rice with the average value, analyzes the structural characteristics of brown rice, generates a structural integrity rating, and calculates the percentage of perfect grains and the percentage of imperfect grains based on the percentage of particles that meet the transmission intensity standard, thereby generating the brown rice yield index.

[0044] The rice grain quality classification module performs image recognition on the milled rice grains, identifies the outline image of a single grain, calculates the proportion of outline defects, and classifies them into broken rice and polished rice, thus obtaining the rice grain quality classification result.

[0045] A weighing and distributing module guides the rice grains into corresponding weighing channels based on the rice grain mass classification result, records the cumulative weight of the channels, compares the total weight of the rice grains, calculates the weight proportion of the milled rice, and generates the whole milled rice rate index;

[0046] A qualified rate calculation module performs image recognition on the milled rice obtained by milling, identifies the proportion of yellow rice based on the image color, counts the proportion of the yellow rice in the total milled rice, combines the husked rice rate and the whole milled rice rate index to calculate the final qualified rate, and outputs the proportion of the yellow rice and the comprehensive quality rating.

[0047] The technical scheme provided by the embodiment of the application has at least the following beneficial effects:

[0048] The position and shape of each rice grain are recognized through high-precision image processing, the effective feed particle index of the input hulling chamber is screened, the infrared transmission intensity of the husked rice directly reflects the structural characteristics of the rice grains, the real-time quality evaluation method optimizes the conversion process from husked rice to milled rice, ensures the optimization of rice quality and yield, and the real-time monitoring of the surface brightness and texture changes during the milling process helps to adjust the milling process, the product quality is optimized through dynamic data analysis, the appearance uniformity is improved, the automatic rice grain quality classification and weight proportion calculation not only improve the calculation accuracy of the whole milled rice rate, but also speed up the processing speed of the production line, so that the overall processing efficiency and product quality are significantly improved. BRIEF DESCRIPTION OF DRAWINGS

[0049] Figure 1 The main step flowchart of the application is shown in the figure;

[0050] Figure 2 The device diagram of the application is shown in the figure. DETAILED DESCRIPTION

[0051] The technical scheme in the embodiments of the application will be described clearly and completely in the description of the embodiments of the application in combination with the drawings. Obviously, the described embodiments are only some of the embodiments of the application, not all the embodiments. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the application.

[0052] In the description of the application, the terms "first" and "second" are used for description purposes only, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" can explicitly or implicitly include one or more of the features. In the description of the application, the meaning of "multiple" is two or more, unless otherwise specifically limited.

[0053] In the description of the present application, the term "for example" is used to mean "serving as an example, instance, or illustration." Any embodiment described as "for example" in the present application is not necessarily to be construed as preferred or advantageous over other embodiments. The following description is presented to enable any person skilled in the art to make and use the application. In the following description, for the purposes of explanation, numerous specific details are set forth in order to provide a thorough understanding of the present application. It will be apparent, however, to one skilled in the art that the present application can be practiced without using these specific details. In other instances, well-known structures and processes are not elaborated upon in order to avoid unnecessary detail and so as to not obscure the present application. Thus, the present application is not intended to be limited by the embodiments shown, but is to be accorded the widest scope consistent with the principles and features disclosed herein.

[0054] Embodiment 1

[0055] In order to make the technical problems, technical solutions and advantages to be solved by the present application clearer, specific embodiments will be described in detail below with reference to the accompanying drawings.

[0056] The embodiment of the present application provides a whole-process detection method for rice quality indexes, as shown in the figure, comprising the following steps: Figure 1

[0057] S1: acquiring the image of the input grain particles through an infrared camera, screening the rice particles according to the image of the input grain particles, extracting the texture and contour coordinates of the rice according to the surface image of the rice particles, identifying the position and shape of the rice and comparing the sample rice texture, and obtaining the effective input grain particle index of the input hulling chamber;

[0058] S2: performing image recognition on the brown rice particles output by the hulling chamber, detecting the infrared transmission intensity of the brown rice, calculating the average transmission intensity, comparing the transmission intensity of the sample brown rice with the average value, analyzing the structural characteristics of the brown rice, generating a structural integrity rating, calculating the perfect grain ratio index and the imperfect grain ratio index according to the proportion of the transmission intensity qualified particles, and generating the husked rate index;

[0059] S3: grinding the brown rice and performing image recognition on the milled rice, identifying the contour image of the single rice, calculating the contour defect ratio, and classifying the milled rice into broken rice and polished rice, and obtaining the milled rice quality classification result;

[0060] S4: based on the milled rice quality classification result, introducing the milled rice into the corresponding weighing channel, recording the cumulative weight of the channel, comparing the total weight of the milled rice, calculating the weight proportion of the polished rice, and generating the whole polished rice rate index;

[0061] S5: performing image recognition on the polished rice obtained by milling, identifying the proportion of yellow rice according to the image color, calculating the proportion of yellow rice in the total polished rice, combining the husked rate and the whole polished rice rate index to calculate the final qualified rate, and outputting the proportion of yellow rice and the comprehensive quality rating. ​

[0062] The effective feed grain indicators include grain size distribution, arrangement density, input uniformity, and perfect grain proportion indicators, and the perfect grain proportion indicators include structure continuity proportion, low transmission defect rate, and complete grain proportion. The rice grain quality classification results include contour completeness level, surface damage degree, and grain type consistency indicators. The head rice rate indicators include weight proportion, head rice stable output ratio, and unit raw material head rice output ratio. The comprehensive quality rating includes color uniformity, head rice purity, and yellow rice proportion level.

[0063] In one aspect, the step of obtaining the effective feed grain indicators of the input hulling chamber includes:

[0064] S101: acquiring a feed grain image by using an infrared camera, screening rice grains according to the feed grain image, detecting a rice image edge contour according to a surface image of the rice grains, calculating a pixel gradient change of a continuous area of the image, locating a rice edge contour coordinate, analyzing a texture interval change trend, and obtaining a rice edge parameter;

[0065] Based on the rice surface image acquired by the infrared camera, specifically, a resolution of 2048x1536 pixels is set for the infrared camera, which is vertically aligned with a conveyer belt moving at a constant speed (for example, 5 cm / s) below. The conveyer belt is single-layered and flat with rice grains. The camera is 30 cm away from the surface of the conveyer belt. An 850 nm infrared light source is used to uniformly irradiate the acquisition area. The surface image data of the rice grains are captured frame by frame. An edge detection operator is applied to each frame of image data, for example, a Canny operator is used, with a low threshold of 50 and a high threshold of 150, to preliminarily identify the edge pixel points. These edge pixel points are connected to form an initial contour. Then, the difference in infrared grayscale value between each pixel point and its adjacent pixels (for example, eight adjacent areas) in the image is calculated to obtain a pixel gradient amplitude graph. The pixel points on the initial contour are traversed, and their gradient amplitude is analyzed. The pixel point sequence with the most significant gradient amplitude change is located as the final rice edge contour coordinate. For example, for a rice grain, its contour coordinate sequence is obtained as follows: , subsequently, within the region inside the located paddy contour, along the long axis of the paddy, texture features are extracted using the gray level co-occurrence matrix analysis method, periodic changes in surface texture (such as the grooves of the rice coat) are identified and located, the pixel distance between adjacent significant texture features (such as local gray minimum points) is calculated, and the physical interval distance is converted according to the pre-calibrated pixel-to-physical size ratio (for example, 0.05 millimeters per pixel), for example, the measured texture interval distance sequence on a certain paddy is {0.82 millimeters, 0.85 millimeters, 0.81 millimeters, 0.88 millimeters}, further, the measured texture interval distance data is analyzed, the mean (for example, (0.82+0.85+0.81+0.88) / 4=0.84 millimeters) and the standard deviation (for example, 0.029 millimeters) are calculated, the standard deviation is compared with the preset stability threshold (for example, 0.05 millimeters, which is set based on the analysis of the statistical distribution of the texture interval standard deviation of a large number of standard paddy samples, taking the 80% percentile of the distribution), if the standard deviation is less than 0.05 millimeters, it is judged that the texture interval change trend is stable, otherwise it is not stable, the edge contour coordinate sequence, the average texture interval distance, the texture interval standard deviation, and other information of the paddy are summarized to obtain the edge parameters of the paddy.

[0066] S102: Based on the paddy edge parameter set, the morphological features of single paddy are identified, the length-width ratio and surface area of standard paddy are compared, the closure degree of each paddy contour is judged, the morphologically abnormal paddy is identified, the paddy morphological distribution interval is analyzed, and the paddy morphological parameters are generated.

[0067] Based on the paddy edge parameter set obtained by S101, the parameter set includes the edge contour coordinate sequence of multiple paddy, the average texture interval distance, the texture interval standard deviation, etc., first, the edge contour coordinate sequence of single paddy is calculated , the minimum circumscribed rectangle is calculated, the length (for example, 10.5 millimeters) and the width (for example, 3.2 millimeters) of the paddy are obtained, the length-width ratio (for example, 10.5 / 3.2≈3.28) is calculated, and the contour perimeter and the area enclosed by the contour (for example, the area is calculated to be 25.8 square millimeters by Green's formula or pixel point counting method) of the paddy are calculated, the calculated length-width ratio and area are compared with the reference values in the pre-stored standard paddy morphological database, for example, the length-width ratio range of standard "medium indica rice" is set to [3.0, 3.5], and the area range is set to [24.0, 28.0] square millimeters (these ranges are obtained by measuring and statistical analysis of more than 1000 particles of "medium indica rice" samples conforming to the national standard, taking the 95% confidence interval), if the length-width ratio and area of a certain paddy and If the area is measured in square millimeters, its morphological characteristics initially meet the standards. Next, the closure of the rice grain's outline is evaluated, specifically by checking the edge outline coordinate sequence. First and last coordinates and Euclidean distance between Set a closing threshold (For example Set to 2 pixel units, determined based on image resolution and acceptable degree of contour breakage. If the outline is closed, it is considered to have good closure; otherwise, it is judged to have an incomplete outline. Then, based on the aspect ratio... ,area The morphological parameters of rice grains are comprehensively identified based on three conditions: whether the morphology is within the standard range, whether the contour closure is good, etc. For example, if a rice grain has an aspect ratio of 2.5 (less than 3.0), an area of ​​30.0 square millimeters (greater than 28.0), or an unclosed contour, it is marked as morphologically abnormal. The morphological parameters (aspect ratio and area) of all rice grains in the batch are statistically analyzed to determine their distribution. For example, histograms of aspect ratio and area are drawn to determine the main distribution range. For example, it is found that 85% of the rice grains have an aspect ratio between 3.1 and 3.4 and an area between 25.0 and 27.5 square millimeters. The parameters of rice grains that meet the standard morphology (aspect ratio and area within the standard range and contour closed) are collected to generate rice morphological parameters.

[0068] S103: Based on the rice morphology parameters, screen the number of rice grains that meet the standards to obtain the effective feed particle index for input into the rice hulling chamber.

[0069] On the one hand, the specific steps of structural integrity rating are as follows:

[0070] S201: Based on the effective feed particles, use an infrared transmission device to perform transmission detection on brown rice after hulling, collect transmission data of each grain of brown rice, analyze the energy change of transmitted light passing through the brown rice, calculate the transmission intensity value of a single grain of brown rice, and generate brown rice transmission intensity characteristics.

[0071] Based on the effective feed particles obtained in S103, which are normal-shaped and have appropriate spacing during the feeding process, the red rice is detected one by one after being processed by the hulling process using an infrared transmission device. The device includes an infrared light source (for example, wavelength 940 nm, power adjustable range 1-10 mW, set to 5 mW) and an infrared detector (for example, InGaAs photodiode). The red rice passes through the fixed light path between the light source and the detector, and the infrared light beam emitted by the light source is received by the detector after penetrating the red rice. The signal intensity data of the detector when each red rice sample passes through the light path is collected, which reflects the energy of the transmitted light, i.e. the transmitted light intensity. The change in energy of the transmitted light when it passes through a single red rice grain is analyzed, mainly recording the lowest signal value measured by the detector when the red rice completely blocks the light path, which is converted into the transmitted light intensity To eliminate the influence of background light and equipment noise, the background light intensity is measured when there is no red rice passing through And the dark current intensity is measured in the area where the light source does not shine on the detector when red rice is passing through The original measurement signal is corrected: Then the transmitted intensity value of a single red rice grain is calculated Using the formula: This formula is used to calculate the theoretical transmitted intensity value of a single red rice grain, where represents the transmitted intensity value of a single red rice grain (unit: mW / cm²), represents the incident light intensity (unit: mW / cm²), which is the measured intensity when the infrared light source directly shines on the detector, and needs to be calibrated in advance, for example, the measured mW / cm², represents the optical absorption coefficient of the red rice for this wavelength of infrared light (unit: cm -1 ), which is related to the internal composition of the red rice (such as moisture and starch structure), and can be obtained by fitting the transmission experiment on standard red rice samples with known composition, for example, for a specific variety of red rice, the coefficient is set to cm -1 , represents the actual thickness of a single red rice grain (unit: cm), which can be measured by a laser thickness meter or image analysis (based on side view) before or at the same time as the transmission detection, for example, the thickness of a certain red rice grain is measured to be mm=0.28cm, represents the scattered light intensity (unit: mW / cm²) generated during the transmission process, which part of the light passes through the red rice but deviates from the direct transmission path, and its intensity is affected by the scattering characteristics of the red rice surface and internal structure. It can be estimated by measuring the scattered light collected by the integrating sphere around the detector, for example, the estimated mW / cm², This represents a dimensionless light scattering correction parameter used to correct for the influence of different thicknesses of brown rice on the scattering effect. This parameter needs to be obtained through experimental calibration using samples of different thicknesses. For example, setting... , The representative thickness (unit: cm) is usually set as the average thickness of the batch or variety of brown rice, for example, by obtaining the average thickness through pre-measurement of a large number of samples. mm = 0.25 cm. Substitute these parameters into the formula to calculate: Calculate the exponential term: Calculate the correction term in the denominator: Calculate the square root term: , Substitute the formula: mW / cm², this measurement and calculation process is performed on each valid grain of brown rice that passes the test, and the transmittance value of each grain of brown rice is recorded. This creates a list of transmission intensity values, such as {7.52, 7.61, 7.48, 7.55, 7.40, 7.65, ...}, which constitutes the transmission intensity characteristics of brown rice. The formula's logic and purpose: This formula aims to estimate the intensity of infrared light transmitted through a single grain of brown rice by comprehensively considering both absorption and scattering, the two main attenuation mechanisms. Basic Part Derived from Beer-Lambert's law, it describes the exponential decay of light due to absorption in a homogeneous medium, where the absorption coefficient... and thickness This is a key factor. However, brown rice is not a uniform transparent medium and exhibits significant scattering; therefore, it is necessary to introduce... A term is used to compensate for the contribution of scattered light outside the direct transmission path (using the square root form is a simplified model based on the intensity of scattered light and the properties of the incident light and the medium). Meanwhile, the thickness of the brown rice is taken into account. It is not constant, and variations in thickness affect the relative contributions of absorption and scattering, thus introducing a correction term in the denominator. This item is obtained through dimensionless parameters. and relative thickness The total transmission intensity is adjusted by adjusting the thickness; the larger the thickness, the larger the correction term, and the relatively smaller the transmission intensity, and vice versa. The absolute value ensures that the final intensity is non-negative. The advantage of the formula is that it considers absorption (…) simultaneously. ),scattering( ) and the combined effect of thickness on these two effects ( Compared to using only Beer-Lambert's law, this method more accurately simulates the actual transmission intensity of infrared light passing through non-uniform, scattering brown rice particles, providing a more reliable physical basis for subsequently judging the internal structural integrity of brown rice based on transmission intensity. The calculation results... mW / cm2 indicates that for this grain of brown rice with a thickness of 0.28 cm, the theoretical transmission intensity value is about 7.52 mW / cm2 under the given incident light intensity, absorption coefficient, scattered light intensity and correction parameters, which is a specific example of the transmission intensity characteristics of the brown rice and will be used for subsequent deviation analysis and structural integrity rating.

[0072] S202: Based on the transmission intensity characteristics of the brown rice, calculate the average transmission intensity of the brown rice, compare it with the transmission data of single brown rice, analyze the deviation of the transmission value, screen the brown rice data with abnormal transmission value, determine the range of abnormal transmission intensity value, and generate the transmission intensity deviation value;

[0073] Based on the transmission intensity characteristics of the brown rice generated in S201, i.e. the set of transmission intensity values calculated for each grain of brown rice {7.52, 7.61, 7.48, 7.55, 7.40, 7.65, …}, first calculate the average transmission intensity of the brown rice sample , for example, if the sample contains 100 grains of brown rice, the sum of the transmission intensity values is 753.5 mW / cm2, then the average value is mW / cm2, take this average value as the reference of the transmission intensity of the batch of brown rice, then compare the transmission intensity value of each grain of brown rice in the data set with the average value , calculate the transmission intensity deviation of each grain of brown rice , for example, for mW / cm2, the deviation is mW / cm2, for mW / cm2, the deviation is mW / cm2, next, analyze the deviation of these deviation values, set a deviation threshold to screen the brown rice data with abnormal transmission value, the setting of the threshold may be based on a certain percentage of the average value, or based on the standard deviation of the sample deviation value, for example, calculate the standard deviation of all deviation values , assuming that mW / cm2 is calculated, set the abnormal criterion as the absolute value of the deviation exceeding times the standard deviation, for example, take , then the abnormal threshold is mW / cm2 (the choice of means that it covers about 95% of the normal fluctuation range, which is based on statistical principles and experience requirements for the uniformity of the internal structure of brown rice), judge whether the absolute value of the deviation of each grain of brown rice is greater than mW / cm2, if , the transmittance intensity of the brown rice is determined to be abnormal, for example, , , determined to be normal, , , also determined to be normal, assuming that there is another brown rice mW / cm2, the deviation mW / cm2, , determined to be abnormal (low), and assuming that there is another brown rice mW / cm2, the deviation mW / cm2, , determined to be abnormal (high), the transmittance intensity value of all brown rice determined to be abnormal is recorded, and the numerical range of the transmittance intensity abnormality is determined, for example, in this batch, the range of abnormal low is mW / cm2, and the range of abnormal high is mW / cm2, each brown rice is associated with its corresponding deviation value to generate a set of transmittance intensity deviation values.

[0074] Table 1: Example of Brown Rice Transmittance Intensity and Deviation

[0075]

[0076] Note: Table 1 assumes incident light intensity , absorption coefficient , scattered light intensity , correction coefficient , reference thickness . Average transmittance intensity .

[0077] As shown in Table 1, the measured thickness of 6 brown rice samples, the transmittance intensity value calculated according to the formula in S201 , the deviation value calculated relative to the batch average value mW / cm2, , and the result of determining whether the transmittance intensity of the brown rice is abnormal according to whether the absolute value of the deviation exceeds the threshold value mW / cm2are listed. Table 1 shows the specific process of abnormal screening based on transmittance intensity characteristics.

[0078] S203: Based on the transmittance deviation value of the brown rice, analyze the transmittance uniformity of the brown rice, calculate the distribution of the transmittance deviation, judge the structural characteristics of the light intensity abnormal brown rice, screen the brown rice that meets the standard structure, divide the structural integrity level, and generate the structural integrity rating.

[0079] On the one hand, the transmission intensity value of a single grain of brown rice is calculated using the formula:

[0080] ;

[0081] Generate the transmission intensity characteristics of brown rice;

[0082] in, The transmittance value represents the intensity of a single grain of brown rice. Represents the intensity of incident light. The optical absorption coefficient representing brown rice. Represents the thickness of a single grain of brown rice. Represents the intensity of transmitted and scattered light. Represents the light scattering correction parameter. This represents the reference thickness.

[0083] Based on the set of brown rice transmission deviation values ​​{(-0.015),(0.075),(-0.135),(0.115),…} generated by S202, the transmission uniformity of the entire batch of brown rice is first analyzed. This can be achieved by calculating the distribution of the deviation values, for example, by calculating the standard deviation of the deviation values. (Already calculated in S202, it is 0.04 mW / cm²) or calculate the coefficient of variation. (For example (i.e., 0.53%), a small standard deviation or coefficient of variation indicates that the transmission intensity of the entire batch of brown rice is relatively uniform. Next, those transmission intensity anomalies identified in S202 (i.e., The structural characteristics of brown rice (such as brown rice number 5) are generally considered to indicate abnormally low transmission intensity. A deviation of -0.135 may indicate the presence of cracks, voids, or impurities within the brown rice. These defects increase light scattering or absorption, leading to reduced transmission intensity. An abnormally high transmission intensity (e.g., brown rice number 6) could indicate the presence of cracks, voids, or impurities within the brown rice. A deviation of +0.115 might mean that the brown rice is thinner than average, or that its internal structure is unusually dense, leading to reduced scattering (the latter being less common). Then, brown rice conforming to a standard structure is selected based on the transmission intensity deviation value. A structural integrity standard is set, for example, the absolute value of the transmission intensity deviation... exist Brown rice with a deviation of less than 0.08 mW / cm² is considered structurally intact. Brown rice is judged to have incomplete structure or suspected defects. Furthermore, the structural integrity can be graded based on the magnitude of the deviation, with multiple thresholds set, for example: Grade A (Excellent). (Right now Grade B (Good): (Right now ) ; Grade C (Medium) : (i.e. ) ; Grade D (Poor) : (i.e. ), the setting of these thresholds is based on a large amount of experimental data, the performance of brown rice in subsequent milling and finished product quality detection (such as broken rice rate, crack rate) is associated with different transmission deviation ranges, for example, for brown rice No. 1 (deviation -0.015), it belongs to grade A, for brown rice No. 2 (deviation +0.075), it belongs to grade B, for brown rice No. 5 (deviation -0.135), due to , it belongs to grade D, for brown rice No. 6 (deviation +0.115), due to , it belongs to grade C, a structural integrity grade (A, B, C, D) is assigned to each brown rice, and this set containing grade information is the final generated structural integrity rating.

[0084] In one aspect, S3 further comprises:

[0085] Based on the structural integrity rating, the polished rice image is denoised, the surface features of the brown rice are analyzed, the brightness fluctuation and the texture interval are calculated, the milling sample is called, the data deviating from the milling range is identified, the uniformity of the milling is judged, and the milling quality analysis result is obtained. Based on the milling quality analysis result, the sample is transmitted to the image acquisition area.

[0086] In one aspect, the step of rice quality classification result is specifically:

[0087] S301: Based on the structural integrity rating, the image data of the brown rice after milling is called, and the image is denoised, the pixel information of the surface area of the rice is extracted, the spatial distribution of the pixel gray value is analyzed, the area with obvious surface pixel brightness is screened, the brightness change value of the area is calculated, and the brightness fluctuation data of the surface of the rice is generated;

[0088] Based on the structural integrity rating (each brown rice has A, B, C, D grades) generated by S203, the optical image data of the polished rice formed after the milling process of these brown rice is called. These images are usually taken by a CCD or CMOS camera under visible light, for example, with a resolution of 1600x1200 pixels. First, the obtained polished rice image is denoised, and a Gaussian filter is applied, for example, with a standard deviation to smooth out sensor noise and slight surface irregularities. Then, the image segmentation technique (such as Otsu threshold method or color space analysis) is used to extract the surface area pixel information of each polished rice, and the background pixels are excluded. For all pixels in the surface area of each polished rice extracted, the spatial distribution of the gray value is analyzed, and the average value and the standard deviation of the pixel gray value in the area are calculated., then, the regions with prominent surface pixel brightness are screened, which can be achieved by setting a brightness threshold, for example, the pixel region with brightness value greater than is identified as a highlight region, or the pixel region with brightness value less than is identified as a dim region (threshold coefficient 1.5 is set based on statistical analysis of the brightness distribution of normal milled rice surface, aiming to capture significant local brightness anomalies), for example, the average gray value of certain milled rice is , the standard deviation is , then the pixel region with brightness greater than or less than is screened, the total area of these prominent regions is calculated as a proportion of the total area of the rice grain, and the difference between the average brightness of each prominent region and is calculated as the brightness change value of the region, for example, the average brightness of a highlight region is 205, then the brightness change value is , the average brightness of a dim region is 150, then the brightness change value is , the total identified prominent regions and their corresponding brightness change values are summarized to form the surface brightness fluctuation data of the rice grain.

[0089] S302: Based on the surface brightness fluctuation data of the rice grain, the distribution of the pixel regions with prominent brightness change is analyzed, the interval between adjacent brightness extreme points is measured, the offset amplitude of the brightness mutation interval is calculated, and the surface texture difference of the rice grain is determined to generate the rice grain texture interval difference;

[0090] Based on the surface brightness fluctuation data of the rice grain generated in S301, which contains the position, area and brightness change value information of each rice grain surface brightness prominent region, first, the distribution of these pixel regions with prominent brightness change on the surface of the rice grain is analyzed, for example, the centroid coordinates of these regions are calculated, whether they are concentrated or dispersed distribution, and whether they are arranged along a certain direction (such as the long axis of the rice grain), then, in order to quantify the roughness or irregularity of the surface texture, the interval between adjacent brightness extreme points needs to be measured, which is usually along the main axis direction of the rice grain, one or more brightness profile lines are extracted on the main axis of the rice grain, the profile line data is smoothed (for example, moving average), then the peak detection algorithm is used to find the positions and brightness values of the brightness maximum points and the brightness minimum points (for example, , is the total number of detected extreme points), the pixel interval between adjacent extreme points (one maximum and one minimum) is calculated, for example, the first extreme point pair: maximum at pixel position 50, minimum at pixel position 68, then Pixel, second extreme point pair: maximum value At pixel location 110, the minimum value is... At pixel position 135, then Pixels, assuming a total of 1000 pixels were detected Five sets of extreme points were obtained. The data is then used to calculate the offset magnitude of the brightness abrupt change interval. The formula used is: This formula calculates the weighted average offset amplitude of the brightness abrupt change interval, where The offset range representing the range of brightness abrupt changes (the unit depends on the brightness and distance units; here it is grayscale level·pixel). Representing the The brightness (grayscale level) of each maximum brightness point. Representing the The brightness (grayscale level) of the minimum brightness point. Representing the The pixel interval (unit: pixels) between adjacent extreme brightness points. The summation sign represents the total number of detected extreme brightness point pairs. Indicates all The extreme points are accumulated, and the numerator is used to calculate the average brightness of each pair of extreme points. (Dividing by 2 was omitted because it's a relative comparison) Multiply by the interval between them Then summing them up is equivalent to weighting the average brightness using intervals, reflecting the sum of the product of the overall brightness fluctuation amplitude and spatial frequency. The denominator is... It is the root mean square (RMS) of all intervals, used to normalize the weighted sum of the molecules, so that the result... It primarily reflects the relative fluctuation range of brightness, reducing dependence on the absolute length of rice grains. The absolute value ensures the result is non-negative. Use the example data above to perform the calculation: Calculate the numerator: Calculate the denominator: ,calculate According to the calculation The value determines the difference in surface texture of rice grains, and a texture difference threshold is set, for example... It was determined to have large texture differences (rough surface or uneven grinding). The texture difference is judged to be moderate. In this case, the texture difference is small (the surface is smooth). The rice grain texture was determined to have a large difference, and the calculated value was used to determine the difference. The value or its corresponding level (large / medium / small) is used as the result to generate the rice grain texture interval difference.

[0091] S303: Based on the rice surface brightness fluctuation data and the rice grain texture interval difference, the rice milling sample data is called, the data points exceeding the conventional milling interval are screened, the surface brightness of each rice grain is compared with the texture change range, the surface smoothness of the rice grain is analyzed, the milling uniformity deviation is calculated, and the milling quality analysis result is obtained;

[0092] Based on the rice surface brightness fluctuation data (including brightness standard deviation and brightness highlight area information) generated by S301 and the rice grain texture interval difference (quantified as value or level, as shown in Table 2) generated by S302, first, the pre-established rice milling sample database is called, which stores a large number of milled rice samples processed by standard milling procedures, these samples are classified according to variety, original brown rice quality (may include S203 structure rating information), milling time, etc., and their corresponding surface brightness standard deviation and texture interval difference statistical range (such as mean, standard deviation, normal interval) are recorded, for example, the database shows that for “first-class high-quality long-grain indica rice” with A-class structure integrity, the normal range of after standard milling is [5, 12], the normal range is [200, 350], then, the calculated of the current rice to be analyzed (for example, the of rice grain 1 and the of rice grain 3 in Table 2) and (for example, the of rice grain 1 and the of rice grain 3 in Table 2) are compared with the corresponding conventional milling interval in the database, and the data points exceeding the conventional milling interval are screened, in this example, assuming that rice grain 1 and rice grain 3 are both A-class brown rice for milling, the of rice grain 1 is within the range of [5, 12], but exceeds the range of [200, 350]; the of rice grain 3 exceeds the range of [5, 12], also exceeds the range of [200, 350], so the surface parameters of rice grain 1 and rice grain 3 are both determined to deviate from the conventional milling result, then, the surface brightness (represented by ) of each rice grain is compared with the texture change range (represented by ), the surface smoothness of the rice grain is analyzed, the smooth rice grain should have lower and lower value (such as rice grain 2 and rice grain 4 in Table 2), otherwise it indicates that the surface is rough or the milling is uneven (such as rice grain 1 and rice grain 3), then, the milling uniformity deviation is calculated, which can be defined as the parameter point The normalized distance to the center point of the normal range of its corresponding category in the database, or simply a statistical method. and The number of items out of range in the two indicators (0, 1, or 2) is used as the level of the offset. For example, grain 1 has one indicator ( The following parameters were found to be out of range: 1) Rice grain 3 had two out-of-range parameters, with an offset level of 2; Rice grains 2 and 4 had values ​​of 0. By analyzing the distribution of offset in the uniformity of rice grain grinding in a batch, such as calculating the proportion of rice grains with the average offset level or high offset levels (e.g., levels 1 and 2), the grinding quality analysis results for the entire batch were obtained. These results summarized the degree and proportion of deviation of rice grain surface parameters from the normal range, reflecting the uniformity and effectiveness of the grinding process.

[0093] S304: Based on the results of milling quality analysis, rice samples are transferred to the image acquisition area using a conveying device. The contour boundary of a single grain of rice is detected, contour edge pixels are selected and connected to form a closed contour. The overlapping area of ​​the contour is calculated for the overlapping area, and the overlapping part is morphologically segmented to obtain the contour features of the rice grain.

[0094] Based on the milling quality analysis result obtained in S303, which indicates which surface characteristics (smoothness, uniformity) of the rice grains may be abnormal, all rice samples (regardless of the milling quality analysis result) are transported to the next high-resolution image acquisition area using a conveying device (for example, a vibrating feeder combined with a narrow belt conveyor, ensuring that the rice grains advance in a single row, separated state), where another camera (for example, a line array CCD camera, combined with a conveyor belt at an appropriate speed, can obtain higher resolution images of the rice grains) is used to image the passing rice grains one by one, detect the contour boundaries of individual rice grains, and again use edge detection algorithms (such as Sobel or Canny operators, with threshold values adjusted according to the current imaging conditions) or region growing or active contour model-based methods to obtain more accurate contour pixel points. These contour edge pixels are screened and connected using a connection algorithm (such as eight-neighbor connection or connection based on the minimum perimeter principle) to form a closed rice grain contour line. During the conveying process, there may be contact or slight overlap between the rice grains. To address this situation, processing is required after contour detection to identify potential overlapping areas (usually manifested as contour depressions or irregular connection points). The contour overlap area of these suspected overlapping areas can be identified by comparing the expected area of a single rice grain (based on the average size) with the detected connected domain area, or by analyzing the curvature change of the contour to identify overlapping points. The confirmed overlapping parts are morphologically segmented, for example, using a watershed algorithm or a concave point-based segmentation method to separate the contours of the adhered rice grains, ensuring that each object corresponds to a complete rice grain. Finally, the accurate contour feature data of each separated rice grain is generated, including contour pixel coordinate sequences and basic geometric parameters (such as area, perimeter, major axis, minor axis) calculated therefrom.

[0095] S305: Based on the rice grain contour features, analyze the pixel distribution on the surface of the rice grain, screen the surface spot area according to the change in pixel brightness, and detect the pixel breakage along the main axis direction, calculate the number and length of cracks, analyze the structural integrity of the rice grain surface, and obtain surface damage data;

[0096] Based on the single rice grain contour features (precise contour coordinate sequences) obtained in S304, first, the pixel analysis of the rice grain surface area inside the contour is performed to check the distribution of pixel gray value. Using the similar method in S301, the surface spot area is screened according to the difference between the pixel brightness and the average brightness of its neighborhood pixels or global / local average brightness. For example, define a pixel, if its gray value is lower than the average value of its 8-neighborhood pixels by a threshold value (for example The grayscale level (the threshold is set based on the analysis of the contrast between typical rice grain spots (such as blemishes and black spots) and normal rice grain surfaces) is then marked as a spot pixel. Connected spot pixels constitute spot regions. The area of ​​each spot region is calculated. At the same time, pixel fracture along the main axis of the rice grain, i.e., cracks, is detected. This can be achieved by applying a line segment detection algorithm (such as Hough transform to detect straight line segments) or a specific crack detection filter (such as based on morphological operations or tensor voting) to the rice grain image. Dark or low-contrast line segments representing cracks in the image are identified. The number of each detected crack and its length are calculated (e.g., by measuring the pixel distance between the two ends of the crack and converting it to millimeters). Then, the spot and crack information are combined to analyze the structural integrity of the rice grain surface. For example, the total area of ​​spots and the total length of cracks on each grain of rice are counted. These measurement results are summarized to obtain the surface damage data of the rice grain. This data may be a vector or record containing the total area of ​​spots (e.g., 0.5 square millimeters) and the total length of cracks (e.g., 1.2 millimeters).

[0097] S306: Based on surface damage data, measure the surface area of ​​rice grains and compare it with the spots and cracks. Calculate the total area ratio of the defective areas and classify the rice grains according to the proportion of contour defects. Rice grains with smooth and intact surfaces are classified as refined rice, while rice grains with defects exceeding the limit are classified as broken rice, thus obtaining the rice grain quality classification results.

[0098] Rice grain surface damage data (including total spot area) obtained based on S305. and total crack length And the surface area in the rice grain profile features obtained by S304. First, measure or call the total surface area of ​​the rice grain. (For example (square millimeters), then, compare the surface damage data with the total area to calculate the proportion of the total area of ​​the defective region. For cracks, their length can be used. Multiply by an assumed average width (For example The area is estimated using millimeters (based on microscopic observation settings). (For example (square millimeters), then the total defect area (For example (square millimeters), calculate the percentage of the defective area. (For example (i.e., 2.34%), based on this proportion of contour loss. The rice grains are then sorted, and a sorting threshold is set. (For example This threshold is set according to the definitions of polished rice and broken rice in national or commercial standards, and is usually based on the allowable content of imperfect grains. ( e.g. 2.34% 5%), and the grain itself is intact (not broken, which can be judged by the integrity of the profile and the aspect ratio), then the grain is judged to be “milled rice”, if , or the grain itself is broken (e.g. less than half the length of a normal grain), then it is classified as “broken rice”, and this classification process is performed for all detected grains, resulting in the final grain quality classification result, i.e. each grain is labeled as “milled rice” or “broken rice”.

[0099] In one aspect, the offset amplitude of the brightness mutation interval is calculated, using the formula:

[0100] ;

[0101] and the difference in grain surface texture interval is determined, generating the grain texture interval difference;

[0102] wherein, represents the offset amplitude of the brightness mutation interval, represents the brightness of the th brightness maximum point, represents the brightness of the th brightness minimum point, represents the pixel interval between adjacent brightness extreme points, represents the total number of detected pairs of brightness extreme points.

[0103] In one aspect, the steps of the milled rice rate index are as follows:

[0104] S401: Based on the grain quality classification result, automatically assign the grain conveying path, and guide the broken rice and milled rice into the corresponding weighing channels respectively, and record the cumulative weight data of each channel in real time to obtain the total amount of each type of grain;

[0105] Based on the grain quality classification result obtained in S306 (each grain is labeled as “milled rice” or “broken rice”, as shown in Table 3), the control device automatically assigns the subsequent conveying path of each grain according to this classification label, for example, a fast-responding sorting device (such as a pneumatic air valve or a fork mechanism) is set at the end of the conveyor belt, when the detection device identifies a grain as “milled rice” (such as grains 1, 2, and 5 in Table 3), the sorting device does not act or guides it to path A, when it identifies it as “broken rice” (such as grains 3 and 4 in Table 3), the sorting device acts (e.g. blows air or moves), guiding it to path B, path A leads to a milled rice collection container, and path B leads to a broken rice collection container, a weighing sensor (such as a high-precision electronic scale or a dynamic weighing module) is installed below each collection container or on the conveying path, which monitors and records the cumulative weight data of the grains flowing into the corresponding channel in real time, for example, the weighing sensor reading of the milled rice channel starts from 0 and accumulates, recording as The weighing sensor reading in the broken rice channel also accumulates from 0 and is recorded as follows: As rice grains are continuously sorted and transported, these two weight data will be continuously updated to obtain the total amount (in weight) of each type of rice grains (polished rice and broken rice) in the current processing batch (or within a certain period of time).

[0106] S402: Call the total amount of rice grains for each type, call the total weight of rice grains, compare the cumulative weight of each type of rice grains with the total weight, calculate the weight ratio of broken rice and polished rice respectively, combine with the total amount of paddy rice, determine the average proportion of polished rice in the total amount of paddy rice, and generate the head rice rate index.

[0107] The total amount of each type of rice grain obtained by calling S401, i.e., the cumulative weight of polished rice. (For example (kg) and total weight of broken rice (For example (kg), calculate the total weight of the rice grains after processing. (For example (kg), compare the cumulative weight of each type of rice grain with the total weight, and calculate the weight ratio of broken rice to polished rice, and the broken rice rate. (For example ), rice milling rate (For example These two percentages reflect the product composition after milling and sorting, and then need to be combined with the total amount of paddy rice in the original processing. To calculate the head rice yield, assume the total amount of rice processed in this batch is... The head rice yield is calculated by measuring kilograms (this data needs to be obtained from the feed weighing or batch records) and determining the average proportion of polished rice (represented by weight) in the total paddy rice. (For example This indicator reflects the total yield from raw paddy rice to the final obtained whole, qualified polished rice. The calculated head rice yield is... (For example, 65.38%) is used as the final output to generate the head rice yield index.

[0108] Please see Figure 2 This invention provides a device for the whole-process detection of rice quality indicators. This device is used to execute the above-mentioned whole-process detection method for rice quality indicators. The device includes:

[0109] The infrared image acquisition module acquires images of the feed particles through an infrared camera, filters rice particles based on the feed particle images, extracts rice texture and contour coordinates based on the surface images of the rice particles, identifies the position and shape of the rice particles, compares the texture of sample rice particles, and obtains the effective feed particle index.

[0110] The transmission intensity detection module detects the infrared transmission intensity of the brown rice based on the effective feeding particle index, calculates the average transmission intensity, compares the transmission intensity of the sample brown rice with the average value, analyzes the structural characteristics of the brown rice, generates a structural integrity rating, calculates the perfect grain ratio index and the imperfect grain ratio index according to the proportion of the transmission intensity qualified grains, and generates a brown rate index;

[0111] The rice grain quality classification module performs image recognition on the milled rice grains, recognizes the contour image of the single rice grain, calculates the contour defect proportion, and classifies the rice grains into broken rice and milled rice to obtain a rice grain quality classification result.

[0112] The weighing and distribution module guides the rice grains into corresponding weighing channels based on the rice grain quality classification result, records the cumulative weight of the channels, compares the total weight of the rice grains, calculates the weight proportion of the milled rice, and generates a whole milled rice rate index.

[0113] The qualified rate calculation module performs image recognition on the milled rice obtained by milling, identifies the proportion of yellow rice based on the image color, calculates the final qualified rate by combining the brown rate and the whole milled rice rate index, and outputs the proportion of yellow rice and a comprehensive quality rating.

[0114] It should be noted that in the above embodiments, the description of each embodiment has its own emphasis, and the parts not described in detail in a certain embodiment can be referred to the related description of other embodiments.

[0115] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, device, or computer program product. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer usable devices (including but not limited to disk memory, CD-ROM, optical memory, etc.) containing computer usable program code.

[0116] The present application is described with reference to flowcharts and / or block diagrams of methods, devices (apparatuses), and computer program products according to embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, as well as the combination of flows and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing devices to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices produce a machine that implements the functions specified in the flowcharts and / or block diagrams. Figure 1 The device that implements the functions specified in one flow or multiple flows and / or blocks. Figure 1 The device that implements the functions specified in one flow or multiple flows and / or blocks.

[0117] These computer program instructions can also be stored in a computer- readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instructions which implement the Figure 1 function specified in the flow or flows and / or blocks Figure 1 of the block or blocks.

[0118] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions that are executed on the computer or other programmable apparatus provide steps for implementing the Figure 1 function specified in the flow or flows and / or blocks Figure 1 of the block or blocks.

[0119] While the preferred embodiments of the application have been described, additional variations and modifications can be made to the embodiments by those skilled in the art once they learn of the basic inventive concepts. Therefore, the appended claims are intended to cover all such additional variations and modifications as fall within the scope of the application.

[0120] Obviously, numerous modifications and variations of the present application are possible in light of the above teachings. It is therefore to be understood that within the scope of the appended claims and their equivalents, the application can be practiced otherwise than as specifically described.

Claims

1. A method for detecting the quality index of paddy throughout the whole process, characterized in that, The method comprises: S1: collecting the images of the input particles by the infrared camera, screening the rice particles according to the images of the input particles, extracting the rice texture and contour coordinates according to the surface images of the rice particles, identifying the position and shape of the rice and comparing the sample rice texture to obtain the effective input particle index of the input hulling chamber; S2: performing image recognition on the brown rice particles output by the hulling chamber, detecting the infrared transmission intensity of the brown rice, calculating the average transmission intensity, comparing the transmission intensity of the sample brown rice with the average value, analyzing the structural characteristics of the brown rice, generating a structural integrity rating, calculating the perfect particle ratio index and the imperfect particle ratio index according to the proportion of the transmission intensity qualified particles, and generating the husked rate index; S3: grinding the brown rice and performing image recognition on the milled rice, identifying the contour image of the single rice, calculating the contour damage ratio, and classifying the milled rice into broken rice and polished rice to obtain the milled rice quality classification result; S4: based on the milled rice quality classification result, the milled rice is introduced into the corresponding weighing channel, the cumulative weight of the channel is recorded, the total weight of the milled rice is compared, the weight proportion of the polished rice is calculated, and the whole polished rice rate index is generated; S5: performing image recognition on the polished rice obtained by grinding, identifying the proportion of yellow rice according to the image color, calculating the proportion of yellow rice in the total polished rice, combining the husked rate and whole polished rice rate indexes to calculate the final qualified rate, and outputting the proportion of yellow rice and the comprehensive quality rating; The structural integrity rating step specifically comprises: S201: based on the effective input particle, using an infrared transmission device to perform transmission detection on the brown rice after hulling, collecting the transmission data of each sample brown rice, analyzing the energy change of the transmission light passing through the brown rice, calculating the transmission intensity value of the single brown rice, and generating the brown rice transmission intensity feature; S202: based on the brown rice transmission intensity feature, calculating the average value of the transmission intensity of the brown rice, comparing with the transmission data of the single brown rice, analyzing the deviation degree of the transmission value, screening the brown rice data with abnormal transmission value, judging the value range of the abnormal transmission intensity, and generating the transmission intensity deviation value; S203: based on the brown rice transmission deviation value, analyzing the transmission uniformity of the brown rice, calculating the distribution of the transmission deviation, judging the structural characteristics of the light intensity abnormal brown rice, screening the brown rice with standard structure, dividing the structure integrity level, and generating the structure integrity rating.

2. The method according to claim 1, wherein, The effective input particle index includes particle size distribution, arrangement density and input uniformity, the perfect particle ratio index specifically includes structure continuity ratio, low transmission defect rate and perfect particle proportion, the milled rice quality classification result includes contour integrity level, surface damage degree and particle type consistency index, the whole polished rice rate index specifically includes weight proportion, polished rice stable output ratio and unit raw material polished rice output ratio, and the comprehensive quality rating includes color uniformity, polished rice purity and yellow rice proportion level.

3. The method according to claim 1, wherein, The step of the effective input particle index of the input hulling chamber specifically comprises: S101: Collecting an image of the input grain by an infrared camera, screening the rice grain according to the image of the input grain, detecting the edge profile of the rice image according to the surface image of the rice grain, calculating the pixel gradient change of the continuous area of the image, locating the edge profile coordinates of the rice, analyzing the trend of the interval change of the texture, and obtaining the edge parameters of the rice; S102: Identifying the morphological features of the single rice grain based on the set of edge parameters of the rice, comparing the length-width ratio and surface area of the standard rice, judging the contour closure degree of each rice grain, identifying the rice with abnormal morphology, analyzing the distribution interval of the rice morphology, and generating the morphological parameters of the rice; S103: Screening the number of rice grains meeting the standard based on the morphological parameters of the rice, and obtaining the effective input grain index of the husking chamber.

4. The method according to claim 1, wherein, The transmission intensity value of the single brown rice is calculated by using the formula: ； The transmission intensity characteristics of the brown rice are generated. wherein, represents a transmission intensity value of a single grain of brown rice, represents an incident light intensity, represents an optical absorption coefficient of brown rice, represents a thickness of a single grain of brown rice, represents a transmission scattered light intensity, represents a light scattering correction parameter, represents a reference thickness.

5. The method for detecting the quality index of paddy according to claim 1, wherein, The S3 further comprises: Based on the structural integrity rating, the polished rice image is denoised, the surface features of the brown rice are analyzed, the brightness fluctuation and the texture interval are calculated, the milling sample is called, the data deviating from the milling range is identified, the uniformity of the milling is judged, the milling quality analysis result is obtained, and based on the milling quality analysis result, the sample is transmitted to the image acquisition area.

6. The method for detecting the quality index of paddy according to claim 1, wherein, The steps of the rice grain quality classification result are specifically: S301: Based on the structural integrity rating, the image data of the brown rice after milling is called, and the image is denoised, the pixel information of the surface area of the rice grain is extracted, the spatial distribution of the pixel gray value is analyzed, the area with high surface pixel brightness is screened, the brightness change value of the area is calculated, and the brightness fluctuation data of the surface of the rice grain is generated; S302: Based on the brightness fluctuation data of the surface of the rice grain, the distribution of the pixel area with high brightness change is analyzed, the interval between adjacent brightness extreme points is measured, the offset amplitude of the brightness mutation interval is calculated, and the texture difference of the surface of the rice grain is determined, and the texture interval difference of the rice grain is generated; S303: Based on the brightness fluctuation data of the surface of the rice grain and the texture interval difference of the rice grain, the rice milling sample data is called, the data points exceeding the conventional milling interval are screened, the surface brightness of each rice is compared with the texture change range, the smoothness of the surface of the rice grain is analyzed, the milling uniformity offset is calculated, and the milling quality analysis result is obtained; S304: Based on the milling quality analysis result, the rice sample is transmitted to the image acquisition area by using the conveying device, the contour boundary of the single rice grain is detected, the contour edge pixel points are screened and connected to form a closed contour, the overlapping area of the contour is calculated, the overlapping part is morphologically segmented, and the contour feature of the rice grain is obtained; S305: Based on the contour feature of the rice grain, the pixel distribution on the surface of the rice grain is analyzed, the surface spot area is screened according to the brightness change of the pixel, and the pixel breakage along the main axis direction is detected, the number and length of the cracks are calculated, the structural integrity of the surface of the rice grain is analyzed, and the surface damage data is obtained; S306: Based on the surface damage data, the surface area of the rice kernel is measured and compared with the spot and crack area, the total area ratio of the defect area is calculated, the rice kernel is classified according to the contour defect ratio, the surface is smooth and complete, and the rice kernel is classified as broken rice, and the rice kernel quality classification result is obtained.

7. The method according to claim 6, wherein, The offset amplitude of the brightness mutation interval is calculated by the formula: ； And determine the texture difference of the rice kernel, generate the texture interval difference of the rice kernel; wherein, an offset magnitude representing a luminance abruptness interval, a luminance representing a first luminance maximum point, a luminance representing a first luminance minimum point, a pixel interval representing a distance between adjacent luminance extreme points, a total number of detected luminance extreme point pairs.

8. The method for detecting the quality index of paddy according to claim 1, wherein, The steps of the whole milled rice rate index are specifically: S401: Based on the rice kernel quality classification result, the conveying path of the rice kernel is automatically distributed, the broken rice and the milled rice are respectively introduced into the corresponding weighing channel, the cumulative weight data of each channel is recorded in real time, and the total amount of each type of rice kernel is obtained; S402: Call the total amount of each type of rice kernel, call the total weight of the rice kernel, compare the cumulative weight of each type of rice kernel with the total weight, calculate the weight ratio of broken rice and milled rice respectively, and judge the average proportion of milled rice in the total amount of paddy according to the total amount of paddy, to generate the whole milled rice rate index.

9. A device for detecting the quality of rice throughout the whole process, characterized by, The device stores a computer software program, which is executed by the processor to realize the rice quality index full-process detection method according to any one of claims 1-8, and the device comprises: An infrared image acquisition module acquires the input particle image through an infrared camera, screens the paddy particles according to the input particle image, extracts the paddy texture and contour coordinates according to the surface image of the paddy particles, identifies the position and shape of the paddy and compares the sample paddy texture to obtain the effective input particle index; A transmission intensity detection module detects the intensity of the infrared transmission of brown rice based on the effective input particle index, calculates the average transmission intensity, compares the transmission intensity of the sample brown rice with the average value, analyzes the structural characteristics of the brown rice, generates a structural integrity rating, calculates the perfect particle ratio index and the imperfect particle ratio index according to the transmission intensity of the qualified particles, and generates the brown rate index; A rice kernel quality classification module identifies the contour image of a single rice kernel, calculates the contour defect ratio, and classifies the rice kernel into broken rice and milled rice to obtain the rice kernel quality classification result; A weighing distribution module introduces the rice kernel into the corresponding weighing channel based on the rice kernel quality classification result, records the cumulative weight of the channel, compares the total weight of the rice kernel, calculates the weight ratio of the milled rice, and generates the whole milled rice rate index; A qualified rate calculation module identifies the milled rice obtained by grinding according to the image color, calculates the proportion of the yellow rice kernel in the total amount of the milled rice, calculates the final qualified rate according to the brown rate and the whole milled rice rate index, and outputs the yellow rice kernel ratio and the comprehensive quality rating.

Images