CN111179266A

CN111179266A - Artificial intelligence analysis system for tobacco insects

Info

Publication number: CN111179266A
Application number: CN202010039484.6A
Authority: CN
Inventors: 范建建; 谷永茂; 郑晓耘; 李晓磊
Original assignee: Hebei Runyi Electromechanical Technology Co ltd
Current assignee: Hebei Runyi Electromechanical Technology Co ltd
Priority date: 2020-01-15
Filing date: 2020-01-15
Publication date: 2020-05-19

Abstract

The invention relates to the technical field of industrial equipment, in particular to an artificial intelligent analysis system for tobacco insects, which adopts the technical design of an embedded system integrated camera and a wireless transmission method, solves the technical problems of time consumption, labor consumption and low working efficiency of the traditional artificial tobacco insect detection, and achieves the technical effects of time and labor saving by directly observing results through an artificial intelligent analysis method; by adopting a total-branch total-modularization design method such as an MCU unit, a GPRS module and a digital camera, the technical problems of complex system and difficult management are solved, and the technical effects of clear division of labor and simple management and maintenance of the whole cigarette worm monitoring system are achieved; the technical design of artificial intelligence image recognition analysis is adopted, so that the analysis error caused by the fact that the insect plate is manually checked for a long time is effectively avoided, and the technical effects of automatic analysis, intelligent recording, data self-organizing storage and accuracy rate guarantee are achieved.

Description

Artificial intelligence analysis system for tobacco insects

Technical Field

The invention relates to the technical field of industrial equipment, in particular to an artificial intelligence analysis system for tobacco worms.

Background

The cigarette is a food which can be sucked, the quality safety of the product becomes one of the key points of attention of consumers along with the gradual enhancement of health consciousness of the consumers, and the quality of the cigarette can be greatly influenced when the cigarette is damaged by tobacco worms in the cigarette production stage. After the cigarettes are damaged by pests, dead bodies, ova, excrement and filaments of the cigarettes pollute the cigarettes, and the cigarettes are smoked to generate foul smell, so that the internal quality of the cigarette products is influenced, the satisfaction degree and market rate of consumers are reduced, and claim disputes are possibly caused. Meanwhile, the cigarette packaging material will affect the appearance quality after being wormhole. Therefore, effective protection against insects and smoke insects detection is of paramount importance.

The cigarette worms as harmful substances have great influence on cigarettes, and consumers are very dislike to cigarettes damaged by the worms, so cigarette manufacturers need to strengthen the control of the cigarette worms, the detection amount of the cigarette worms in different seasons and different parts is different from the analysis of the detection amount of the cigarette worms, and the cigarette worm killing is carried out in time according to the detection and analysis results, so that the cigarette worm killing method is an important means for ensuring the quality of the cigarettes.

The existing cigarette insect analysis is that each box is opened on site through people, a plurality of insects on the insect staining plate are recorded on paper, the large insect staining plate in the workshop area is wide in distribution, and the method consumes time and labor and fails to analyze results and records integrally.

Disclosure of Invention

The technical problems to be solved by the invention are that aiming at the technical defects, the artificial intelligent analysis system for the tobacco pests is provided, the technical design that an embedded system integrates a camera and a wireless transmission method is adopted, the technical problems that the traditional artificial tobacco pest detection is time-consuming and labor-consuming and the working efficiency is low are solved, and the technical effects that the result can be directly observed by the artificial intelligent analysis method and the time and the labor are saved are achieved; by adopting a total-branch total-modularization design method such as an MCU unit, a GPRS module and a digital camera, the technical problems of complex system and difficult management are solved, and the technical effects of clear division of labor and simple management and maintenance of the whole cigarette worm monitoring system are achieved; the technical design of artificial intelligent image recognition and analysis is adopted, so that the analysis error caused by the fact that the insect plate is manually checked for a long time is effectively avoided, and the technical effects of automatic analysis, intelligent recording, data self-organizing and storing and accuracy rate guaranteeing are achieved; by adopting the technical design of GPRS network wireless connection and networking type wireless intelligent transmission management, the technical problems of complex wiring and easy interference are solved, and the equipment achieves the intelligent, planned and reasonable monitoring effect.

In order to solve the technical problems, the technical scheme adopted by the invention is as follows: the system comprises a workshop, an insect attracting monitoring board, a GPRS terminal, an Internet network and a GPRS network; a GPRS terminal and a plurality of insect trapping monitoring boards are installed in the workshop, and the insect trapping monitoring boards are connected to a GPRS network through the GPRS terminal; one end of the GPRS network interacts with the Internet network, and one end of the GPRS network is connected with a monitoring mobile phone; a remote control platform is connected to the Internet network;

the insect attracting monitoring board comprises an insect attracting board and a wireless image acquisition device; the insect attracting plate is arranged on the lower side of the wireless image acquisition device; the insect attracting plate is of a plate-shaped structure, and transparent glue is coated on the insect attracting plate; a plurality of drops of insect-attracting paste are coated on the transparent adhesive; the wireless image acquisition device is of a box-type structure and comprises a circuit board, a battery, a digital camera and an antenna; the circuit board is arranged in the wireless image acquisition device, and the antenna is arranged at the top end of the wireless image acquisition device; the battery and the digital camera are arranged on the circuit board; the camera end of the digital camera is aligned with the insect attracting plate;

further optimizing the technical scheme, the circuit board comprises an MCU unit, a signal processing channel, a GPRS module, a power supply module, a control output module and a Bluetooth module; the I/O port of the MCU unit is connected with a GPRS module and a memory, and the MCU unit is connected with a Bluetooth module in parallel through the GPRS module; one side of the MCU unit is connected with a signal processing channel, and the input end of the signal processing channel is connected with a sensor; the other side of the MCU unit is connected with a control output module, and the output end of the control output module is connected with a display screen; the output side of the power supply module is connected with a power pin of the MCU unit, and the input side of the power supply module is connected with a battery; the MCU unit is connected with the GPRS terminal through the GPRS module;

further optimizing the technical scheme, the input end of the antenna is arranged at the signal output end of the GPRS module;

further optimizing the technical scheme, the sensor comprises a temperature sensor and a humidity sensor;

further optimizing the technical scheme, the signal processing channel comprises a filtering amplifying circuit and an A/D conversion circuit; the input end of the filtering amplification circuit is connected with the input ends of the temperature sensor and the humidity sensor; the output end of the filtering amplifying circuit is connected with the input end of the A/D conversion circuit; the output end of the input end of the A/D conversion circuit is connected with the MCU;

further optimizing the technical scheme, the MCU is an STM32F103RBT6 chip, and the control output module is a TFT _ ILI9341 chip; a DB pin of the control output module is connected with a PB pin of the MCU; the CS pin of the control output module is connected with the PC9 pin of the MCU unit; the RD pin of the control output module is connected with the PC10 pin of the MCU unit; the WR pin of the control output module is connected with the PC11 pin of the MCU unit; the RS pin of the control output module is connected with the PC12 pin of the MCU unit; the RST pin of the control output module is connected with the PC8 pin of the MCU unit;

further optimizing the technical scheme, the A/D conversion circuit is an ADC0809 conversion circuit chip;

further optimizing the technical scheme, the digital camera is an OV7670 camera; the data pin of the digital camera is connected with the PA pin of the MCU; the power supply module comprises an AMS1117 chip, a plurality of capacitors and a light-emitting diode; the input voltage of the power supply module is +5V, and the output voltage of the power supply module is +3.3V

Further optimizing the technical scheme, the remote control platform is internally provided with image analysis software, and the software functions comprise image preprocessing, image segmentation, image refinement and edge detection, image feature extraction, image feature calculation, error analysis and result output.

Compared with the prior art, the invention has the following advantages: 1. the device adopts a total value modular design, the task of the cigarette worm detection device is clear, automatic control, automatic image processing and analysis and result output are realized, and the device has a simple and convenient structure; 2. ADC0809 has the characteristics of high speed, high precision, low temperature drift, excellent long-term precision, repeatability and low power consumption, and is provided with a plurality of serial ports, so that the device has an extensible function, and a sensing device can be added according to the requirements of customers; 3. the device saves a large amount of manual time when working, is installed in a wireless mode during installation, does not need a wired power supply to carry a battery, and saves the trouble of wiring; 4. the device can also detect and observe the state of the insect-attracting plate in a short distance by using the Bluetooth of the mobile phone, and is convenient and quick; 5. the device saves a large amount of manual time when in work, and effectively avoids visual fatigue caused by manually checking the insect plate for a long time and analysis errors.

Drawings

FIG. 1 is a diagram of a wireless transmission architecture;

FIG. 2 is a view of the structure of the insect attracting plate;

FIG. 3 is a side cutaway view of a wireless image capture device;

FIG. 4 is a front view of a wireless image capture device;

FIG. 5 is a schematic block diagram of a circuit board;

FIG. 6 is a diagram of a signal processing channel architecture;

FIG. 7 is a flowchart of an image artificial intelligence recognition process;

FIG. 8 is a circuit configuration diagram of an MCU unit;

FIG. 9 is a circuit configuration diagram of the single chip microcomputer unit;

FIG. 10 is a circuit block diagram of a power module;

fig. 11 is a circuit configuration diagram of a signal processing channel.

In the figure, 1, a workshop; 2. a pest trapping monitoring plate; 3. a GPRS terminal; 4. an Internet network; 5. a GPRS network; 6. a remote control platform; 7. monitoring the mobile phone; 8. a pest trapping plate; 9. transparent glue; 10. insect attracting paste; 11. a wireless image acquisition device; 12. a circuit board; 13. a battery; 14. a digital camera; 15. an antenna; 16. an MCU unit; 17. a sensor; 18. a signal processing channel; 19. a control output module; 20. a display screen; 21. a memory; 22. a Bluetooth module; 23. a power supply module; 24. a GPRS module; 25. an A/D conversion circuit; 26. a temperature sensor; 27. a humidity sensor; 28. and a filtering and amplifying circuit.

Detailed Description

In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention will be described in further detail with reference to the accompanying drawings in conjunction with the following detailed description. It should be understood that the description is intended to be exemplary only, and is not intended to limit the scope of the present invention. Moreover, in the following description, descriptions of well-known structures and techniques are omitted so as to not unnecessarily obscure the concepts of the present invention.

The first embodiment is as follows: as shown in fig. 1 to 11, the system comprises a workshop 1, an insect attracting monitoring board 2, a GPRS terminal 3, an Internet network 4 and a GPRS network 5; a GPRS terminal 3 and a plurality of insect attracting monitoring boards 2 are installed in the workshop 1, and the insect attracting monitoring boards 2 are connected to a GPRS network 5 through the GPRS terminal 3; one end of the GPRS network 5 is interacted with the Internet network 4, and the other end is connected with a monitoring mobile phone 7; the Internet network 4 is connected with a remote control platform 6;

the insect trapping monitoring board 2 comprises an insect trapping board 8 and a wireless image acquisition device 11; the insect attracting plate 8 is arranged at the lower side of the wireless image acquisition device 11; the insect attracting plate 8 is of a plate-shaped structure, and transparent glue 9 is coated on the insect attracting plate 8; a plurality of drops of insect-attracting paste 10 are coated on the transparent adhesive 9; the wireless image acquisition device 11 is of a box-type structure, and the wireless image acquisition device 11 comprises a circuit board 12, a battery 13, a digital camera 14 and an antenna 15; the circuit board 12 is arranged inside the wireless image acquisition device 11, and the antenna 15 is arranged at the top end of the wireless image acquisition device 11; a battery 13 and a digital camera 14 are mounted on the circuit board 12; the camera end of the digital camera 14 is aligned with the insect attracting plate 8;

the circuit board 12 comprises an MCU unit 16, a signal processing channel 18, a GPRS module 24, a power module 23, a control output module 19 and a Bluetooth module 22; an I/O port of the MCU unit 16 is connected with a GPRS module 24 and a memory 21, and the MCU unit 16 is connected with a Bluetooth module 22 in parallel through the GPRS module 24; one side of the MCU unit 16 is connected with a signal processing channel 18, and the input end of the signal processing channel 18 is connected with a sensor 17; the other side of the MCU unit 16 is connected with a control output module 19, and the output end of the control output module 19 is connected with a display screen 20; the output side of the power supply module 23 is connected with a power pin of the MCU unit 16, and the input side of the power supply module 23 is connected with the battery 13; the MCU unit 16 is connected with the GPRS terminal 3 through a GPRS module 24;

the input end of the antenna 15 is arranged at the signal output end of the GPRS module 24; the sensor 17 comprises a temperature sensor 26 and a humidity sensor 27;

the signal processing channel 18 comprises a filter amplifying circuit 28 and an A/D conversion circuit 25; the input end of the filter amplifying circuit 28 is connected with the input ends of the temperature sensor 26 and the humidity sensor 27; the output end of the filter amplifying circuit 28 is connected with the input end of the A/D conversion circuit 25; the output end of the input end of the A/D conversion circuit 25 is connected with the MCU unit 16;

the MCU unit 16 is an STM32F103RBT6 chip, and the control output module 19 is a TFT _ ILI9341 chip; the DB pin of the control output module 19 is connected with the PB pin of the MCU unit 16; the CS pin of the control output module 19 is connected with the PC9 pin of the MCU unit 16; the RD pin of the control output module 19 is connected with the PC10 pin of the MCU unit 16; the WR pin of the control output module 19 is connected with the PC11 pin of the MCU unit 16; the RS pin of the control output module 19 is connected with the PC12 pin of the MCU unit 16; the RST pin of the control output module 19 is connected with the PC8 pin of the MCU unit 16;

the A/D conversion circuit 25 is an ADC0809 conversion circuit chip; the digital camera 14 is an OV7670 camera; the data pin of the digital camera 14 is connected with the PA pin of the MCU unit 16; the power module 23 comprises an AMS1117 chip, a plurality of capacitors and a light emitting diode; the input voltage of the power module 23 is +5V, and the output voltage of the power module 23 is + 3.3V;

the remote control platform 6 is internally provided with image analysis software, and the software functions comprise image preprocessing, image segmentation, image thinning and edge detection, image feature extraction, image feature calculation, error analysis and result output;

the specific operation steps of the wireless monitoring device for the tobacco insects are as follows:

step 1: a plurality of insect trapping monitoring boards 2 distributed in the workshop 1 are in a dormant state for a long time, and only a clock works; during the time period, the insect attracting plate 8 in the insect attracting monitoring plate 2 attracts the tobacco insects to be attached to the insect attracting plate 8 through the insect attracting paste 10; the insect attracting function of the insect attracting plate 8 is realized according to the following principle: as shown in the structure of the insect attracting plate 8 in fig. 2, transparent glue is coated on the insect attracting plate 8, and a drop of insect attracting paste 10 is smeared on the transparent glue to attract the tobacco insects; when the cigarette insect taste is the taste of the insect attracting paste 10, the cigarette insect is close to the insect attracting plate 8, and after the cigarette insect flies to the insect attracting plate 8, the transparent adhesive 9 attached to the insect attracting plate 8 sticks the cigarette insect, so that the premise is provided for realizing image acquisition;

step 2: as shown in fig. 1, when a specified time (which can be set) is reached or a specified time interval is set, the pest trapping monitoring board 2 is remotely activated by an upper computer, namely the remote control platform 6, to be in a working state, namely the power of the wireless image acquisition device 11 in the pest trapping monitoring board 2 is switched on, and the control board controls the digital camera 14 to perform multiple times (to prevent errors) of image acquisition on the pest trapping board 8 through the MCU unit 16;

and step 3: the collected images are sorted and collected by the circuit board 12; the signal is transmitted by the wireless device connection antenna 15, and the GPRS terminal 3 located in the workshop receives the signal, as shown in fig. 1; the working principle of the circuit board 12 is as follows: as shown in fig. 5, it is an internal structure diagram of the circuit board 12, in which the MCU unit 16 is a core processing unit, which monitors the states of various components in the system through the sensor 17 and the digital camera 14; secondly, the display screen 20 is controlled to display images and the quantity of the tobacco insects through the control output module 19; the other side of the MCU unit 16 is connected with a mobile phone in a short distance through a Bluetooth module 22; finally, the MCU 16 is wirelessly connected to the GPRS network 3 through the GPRS module 24, so as to realize the image transmission and calculate the quantity of the tobacco insects at regular time; in fig. 6, the sensor 17 includes a temperature sensor 26 and a humidity sensor 27, which implement monitoring of ambient temperature and humidity;

and 4, step 4: the other side of the GPRS terminal 3 is connected with a GPRS network, collected images are numbered and then transmitted to the GPRS network 5, the GPRS network 5 interacts with an Internet network, and at the moment, the images are uploaded to the network; the user can receive the collected images through a remote control platform 6, such as a PC (personal computer), a monitoring mobile phone 7 and the like;

and 5: the images are collected, but the data results such as the quantity of the tobacco insects and the like can be obtained only after the images are processed, so that image recognition software attached to the device needs to be downloaded by an upper computer, and the collected images are intelligently recognized by artificial intelligence software;

step 6: the image is automatically downloaded and received by the upper computer software; as shown in fig. 7, the steps of the software for performing artificial intelligence automatic identification are as follows: I. firstly, preprocessing an image, adjusting color and contrast, and judging whether the image is a qualified image or not; II. The image is segmented, and useless parts (outside the insect attracting plate) are deleted, so that the effectiveness of the image is guaranteed; III, thinning and edge detection are carried out on the image, so that the definition of the image is ensured, and the image characteristics are conveniently identified; identifying image characteristics, namely a cigarette worm monitoring device which is mainly used for observing the number of the cigarette worms on the insect attracting plate 8, wherein the image characteristics are the shapes of the individual cigarette worms, namely the number of the cigarette worms is judged by extracting the characteristics, namely the shapes of the cigarette worms, and the number is counted; v, calculating characteristics to finish the judgment and calculation of the quantity of the tobacco insects; VI, secondary feature calculation is carried out, and the calculation accuracy of the device is guaranteed; if the error is larger, analyzing again; VII, outputting a tobacco insect picture and a quantity result;

and 7: when the quantity of the tobacco worms is counted at intervals, a curve graph of the quantity of the tobacco worms changing along with the change of time is calculated, and a result and a change curve are output on a remote control platform;

and 8: the remote control platform controls the dormancy of the insect trapping monitoring board 2;

and step 9: a user observes the quantity and the change rule of all the insects luring plates 8 in the whole workshop 1 through the remote control platform 6; or the monitoring mobile phone 7 is used for downloading data for watching;

the device mainly analyzes the quantity of the tobacco insects on the insect attracting plate by intelligent analysis software of an upper computer; the time for personnel to go to the field for collecting and recording is saved, and the automatic analysis can be realized through artificial intelligence software. The visual fatigue caused by manually checking the insect plate for a long time is effectively avoided, and the analysis error is avoided. The accuracy of the result is ensured.

Secondly, fig. 8 is a circuit structure diagram of a single chip microcomputer system, the system uses an STM32 single chip microcomputer system as a CPU chip, integrates a memory with a large capacity and abundant and powerful hardware interface circuits, has extremely high performance and price ratio, and greatly increases the feasibility and practicability of the system; and the digital camera 14 adopts an OV7670 chip and an image sensor, has small volume and low working voltage, and provides all functions of a single-chip VGA camera and an image processor. Through the control of the SCCB bus, 8-bit image data with various resolutions such as whole frame, sub-sampling, window taking and the like can be input. The VGA image of the product is up to 30 frames/sec. The user can control the image quality, data format and transmission completely. In fig. 11, the circuit adopts a fourth-order band-pass circuit as a filter circuit, so that harmonic signals in the circuit are effectively weakened, and the quality of output signals is ensured.

It is to be understood that the above-described embodiments of the present invention are merely illustrative of or explaining the principles of the invention and are not to be construed as limiting the invention. Therefore, any modification, equivalent replacement, improvement and the like made without departing from the spirit and scope of the present invention should be included in the protection scope of the present invention. Further, it is intended that the appended claims cover all such variations and modifications as fall within the scope and boundaries of the appended claims or the equivalents of such scope and boundaries.

Claims

1. Artificial intelligence analytic system of tobacco worm, its characterized in that: the system comprises a workshop (1), an insect attracting monitoring board (2), a GPRS terminal (3), an Internet network (4) and a GPRS network (5); a GPRS terminal (3) and a plurality of insect trapping monitoring boards (2) are installed in the workshop (1), and the insect trapping monitoring boards (2) are connected to a GPRS network (5) through the GPRS terminal (3); one end of the GPRS network (5) is interacted with the Internet network (4), and the other end is connected with a monitoring mobile phone (7); the Internet network (4) is connected with a remote control platform (6); the insect trapping monitoring board (2) comprises an insect trapping board (8) and a wireless image acquisition device (11); the insect attracting plate (8) is arranged at the lower side of the wireless image acquisition device (11); the insect attracting plate (8) is of a plate-shaped structure, and transparent glue (9) is coated on the insect attracting plate (8); a plurality of drops of insect-attracting paste (10) are coated on the transparent adhesive (9); the wireless image acquisition device (11) is of a box-type structure, and the wireless image acquisition device (11) comprises a circuit board (12), a battery (13), a digital camera (14) and an antenna (15); the circuit board (12) is arranged inside the wireless image acquisition device (11), and the antenna (15) is arranged at the top end of the wireless image acquisition device (11); the battery (13) and the digital camera (14) are arranged on the circuit board (12); the camera end of the digital camera (14) is aligned with the insect attracting plate (8).

2. The artificial intelligence analysis system for tobacco worms of claim 1 further comprising: the circuit board (12) comprises an MCU unit (16), a signal processing channel (18), a GPRS module (24), a power supply module (23), a control output module (19) and a Bluetooth module (22); an I/O port of the MCU unit (16) is connected with a GPRS module (24) and a memory (21), and the MCU unit (16) is connected with a Bluetooth module (22) in parallel through the GPRS module (24); one side of the MCU unit (16) is connected with a signal processing channel (18), and the input end of the signal processing channel (18) is connected with a sensor (17); the other side of the MCU unit (16) is connected with a control output module (19), and the output end of the control output module (19) is connected with a display screen (20); the output side of the power supply module (23) is connected with a power pin of the MCU unit (16), and the input side of the power supply module (23) is connected with the battery (13); the MCU unit (16) is connected with the GPRS terminal (3) through the GPRS module (24).

3. The artificial intelligence analysis system for tobacco worms of claims 1 and 2 further comprising: the input end of the antenna (15) is arranged at the signal output end of the GPRS module (24).

4. The artificial intelligence analysis system for tobacco worms of claim 2 wherein: the sensor (17) includes a temperature sensor (26) and a humidity sensor (27).

5. The artificial intelligence analysis system for tobacco worms of claim 1 further comprising: the signal processing channel (18) comprises a filtering and amplifying circuit (28) and an A/D conversion circuit (25); the input end of the filter amplifying circuit (28) is connected with the input ends of the temperature sensor (26) and the humidity sensor (27); the output end of the filtering amplification circuit (28) is connected with the input end of the A/D conversion circuit (25); the output end of the input end of the A/D conversion circuit (25) is connected with the MCU unit (16).

6. The artificial intelligence analysis system for tobacco worms of claim 1 further comprising: the MCU unit (16) is an STM32F103RBT6 chip, and the control output module (19) is a TFT _ ILI9341 chip; a DB pin of the control output module (19) is connected with a PB pin of the MCU unit (16); the CS pin of the control output module (19) is connected with the PC9 pin of the MCU unit (16); the RD pin of the control output module (19) is connected with the PC10 pin of the MCU unit (16); the WR pin of the control output module (19) is connected with the PC11 pin of the MCU unit (16); the RS pin of the control output module (19) is connected with the PC12 pin of the MCU unit (16); the RST pin of the control output module (19) is connected with the PC8 pin of the MCU unit (16).

7. The artificial intelligence analysis system for tobacco worms of claim 2 wherein: the A/D conversion circuit (25) is an ADC0809 conversion circuit chip.

8. The artificial intelligence analysis system for tobacco worms of claim 1 further comprising: the digital camera (14) is an OV7670 camera; and a data pin of the digital camera (14) is connected with a PA pin of the MCU unit (16).

9. The artificial intelligence analysis system for tobacco worms of claim 2 wherein: the power supply module (23) comprises an AMS1117 chip, a plurality of capacitors and a light-emitting diode; the input voltage of the power supply module (23) is +5V, and the output voltage of the power supply module (23) is + 3.3V.

10. The artificial intelligence analysis system for tobacco worms of claim 1 further comprising: image analysis software is installed in the remote control platform (6), and the software functions comprise image preprocessing, image segmentation, image thinning and edge detection, image feature extraction, image feature calculation, error analysis and result output.