Grain impurity measurement device and method

Abstract

Provided are a grain impurity measurement device and method. The measurement method comprises: a filtering component relies on the gravity of grain to filter out primary impurities from flowing grains thereon, and a first weight sensor (2) measures a first weight of the grains having the primary impurities filtered out on the filtering component in real time; an impurity collecting component (3) collects the primary impurities filtered out below the filtering component, and a second weight sensor (4) measures a second weight of the primary impurities in real time; and an image acquisition unit (5) acquires sample images of the grains having the primary impurities filtered out on the filtering component in real time, and a controller determines a third weight of secondary impurities in the grains on the basis of the sample images, and then determines the impurity ratio of the grains on the basis of the first weight, the second weight and the third weight. The measurement device and the measurement method aim to solve the problem of poor accuracy of existing grain impurity measurement methods, reduce labor intensity of field testing personnel and improve the accuracy of grain impurity content measurement, and are suitable for grain quality testing.

Description

Grain impurity detection device and method Technical Field

[0001] The present application relates to the technical field of grain detection, and in particular to a device and method for detecting grain impurities. Background Art

[0002] Grain quality inspection is the first hurdle before grain enters storage. Accurately assessing grain quality is a prerequisite for ensuring safe and scientific grain storage management. Impurity content is a key indicator of grain quality, directly affecting the purchase price and whether the grain meets purchase requirements.

[0003] Traditional impurity detection methods typically rely on fixed-point sampling. This involves sampling a portion of grain from a batch of grain. Each grain in the sample is then individually tested, and the impurity content of the entire batch is determined based on the test results. This method is complex, time-consuming, and labor-intensive. Furthermore, due to factors such as uneven impurity distribution and the influence of sampling location, the measurement results may not accurately assess the impurity level of the entire grain.

[0004] Summary of the Invention

[0005] This application aims to solve the problems of existing grain impurity detection methods being time-consuming, labor-intensive and poorly accurate, and to propose another grain impurity detection device and method.

[0006] The technical solution adopted by this application to solve the above technical problems is:

[0007] In a first aspect, the present application provides a device for detecting impurities in grain, the device comprising:

[0008] The filtering component is used to filter the first impurities in the grain flowing above it by relying on the gravity of the grain;

[0009] a first weight sensor for detecting in real time a first weight of the grain above the filter component after the first impurities have been filtered out;

[0010] An impurity collecting component, used for collecting first impurities filtered out from below the filter component;

[0011] a second weight sensor, configured to detect in real time a second weight of the first impurities filtered out by the filter component;

[0012] An image collector, used for collecting in real time a sample image of the grain above the filter component after being filtered by the first impurities;

[0013] The controller is configured to determine a third weight of a second impurity in the grain according to the sample image, and determine an impurity rate of the grain according to the first weight, the second weight, and the third weight.

[0014] Furthermore, the filtering component includes a support frame and a screen obliquely arranged on the support frame, and the first weight sensor is arranged below the support frame.

[0015] Furthermore, the impurity collecting component is arranged below the screen and is suspended and fixed in the bracket, and the second weight sensor is arranged below the impurity collecting component.

[0016] Furthermore, determining a third weight of the second impurity in the grain according to the sample image specifically includes:

[0017] Identify grain images and impurity images in sample images based on a pre-trained impurity recognition model;

[0018] Determining a first pixel area of ​​a region formed by a grain outline in the sample image, and determining a second pixel area of ​​a region formed by an impurity outline in each impurity image;

[0019] A third weight of a second impurity in the grain is determined based on the first weight, the first pixel area, and the second pixel area.

[0020] Further, determining a third weight of a second impurity in the grain according to the first weight, the first pixel area, and the second pixel area specifically includes:

[0021] Determining a first pixel area sum of all grain images and a second pixel area sum of all impurity images in the sample image, and determining a pixel area sum of the first pixel area sum and the second pixel area sum;

[0022] determining a mass proportion of a second impurity in the grain based on a ratio of the first pixel area sum to the total pixel area;

[0023] A third weight of the second impurity in the grain is determined based on the first weight and the mass ratio.

[0024] Furthermore, the calculation formula of the impurity rate is as follows:

[0025] Here, M represents the impurity rate, M1 represents the first weight, M2 represents the second weight, and M3 represents the third weight.

[0026] In a second aspect, the present application provides a method for detecting food impurities, the method comprising:

[0027] filtering first impurities in the grain flowing above the filtering component by relying on the grain gravity, and detecting in real time a first weight of the grain above the filtering component after the first impurities have been filtered;

[0028] collecting the first impurities filtered out from below the filter component by the impurity collecting component, and detecting in real time the second weight of the first impurities filtered out by the filter component;

[0029] A sample image of the grain after the first impurity is filtered out above the filter component is collected in real time, a third weight of the second impurity in the grain is determined based on the sample image, and an impurity rate of the grain is determined based on the first weight, the second weight and the third weight.

[0030] Furthermore, determining a third weight of the second impurity in the grain according to the sample image specifically includes:

[0031] Identify grain images and impurity images in sample images based on a pre-trained impurity recognition model;

[0032] Determining a first pixel area of ​​a region formed by a grain outline in the sample image, and determining a second pixel area of ​​a region formed by an impurity outline in each impurity image;

[0033] A third weight of a second impurity in the grain is determined based on the first weight, the first pixel area, and the second pixel area.

[0034] Further, determining a third weight of a second impurity in the grain according to the first weight, the first pixel area, and the second pixel area specifically includes:

[0035] Determining a first pixel area sum of all grain images and a second pixel area sum of all impurity images in the sample image, and determining a pixel area sum of the first pixel area sum and the second pixel area sum;

[0036] determining a mass proportion of a second impurity in the grain based on a ratio of the first pixel area sum to the total pixel area;

[0037] A third weight of the second impurity in the grain is determined based on the first weight and the mass ratio.

[0038] Furthermore, the calculation formula of the impurity rate is as follows:

[0039] Here, M represents the impurity rate, M1 represents the first weight, M2 represents the second weight, and M3 represents the third weight.

[0040] The beneficial effects of the present application are: the grain impurity detection device and method provided by the present application can automatically detect the weight of the first impurity and the second impurity in the grain when the grain flows above the filter component, and the impurity content of the grain can be determined by the sum of the weight of the first impurity and the second impurity and the total weight of the grain containing impurities. The present application can detect the impurity content in the grain in real time online, which can greatly reduce the labor intensity of on-site inspection personnel and can make a more accurate assessment of the impurity content of the entire grain. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] FIG1 is a schematic structural diagram of a grain impurity detection device provided in an embodiment of the present application;

[0042] FIG2 is a schematic diagram of a process for detecting food impurities according to an embodiment of the present application;

[0043] Explanation of reference numerals: 11 - support frame, 12 - screen; 2 - first weight sensor; 3 - impurity collecting component; 4 - second weight sensor; 5 - image collector. DETAILED DESCRIPTION

[0044] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.

[0045] In some of the processes described in the specification of this application and the above-mentioned figures, multiple operations that appear in a specific order are included, but it should be clearly understood that these operations may not be executed in the order in which they appear in this article or may be executed in parallel. The sequence numbers of the operations are only used to distinguish between different operations, and the sequence numbers themselves do not represent any order of execution. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions of "first", "second", etc. in this article are used to distinguish different messages, devices, modules, etc., and do not represent the order of precedence, nor do they limit "first" and "second" to different types.

[0046] The technical solution of the embodiment of the present application is applicable to application scenarios where impurity content of grains needs to be detected, such as rice grains, wheat grains, legume grains or coarse grains.

[0047] At present, the impurity content of grain is basically detected through sampling. This method is not only complicated in operation process, time-consuming and labor-intensive, but also has poor accuracy due to the influence of factors such as uneven distribution of impurities and sampling position.

[0048] Based on this, the technical solution of the present application is proposed. In an embodiment of the present application, a filtering component is used to filter the first impurities in the flowing grain above the filtering component by relying on the gravity of the grain, and a first weight sensor is used to detect in real time the first weight of the grain above the filtering component after the first impurities are filtered out; an impurity collecting component is used to collect the first impurities filtered out below the filtering component, and a second weight sensor is used to detect in real time the second weight of the first impurities filtered out by the filtering component; an image collector is used to collect in real time a sample image of the grain above the filtering component after the first impurities are filtered out, and a controller is used to determine the third weight of the second impurities in the grain based on the sample image, and the impurity rate of the grain is determined based on the first weight, the second weight and the third weight.

[0049] Specifically, under the action of the gravity of the grain, the grain flows above the filter component, and the first impurities in the grain are filtered to the bottom of the filter component. The first impurities are collected by the impurity collecting component and weighed by the second weight sensor to obtain the second weight of the first impurities, wherein the first impurities are impurities that can be filtered by the screen. The grain and the second impurities are retained above the filter component, and the second impurities are impurities that cannot be filtered by the screen. The first weight of the grain and the second impurities are weighed by the first weight sensor, and a sample image is collected by the image collector. The controller performs image recognition on the collected sample image and determines the third weight of the second impurity therein. Finally, the impurity rate of the grain can be determined based on the first weight, the second weight, and the third weight. In actual application, the inspection personnel only need to place the grain above the filter component at a certain flow rate to realize automatic impurity content detection of the flowing grain, realize real-time online detection of the impurity content of the grain, reduce the labor intensity of on-site inspection personnel, and improve the accuracy of grain impurity content detection.

[0050] The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments.

[0051] Referring to Figure 1 , the grain impurity detection device provided in an embodiment of the present application includes a filter component, an impurity collection component 3, a first weight sensor 2, a second weight sensor 4, an image collector 5, and a controller. The first weight sensor 2, the second weight sensor 4, and the image collector 5 are electrically connected to the controller.

[0052] In the embodiment of the present application, the filter component is used to rely on the gravity of the grain to filter out first impurities from the grain flowing above it. The filter component may include a support frame 11 and a screen 12 tilted on the support frame. The screen 12 is provided with a plurality of sieve holes with a pore size smaller than the grain particles. The impurity collection component 3 is used to collect first impurities filtered out from below the filter component. First impurities are impurities that can be filtered by the screen, such as undersize and organic or inorganic impurities that can be filtered by the screen. The impurity collection component 3 is disposed below the screen 12 and is suspended and fixed within the support frame 11.

[0053] In the embodiment of the present application, the first weight sensor 2 is disposed below the support frame 11 and is used to detect in real time the first weight of the grain above the filter component after the first impurities have been filtered out, i.e., the total weight of the grain and the second impurities, and transmit the weight to the controller. The second weight sensor 4 is disposed below the impurity collecting component 3 and is used to detect in real time the second weight of the first impurities filtered out by the filter component and transmit the weight to the controller. The image collector 5 is used to capture in real time a sample image of the grain above the filter component after the first impurities have been filtered out and transmit the image to the controller. The image collector 5 may be a camera disposed above the screen 12.

[0054] Referring to FIG. 2 , based on the above-mentioned detection device, the grain impurity detection method provided in the embodiment of the present application includes the following steps:

[0055] Step 201: filtering first impurities in the grain flowing above the filter component by relying on the grain's gravity, and detecting in real time a first weight of the grain above the filter component after the first impurities have been filtered out;

[0056] In actual use, testing personnel place grain above the filter element at a certain flow rate. As the grain flows through the inclined screen under the action of gravity, the screen filters out the first impurities in the grain. After the filter, the grain is still above the screen. At this time, the first weight sensor below the support frame detects the first weight and sends it to the controller. The first weight is the weight of the filtered grain. The filtered grain may contain second impurities, which are impurities that cannot be filtered by the screen.

[0057] Step 202: collecting first impurities filtered out from below the filter component by the impurity collecting component, and detecting a second weight of the first impurities filtered out by the filter component in real time;

[0058] During the process of filtering the grain through the screen, the first impurity falls into the impurity collecting component below the screen and is collected. At this time, the second weight is detected by the second weight sensor below the impurity collecting component and sent to the controller. The second weight is the weight of the first impurity filtered out by the screen.

[0059] Step 203: collect a sample image of the grain after impurities are filtered above the filter component in real time, determine the third weight of the second impurity in the grain based on the sample image, and determine the impurity rate of the grain based on the first weight, second weight and third weight.

[0060] The image collector arranged above the screen collects sample images of the filtered grain in real time and sends them to the controller. The sample image contains grain and may also contain second impurities, that is, impurities that cannot be filtered by the screen, such as organic impurities or inorganic impurities that cannot be filtered by the screen.

[0061] The controller determines the third weight of the second impurity according to the received sample image, specifically including the following steps:

[0062] Step 2031: Identify grain images and impurity images in the sample image according to the pre-trained impurity identification model;

[0063] In practical applications, multiple secondary impurities in grain can be collected and imaged grain by grain. The obtained impurity images are preprocessed and used as negative sample images. Multiple grain images are then acquired and preprocessed as positive sample images. Finally, a deep learning model is trained based on the positive and negative sample images to obtain an impurity recognition model. By inputting the sample images into the impurity recognition model, the grain and secondary impurities in the sample images can be determined, enabling classification of the secondary impurities in the grain.

[0064] Step 2032: determining a first pixel area of ​​a region formed by a grain outline in the sample image, and determining a second pixel area of ​​a region formed by an impurity outline in each impurity image;

[0065] In actual application, after determining the grain image and impurity image, the corresponding pixel area can be determined according to the number of pixels in the area formed by the outline of the target object in the image. The corresponding first pixel area is determined for the grain image, and the corresponding second pixel area is determined for the impurity image.

[0066] Step 2033: Determine a third weight of a second impurity in the grain based on the first weight, the first pixel area, and the second pixel area.

[0067] Generally, the larger the pixel area of ​​the area formed by the outline of the target object in a single image, the heavier the corresponding target object. Based on this, the embodiment of the present application first determines the second pixel area sum of the area formed by the outlines of all second impurities, then determines the total pixel area of ​​the area formed by the impurity outlines of all grains and second impurities. Finally, based on the ratio of the second pixel area sum to the total pixel area and combined with the correlation coefficient, the mass percentage of the second impurity in the grain can be estimated. The product of the first mass and the mass percentage is the third weight of the second impurity in the grain.

[0068] After the controller obtains the first weight, the second weight, and the third weight, it can calculate the impurity rate of the grain. The calculation formula is as follows:

[0069] Here, M represents the impurity rate, M1 represents the first weight, M2 represents the second weight, and M3 represents the third weight.

[0070] In summary, the grain impurity detection device and method provided in the embodiments of the present application can automatically detect the weight of the first impurity and the second impurity in the grain when the grain flows over the filter component, and the impurity content of the grain can be determined by the sum of the weights of the first impurity and the second impurity and the total weight of the grain containing the impurities. In actual application, the inspection personnel only need to place the grain over the filter component at a certain flow rate to automatically detect the impurity content of the flowing grain, realize real-time online detection of the impurity content of the grain, reduce the labor intensity of on-site inspection personnel, and improve the accuracy of grain impurity content detection.

Claims

1. A grain impurity detection device, characterized in that, The device includes: A filtering component for filtering the first impurities in the flowing grain above it by relying on the gravity of the grain; A first weight sensor for real-time detecting the first weight of the grain after the first impurities are filtered above the filtering component; The filtering component includes a support frame and a sieve inclinedly arranged on the support frame, and the first weight sensor is arranged below the support frame; An impurity collection component for collecting the first impurities filtered out below the filtering component; A second weight sensor for real-time detecting the second weight of the first impurities filtered out by the filtering component; The impurity collection component is arranged below the sieve and is suspended and fixed inside the support frame, and the second weight sensor is arranged below the impurity collection component; An image collector for real-time collecting a sample image of the grain after the first impurities are filtered above the filtering component; A controller for determining the third weight of the second impurities in the grain according to the sample image, and determining the impurity rate of the grain according to the first weight, the second weight and the third weight.

2. The grain impurity detection device according to claim 1, characterized in that Determining the third weight of the second impurities in the grain according to the sample image specifically includes: Identifying the grain image and the impurity image in the sample image according to a pre-trained impurity recognition model; Determining the first pixel area of the area formed by the grain contour in the sample image, and respectively determining the second pixel area of the area formed by the impurity contour in each impurity image; Determining the third weight of the second impurities in the grain according to the first weight, the first pixel area and the second pixel area.

3. The grain impurity detection device according to claim 2, characterized in that, Determining the third weight of the second impurities in the grain according to the first weight, the first pixel area and the second pixel area specifically includes: Determining the sum of the first pixel areas of all the grain images and the sum of the second pixel areas of all the impurity images in the sample image, and determining the total pixel area of the sum of the first pixel area and the sum of the second pixel areas; Determining the mass proportion of the second impurities in the grain according to the ratio of the sum of the first pixel areas to the total pixel area; Determining the third weight of the second impurities in the grain according to the first weight and the mass proportion.

4. The grain impurity detection device according to claim 1, wherein, The calculation formula for the impurity rate is as follows: Wherein, M represents the impurity rate, M1 represents the first weight, M2 represents the second weight, and M3 represents the third weight.

5. A method for detecting grain impurities, characterized in that, Applied to the grain impurity detection device according to any one of claims 1 to 4, the method includes: Filtering the first impurities in the flowing grain above it by relying on the gravity of the grain through the filtering component, and real-time detecting the first weight of the grain after the first impurities are filtered above the filtering component; Collecting the first impurities filtered out below the filtering component through the impurity collection component, and real-time detecting the second weight of the first impurities filtered out by the filtering component; Real-time collecting a sample image of the grain after the first impurities are filtered above the filtering component, determining the third weight of the second impurities in the grain according to the sample image, and determining the impurity rate of the grain according to the first weight, the second weight and the third weight.

6. The method for detecting grain impurities according to claim 5, characterized in that, Determining the third weight of the second impurities in the grain according to the sample image specifically includes: Identifying the grain image and the impurity image in the sample image according to a pre-trained impurity recognition model; Determine the first pixel area of the region formed by the grain contour in the sample image, and determine the second pixel area of the region formed by the impurity contour in each impurity image respectively; Determine the third weight of the second impurity in the grain according to the first weight, the first pixel area and the second pixel area.

7. The method for detecting grain impurities according to claim 6, wherein, Determine the third weight of the second impurity in the grain according to the first weight, the first pixel area and the second pixel area, specifically including: Determine the sum of the first pixel areas of all grain images and the sum of the second pixel areas of all impurity images in the sample image, and determine the total pixel area of the sum of the first pixel area and the sum of the second pixel area; Determine the mass proportion of the second impurity in the grain according to the ratio of the sum of the first pixel areas to the total pixel area; Determine the third weight of the second impurity in the grain according to the first weight and the mass proportion.

8. The method for detecting grain impurities according to claim 5, characterized in that, The calculation formula for the impurity rate is as follows: Wherein, M represents the impurity rate, M1 represents the first weight, M2 represents the second weight, and M3 represents the third weight.

Images