Movable river crab rapid classification and detection device and method thereof

Abstract

The invention relates to the technical field of aquatic product detection, in particular to a movable river crab rapid classification and detection device and a method thereof.The movable river crab rapid classification and detection device comprises a movable box body, a sliding table motor and a crab outlet are installed on the two sides of the movable box body respectively, and a movable sliding table is installed in the movable box body; a sliding block capable of moving in the direction of the movable sliding table is mounted on the movable sliding table, and a crab carrying box capable of moving in the direction of the sliding block is arranged on the sliding block; a spectrum collecting device is arranged in the movable box body and located above the movable sliding table, and the spectrum collecting device is used for collecting spectrum information of the river crabs in the crab carrying box. The spectrum collecting device is used for collecting spectrums of the river crabs, and the near-infrared spectrometer is sensitive to distinguishing different varieties of river crabs; spectral information of the river crabs can be effectively obtained, and the river crabs are conveyed to different crab outlets to be discharged through cooperation of the crab carrying box and the movable sliding table according to the detection result.

Description

Technical Field

[0001] The invention relates to the technical field of aquatic product detection, and in particular to a movable river crab rapid classification detection device and a method thereof. Background Art

[0002] The scientific name of river crab is Chinese mitten crab, which is a precious freshwater product. It tastes delicious, is rich in nutrients and has high economic value. The muscle tissue of the crab is rich in protein, and the crab meat contains amino acids such as alanine, glycine, serine, etc. It is a food that is deeply loved by consumers.

[0003] There are significant differences in the components of fatty acids, flavor substances and mineral elements in river crabs from the same origin. The ω-3 and ω-6 fatty acids in fatty acids can reduce the risk of cardiovascular and cerebrovascular diseases. Flavor substances such as amino acids, nucleotides, and organic acids directly affect the taste of river crabs. Mineral elements such as calcium, phosphorus, potassium, and magnesium are essential to human health. Therefore, the taste and nutritional content of river crabs from different origins are obviously different, which leads to huge differences in market prices. However, river crabs on the market have been exploited by unscrupulous merchants, resulting in problems such as inferior products being sold as good ones and low-priced crabs being passed off as crabs from famous origins, which has seriously damaged the rights and interests of consumers.

[0004] In the existing technology, the origin of river crabs is mainly identified by physical and chemical destructive testing and stable isotope method. However, when using existing technology for testing, there are problems such as cumbersome operation, time-consuming and inefficient, and it is difficult to identify the origin of river crabs in the circulation link. Summary of the invention

[0005] The purpose of the present invention is to provide a movable river crab rapid classification and detection device and method thereof to solve the problems raised in the above-mentioned background technology.

[0006] The purpose of the present invention can be achieved through the following technical solutions:

[0007] A movable river crab rapid classification and detection device comprises a movable box, a slide motor and a crab outlet are respectively installed on both sides of the movable box, a movable slide is installed inside the movable box, a slider that can move along the direction of the movable slide is installed on the movable slide, and a crab carrying box that can move along the direction of the slider is arranged on the slider;

[0008] A spectral acquisition device is arranged inside the movable box and above the movable slide. The spectral acquisition device is used to collect spectral information of river crabs in the crab box. A Raspberry Pi is installed in the movable box. The spectral acquisition device is electrically connected to the Raspberry Pi and is used to transmit the collected spectral information to the Raspberry Pi through optical fiber. A position sensor is installed on the movable slide. The position sensor is electrically connected to the Raspberry Pi and is used to feed back the position information of the crab box to the Raspberry Pi so that the Raspberry Pi can adjust and control the position of the crab box.

[0009] Preferably, the movable box is rectangular and made of stainless steel. The movable box is sprayed with black paint, and a universal pulley is installed at the bottom of the movable box.

[0010] Preferably, a slider motor is fixedly mounted on the slider, and an output end of the slider motor and the crab carrying box are fixedly mounted to drive the crab carrying box to slide horizontally on the slider.

[0011] Preferably, a slide motor and a crab outlet are respectively provided on both sides of the movable box body, and there are multiple crab outlets for classifying and discharging river crabs from different origins. A photoelectric sensor is installed on the inner wall of the crab outlet, and the photoelectric sensor is a slot-type photoelectric sensor.

[0012] Preferably, a spectral device bracket is installed inside the movable box, the spectral acquisition device is fixedly installed on the spectral device bracket, a Raspberry Pi display screen is installed on the door panel of the movable box, and the Raspberry Pi display screen is electrically connected to the Raspberry Pi for controlling and displaying the set parameters of the Raspberry Pi.

[0013] Preferably, a crab box motor is installed on the outer surface of the crab box, and a push plate is fixedly connected to the output end of the crab box motor and located on the crab box, and the push plate is used to control the entry and exit of river crabs. A pneumatic plate is provided on the crab box, and the pneumatic plate of the air pump is connected to control the opening and closing of the pneumatic plate. A pneumatic module is installed on one side of the crab box, and the pneumatic module and the air pump are electrically connected.

[0014] The rapid classification and detection method for river crabs is used in the movable rapid classification and detection device for river crabs to classify and detect river crabs. The rapid classification and detection method for river crabs comprises the following steps:

[0015] S1. Build a collection platform for river crab spectra, and collect river crab spectrum information through the collection platform;

[0016] S2, processing the collected spectral information of river crabs to obtain a two-dimensional spectral feature matrix;

[0017] S3. Building a river crab classification model based on a convolutional neural network and a two-dimensional spectral feature matrix, and training the river crab classification model;

[0018] S4. Process the spectral information of the crabs to be classified and convert it into a two-dimensional spectral feature matrix. Input the two-dimensional spectral feature matrix into the trained crab classification model to complete the detection of different species of crabs and classify them.

[0019] Preferably, the rapid classification and detection method for river crabs is executed and processed by a Raspberry Pi, and instructions are issued to each execution device to classify the river crabs being tested.

[0020] Preferably, in step S2, the collected crab spectral information is processed, including the following steps:

[0021] S21, intercepting effective spectral information, and performing SG smoothing and standard normal variable transformation for preprocessing to obtain a sample set;

[0022] S22, labeling the sample set, and using the SPXY algorithm to divide the sample set into a training set and a test set;

[0023] S23, selecting characteristic wavelength points for samples in the training set using the CARS algorithm;

[0024] S24. Based on the characteristic wavelength points selected from the training set samples, a specific method for constructing a two-dimensional spectral feature matrix based on the characteristic wavelength points includes: obtaining a one-dimensional spectral feature matrix based on the characteristic wavelength points, and multiplying the one-dimensional spectral feature matrix with the transposed matrix of the one-dimensional spectral feature matrix to obtain a two-dimensional spectral feature matrix.

[0025] Beneficial effects of the present invention:

[0026] 1. The present invention utilizes a spectral acquisition device to collect the spectrum of river crabs. The near-infrared spectrometer is more sensitive to the differences between different species of river crabs and can effectively obtain the spectral information of river crabs. According to the detection results, the crabs are transported to different crab outlets for crab discharge by using the cooperation of a crab carrying box and a mobile slide. At the same time, the present invention constructs a river crab recognition model based on a convolutional neural network, which can perform fast and accurate non-destructive classification detection of different species of river crabs.

[0027] 2. The present invention performs effective spectral information interception and noise data removal on the collected spectral information of different species of river crabs, and obtains wavelength points that can reflect the differences in different species of river crabs through the selection of characteristic wavelength points, which effectively reduces the amount of calculation and improves the accuracy of river crab classification and identification; by converting the one-dimensional spectral feature matrix of river crabs into a two-dimensional spectral feature matrix, the spectral characteristics of river crabs can be more accurately represented, thereby improving the accuracy of river crab classification and identification. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0029] Figure 1 It is a structural schematic diagram of the device for rapid classification and detection of river crabs proposed by the present invention;

[0030] Figure 2 The present invention Figure 1 Schematic diagram of the structure of the mid-load crab box section;

[0031] Figure 3 It is a schematic diagram of the process of the rapid classification and detection method of river crabs proposed by the present invention.

[0032] The reference numerals in the figures are as follows:

[0033] 1. Movable box, 2. Movable slide, 3. Slide motor, 4. Air pump, 5. Spectrum acquisition device, 6. Spectrum device bracket, 7. Slider, 8. Slider motor, 9. Crab box, 10. Raspberry Pi, 11. Position sensor, 12. Crab outlet, 13. Photoelectric sensor, 14. Raspberry Pi display, 15. Power socket, 16. Pneumatic plate, 17. Push plate, 18. Crab box motor, 19. Pneumatic module. DETAILED DESCRIPTION

[0034] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0035] A movable river crab rapid classification and detection device comprises a movable box 1, a slide motor 3 and a crab outlet 12 are respectively installed on both sides of the movable box 1, a movable slide 2 is installed inside the movable box 1, a slide block 7 that can move along the direction of the movable slide 2 is installed on the movable slide 2, and a crab carrying box 9 that can move along the direction of the slide block 7 is arranged on the slide block 7;

[0036] A spectrum collection device 5 is arranged inside the movable box 1 and above the movable slide 2. The spectrum collection device 5 is used to collect spectrum information of the crabs in the crab box 9. A Raspberry Pi 10 is installed in the movable box 1. The spectrum collection device 5 and the Raspberry Pi 10 are electrically connected to transmit the collected spectrum information to the Raspberry Pi 10 through optical fiber. A position sensor 11 is installed on the movable slide 2. The position sensor 11 and the Raspberry Pi 10 are electrically connected to feed back the position information of the crab box 9 to the Raspberry Pi 10 so that the Raspberry Pi 10 can adjust and control the position of the crab box 9.

[0037] like Figure 1 and Figure 2, the Raspberry Pi 10 is used to identify the crab species according to the spectral information, and send instructions to the various execution parts by analyzing the identification results; in this embodiment, the hardware configuration of the Raspberry Pi 10 is CPU Intel Core i7-1210M 2.50GHz processor, RAM (8G), windows64-bit operating system; the Raspberry Pi 10 is embedded with spectral data processing software, and the spectral information processing process of the processing software includes:

[0038] A. Read spectral data;

[0039] B. Intercept effective spectral information and perform spectral preprocessing;

[0040] C. Select characteristic wavelength points;

[0041] D. Convert one-dimensional spectral data into a two-dimensional spectral matrix;

[0042] E. Import the spectral two-dimensional matrix into the trained discriminant model;

[0043] F. Get the classification results.

[0044] The spectrum acquisition device 5 adopts a near-infrared spectrum acquisition instrument from PerkinElmer Spectrum Two, with a collection range of 10,000-4000cm -1 (1000-2500nm), resolution 4cm -1 The spectrum information collected by the spectrum collection device 5 is transmitted to the Raspberry Pi 10 via optical fiber.

[0045] The movable box 1 is rectangular and made of stainless steel. The inside of the movable box 1 is sprayed with black paint, and a universal pulley is installed at the bottom of the movable box 1.

[0046] The slider 7 is fixedly mounted with a slider motor 8 , and an output end of the slider motor 8 and a crab carrying box 9 are fixedly mounted to drive the crab carrying box 9 to slide horizontally on the slider 7 .

[0047] The slider motor 8 is the power output of the movable slide 2, and the movable slide 2 is used to carry the slider 7 and move it back and forth.

[0048] The two sides of the movable box body 1 are respectively provided with a slide motor 3 and a crab outlet 12. There are multiple crab outlets 12 for classifying and discharging river crabs from different origins. A photoelectric sensor 13 is installed on the inner wall of the crab outlet 12. The photoelectric sensor 13 is a slot-type photoelectric sensor.

[0049] The photoelectric sensor 13 is a prior art and is electrically connected to the Raspberry Pi 10 . When the photoelectric sensor 13 detects that the crab box 9 has arrived at the designated crab outlet 12 , the photoelectric sensor 13 will feed back a signal to the Raspberry Pi 10 .

[0050] A spectral device bracket 6 is installed inside the movable box 1, and the spectral acquisition device 5 is fixedly installed on the spectral device bracket 6. A Raspberry Pi display screen 14 is installed on the door panel of the movable box 1. The Raspberry Pi display screen 14 is electrically connected to the Raspberry Pi 10 for controlling and displaying the setting parameters of the Raspberry Pi 10.

[0051] like Figure 1 , the Raspberry Pi display screen 14 is embedded in the door panel; the power supply of the spectrum acquisition device 5, the power supply of the movable slide motor 3 and its sensor, the power supply of the slider motor 8, the power supply of the crab box pneumatic module 19 and the crab box motor 18, and the power supply of the Raspberry Pi 10 are all connected in series to the power socket 15 embedded in the side panel of the movable box 1, and then the device is powered by connecting an external power supply.

[0052] The Raspberry Pi display 14 is an external 7-inch LCD display for the Raspberry Pi 4B.

[0053] A crab box motor 18 is installed on the outer surface of the crab box 9. A push plate 17 is fixedly connected to the output end of the crab box motor 18 and located on the crab box 9. The push plate 17 is used to control the entry and exit of river crabs. A pneumatic plate 16 is provided on the crab box 9. The pneumatic plate 16 of the air pump 4 is connected to control the opening and closing of the pneumatic plate 16. A pneumatic module 19 is installed on one side of the crab box 9. The pneumatic module 19 and the air pump 4 are electrically connected.

[0054] The pneumatic plate 16 is a prior art, which is electrically connected to the air pump 4 through an excess hose. The opening and closing of the pneumatic plate 16 can be controlled by the air pump 4 without affecting the activity of the crab carrying box 9.

[0055] The method for rapid classification and detection of river crabs is used in the movable rapid classification and detection device for river crabs to classify and detect river crabs. The method for rapid classification and detection of river crabs comprises the following steps:

[0056] S1. Build a collection platform for river crab spectra, and collect river crab spectrum information through the collection platform;

[0057] S2, processing the collected spectral information of river crabs to obtain a two-dimensional spectral feature matrix;

[0058] S3. Building a river crab classification model based on a convolutional neural network and a two-dimensional spectral feature matrix, and training the river crab classification model;

[0059] S4. Process the spectral information of the crabs to be classified and convert it into a two-dimensional spectral feature matrix. Input the two-dimensional spectral feature matrix into the trained crab classification model to complete the detection of different species of crabs and classify them.

[0060] The method for rapid classification and detection of river crabs is executed by the Raspberry Pi 10, which simultaneously issues instructions to each execution component to classify the river crabs being tested.

[0061] In step S2, the collected crab spectral information is processed, including the following steps:

[0062] S21, intercepting effective spectral information, and performing SG smoothing and standard normal variable transformation for preprocessing to obtain a sample set;

[0063] S22, labeling the sample set, and using the SPXY algorithm to divide the sample set into a training set and a test set;

[0064] S23, selecting characteristic wavelength points for samples in the training set using the CARS algorithm;

[0065] S24. Based on the characteristic wavelength points selected from the training set samples, a specific method for constructing a two-dimensional spectral feature matrix based on the characteristic wavelength points includes: obtaining a one-dimensional spectral feature matrix based on the characteristic wavelength points, and multiplying the one-dimensional spectral feature matrix with the transposed matrix of the one-dimensional spectral feature matrix to obtain a two-dimensional spectral feature matrix.

[0066] The present invention develops spectrum acquisition software, uses a language secondary development API provided by the spectrum acquisition device, completes the calling of the spectrum acquisition device in the Qt development platform; uses the QLabel class of Qt5.12 to display the spectrum information of river crabs and the classification detection results of river crabs on a display screen provided by the Raspberry Pi; uses a software control button to control the output of a GPIO port of the Raspberry Pi, and when the classification detection of river crabs begins, the control button is pressed, and the Raspberry Pi drives a motor to operate, thereby transporting the river crabs to a designated location; implements the calling of a trained model in the Qt development environment; and builds the software system under Release, packages it, and publishes it, and deploys it to the Raspberry Pi.

[0067] The working principle of the movable crab rapid classification and detection device and method provided by the present invention is as follows:

[0068] Move the crab box 9 to the rightmost end of the movable slide 2, and manually put the crabs into the crab box 9; the crab box 9 moves back and forth with the movable slide 2. When the crab box 9 moves to directly below the spectrum acquisition device 5, the spectrum acquisition device 5 obtains the spectrum information of the crabs and transmits the spectrum information to the Raspberry Pi 10. The Raspberry Pi 10 processes the spectrum information, identifies the spectrum information, and displays it on the Raspberry Pi display 14. When the Raspberry Pi 10 detects and analyzes the species of the crabs, it will issue instructions to the movable slide 2 and the slider 7, and through the position sensor 11, the crab box 9 will reach the designated crab outlet 12.

[0069] When the crab box 9 reaches the designated crab outlet 12, the photoelectric sensor 13 will feedback a signal to the Raspberry Pi 10, which will then send an instruction to the crab box motor 18 to control the push plate 17 to move forward. At the same time, the instruction of the Raspberry Pi 10 will cause the pneumatic module 19 of the crab box 9 to open the pneumatic plate 16 by controlling the air pump 4, and send the river crabs out of the designated crab outlet 12, thereby completing the classification and detection of river crabs.

[0070] Compared with the related art, the movable crab rapid classification and detection device and method provided by the present invention have the following beneficial effects:

[0071] The present invention utilizes a spectrum acquisition device 5 to collect the spectrum of river crabs. The near-infrared spectrometer is more sensitive to the differences between different species of river crabs and can effectively obtain the spectral information of river crabs. According to the detection results, the crabs are transported to different crab outlets 12 for crab discharge by utilizing the cooperation of the crab carrying box 9 and the movable slide 2. At the same time, the present invention constructs a river crab recognition model based on a convolutional neural network, which can perform fast and accurate non-destructive classification detection on different species of river crabs.

[0072] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments, and the above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, and these changes and improvements all fall within the scope of the present invention to be protected.

Claims

1. A movable river crab rapid classification and detection device, comprising a movable box (1), It is characterized in that A slide motor (3) and a crab outlet (12) are respectively installed on both sides of the movable box (1); a movable slide (2) is installed inside the movable box (1); a slider (7) movable along the direction of the movable slide (2) is installed on the movable slide (2); and a crab carrying box (9) movable along the direction of the slider (7) is arranged on the slider (7); A spectrum collection device (5) is arranged inside the movable box (1) and above the movable slide (2). The spectrum collection device (5) is used to collect spectrum information of crabs in the crab carrying box (9). A Raspberry Pi (10) is installed in the movable box (1). The spectrum collection device (5) and the Raspberry Pi (10) are electrically connected and are used to transmit the collected spectrum information to the Raspberry Pi (10) through optical fiber. A position sensor (11) is installed on the movable slide (2). The position sensor (11) and the Raspberry Pi (10) are electrically connected and are used to feed back the position information of the crab carrying box (9) to the Raspberry Pi (10), so that the Raspberry Pi (10) can adjust and control the position of the crab carrying box (9).

2. The movable crab rapid classification and detection device according to claim 1, It is characterized in that The movable box (1) is rectangular and made of stainless steel. The inside of the movable box (1) is sprayed with black paint, and a universal pulley is installed at the bottom of the movable box (1).

3. The movable crab rapid classification and detection device according to claim 2, It is characterized in that A slider motor (8) is fixedly mounted on the slider (7); an output end of the slider motor (8) and a crab carrying box (9) are fixedly mounted to drive the crab carrying box (9) to slide horizontally on the slider (7).

4. The movable crab rapid classification and detection device according to claim 3, It is characterized in that The two sides of the movable box (1) are respectively provided with a slide motor (3) and a crab outlet (12), and a plurality of crab outlets (12) are provided for classifying and discharging river crabs from different origins. A photoelectric sensor (13) is installed on the inner wall of the crab outlet (12), and the photoelectric sensor (13) is a slot-type photoelectric sensor.

5. The movable crab rapid classification and detection device according to claim 1, It is characterized in that A spectrum device bracket (6) is installed inside the movable box (1), and the spectrum collection device (5) is fixedly installed on the spectrum device bracket (6). A Raspberry Pi display screen (14) is installed on the door panel of the movable box (1), and the Raspberry Pi display screen (14) is electrically connected to the Raspberry Pi (10) for controlling and displaying the setting parameters of the Raspberry Pi (10).

6. The movable rapid classification and detection device for river crabs according to claim 1, It is characterized in that The outer surface of the crab carrying box (9) is equipped with a crab carrying box motor (18); the output end of the crab carrying box motor (18) is fixedly connected to the crab carrying box (9) with a push plate (17); the push plate (17) is used to control the entry and exit of river crabs; a pneumatic plate (16) is arranged on the crab carrying box (9); the pneumatic plate (16) of the air pump (4) is connected to control the opening and closing of the pneumatic plate (16); a pneumatic module (19) is installed on one side of the crab carrying box (9); the pneumatic module (19) and the air pump (4) are electrically connected.

7. Rapid classification and detection method for river crabs, It is characterized in that The method for rapid classification and detection of river crabs is used in the movable rapid classification and detection device for river crabs to classify and detect river crabs. The method for rapid classification and detection of river crabs comprises the following steps: S1. Build a collection platform for river crab spectra, and collect river crab spectrum information through the collection platform; S2, processing the collected spectral information of river crabs to obtain a two-dimensional spectral feature matrix; S3. Building a river crab classification model based on a convolutional neural network and a two-dimensional spectral feature matrix, and training the river crab classification model; S4. Process the spectral information of the crabs to be classified and convert it into a two-dimensional spectral feature matrix. Input the two-dimensional spectral feature matrix into the trained crab classification model to complete the detection of different species of crabs and classify them.

8. The method for rapid classification and detection of river crabs according to claim 7, It is characterized in that The method for rapid classification and detection of river crabs is executed by a Raspberry Pi (10), and at the same time, instructions are sent to each execution device to classify the river crabs being tested.

9. The method for rapid classification and detection of river crabs according to claim 7, It is characterized in that In step S2, the collected crab spectral information is processed, including the following steps: S21, intercepting effective spectral information, and performing SG smoothing and standard normal variable transformation for preprocessing to obtain a sample set; S22, labeling the sample set, and using the SPXY algorithm to divide the sample set into a training set and a test set; S23, selecting characteristic wavelength points for samples in the training set using the CARS algorithm; S24. Based on the characteristic wavelength points selected from the training set samples, a specific method for constructing a two-dimensional spectral feature matrix based on the characteristic wavelength points includes: obtaining a one-dimensional spectral feature matrix based on the characteristic wavelength points, and multiplying the one-dimensional spectral feature matrix with the transposed matrix of the one-dimensional spectral feature matrix to obtain a two-dimensional spectral feature matrix.

Images