Durian condition grading system

Abstract

The invention discloses a durian condition grading system, which comprises a feeding line, a detection device and shunting conveying lines, the feeding line is connected with a plurality of shunting conveying lines through the detection device, and the detection device comprises a base, and a visual identification system, a transplanting mechanical arm, a beating mechanism and a stem cutting and pesticide spraying mechanism which are arranged on the base; the visual recognition system is used for recognizing the position of a fruit stem of the durian, the transplanting mechanical arm grabs the fruit stem of the durian under the guidance of the visual recognition system and then transfers the fruit stem to the position of the beating mechanism, the durian shell is beaten through the beating mechanism, and the transplanting mechanical arm is provided with a sound collection and analysis system. The sound acquisition and analysis system performs spectral analysis according to the feedback sound to judge the fruit maturity, and on the premise that the durian is judged to be mature enough, the transplanting mechanical arm conveys the durian to the stem cutting and pesticide spraying mechanism to cut off redundant fruit stems and spray insect prevention liquid medicine; and finally, according to different fruit grades, the fruits are transferred to different shunting conveying lines to the next process. The method can realize automatic detection and grading of durian quality.

Description

Technical Field

[0001] The invention relates to a fruit appearance grading device, in particular to a durian appearance grading system. Background Art

[0002] With the increase in the production capacity of the durian planting industry, the grading and testing of durian fruits are currently mainly carried out manually. Fruit dealers / processing factories first use their naked eyes to identify the number of petals in the durian flesh, whether the fruit shape is relatively regular, and whether the fruit thorns are standard. Then they use a special stick to knock on the fruit shell to judge whether the fruit is ripe enough by listening to the sound. The ripe fruit will make a lower sound when knocked. According to the above standards, the durian is divided into ABCD grades and then sent to different processing workshops.

[0003] All the above items are currently operated manually, which is labor-intensive and relies on subjective judgment by humans. The standards are inconsistent and there are problems such as misjudgment. Summary of the invention

[0004] The purpose of the present invention is to overcome the shortcomings of the prior art and provide a durian appearance grading system, which can realize automatic detection and grading of durian quality, greatly reduce the labor intensity of manual work, and ensure the uniformity of durian grading standards.

[0005] The technical solution of the present invention is as follows:

[0006] A durian grading system comprises a feeding line, a detection device and a plurality of shunt conveying lines, wherein the feeding line is connected with the plurality of shunt conveying lines through the detection device, the detection device comprises a base and a visual recognition system, a transplanting mechanical arm, a knocking mechanism and a handle cutting and spraying mechanism arranged on the base, the visual recognition system is close to the feeding end of the detection device and is used to identify the position of the fruit handle of the durian, the transplanting mechanical arm, the knocking mechanism and the handle cutting and spraying mechanism are respectively located on both sides of the feeding line, the transplanting mechanical arm grasps the fruit handle of the durian under the guidance of the visual recognition system, and then transfers to the position of the knocking mechanism, and knocks the durian shell through the knocking mechanism, the transplanting mechanical arm is provided with a sound collection and analysis system, and the sound collection and analysis system determines the maturity of the fruit according to the frequency spectrum analysis of the feedback sound, and on the premise of determining that the durian is sufficiently mature, the transplanting mechanical arm then sends the durian to the handle cutting and spraying mechanism to cut off the excess fruit handle and spray insect repellent, and finally transfers to different shunt conveying lines to the next process according to different fruit grades.

[0007] Furthermore, the detection device also includes a protective shell and an operating screen. The protective shell cover is arranged on the base to cover the visual recognition system, the transplanting robot arm, the knocking mechanism, and the cutting handle spraying mechanism. The operating screen is arranged on the front of the protective shell.

[0008] Further, the durians are placed on a tray for feeding.

[0009] Furthermore, the visual recognition system includes a first 3D camera and a second 3D camera, wherein the first 3D camera is located on one side of the feeding line and is used to identify the 3D shape of the durian, and the second 3D camera is located above the feeding line and is used to identify the position of the durian stalk.

[0010] Furthermore, the transplanting robot arm is a six-axis robot arm.

[0011] Furthermore, the striking mechanism includes a motor seat, a rotating motor, a rotating wheel, a striking plate and a spring. The rotating motor is arranged in the motor seat, the motor shaft of the rotating motor faces upward and is connected to the rotating wheel, the striking plate is pivotally arranged on the motor seat through a pivot, and the striking plate is connected to the rotating wheel through a spring.

[0012] Furthermore, the sound collection and analysis system includes a probe and a microphone, a data collection module, a microphone and a data analysis module. The probe and the microphone, the data collection module, the microphone and the data analysis module are connected in sequence. The probe and the microphone are moved close to the durian knocking position by a transplanting robotic arm, and a high level is sent to the data collection module. The data collection module triggers sampling, completes the electroacoustic signal collection of the microphone under the excitation of a sweep frequency meter, and transmits the data to the data analysis module. The data analysis module completes the electroacoustic signal analysis and outputs a judgment result according to the detection model.

[0013] Furthermore, the microphone adopts MPA201 microphone, and the data acquisition module adopts USB8812 data acquisition card.

[0014] Furthermore, the data analysis module is divided into a data layer, a model layer and a detection layer. The data layer includes a production database and a training database. The production database is connected to the training database through a data marking module. The model layer collects and analyzes early data samples, including noise classification, feature extraction, and integrated learning to build a machine learning model, and then outputs the machine learning model to the detection layer. The detection layer includes an intelligent detection module. The production database transmits the data to be tested to the detection layer through a data navigation module, determines the maturity of the fruit according to the machine learning model, and transmits the detection results to the intelligent detection module, and generates a detection log and stores the detection log in the production database through the data navigation module.

[0015] Furthermore, the handle cutting and spraying mechanism comprises handle cutting scissors and a spray nozzle, and the redundant fruit handles are cut off by the handle cutting scissors. The spray nozzle is arranged above the handle cutting scissors and is used for spraying insect repellent.

[0016] Compared with the prior art, the beneficial effects of the present invention are as follows: the present invention can realize automatic feeding, automatic identification of durian quality and automatic detection and grading, which effectively solves the problems of inconsistent durian grading standards, easy misjudgment and high labor intensity in the existing manual work. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0018] Figure 1 A schematic diagram of the appearance of a durian quality grading system provided by the present invention;

[0019] Figure 2 An internal schematic diagram of a durian quality grading system provided by the present invention;

[0020] Figure 3 It is a structural schematic diagram of the striking mechanism of the present invention;

[0021] Figure 4 This is a system composition block diagram of the sound collection and analysis system of the present invention;

[0022] Figure 5 This is an architecture diagram of the data analysis module of the present invention;

[0023] Figure 6 It is a structural schematic diagram of the handle cutting and spraying mechanism of the present invention. DETAILED DESCRIPTION

[0024] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0025] In order to illustrate the technical solution of the present invention, a specific embodiment is provided below for illustration.

[0026] Example

[0027] See also Figure 1 , Figure 2The present embodiment provides a durian grading system, including a feed line 1, a detection device 2 and a plurality of shunt conveying lines 3, wherein the feed line 1 is connected to the plurality of shunt conveying lines 3 through the detection device 2. Specifically, the detection device 2 includes a base 21, a protective shell 22, a visual recognition system 23, a transplanting mechanical arm 24, a knocking mechanism 25, a cutting handle spraying mechanism 26 and an operating screen 27, wherein the visual recognition system 23, the transplanting mechanical arm 24, the knocking mechanism 25, and the cutting handle spraying mechanism 26 are arranged on the base 21, and the protective shell 22 is covered on the base 21 to cover the visual recognition system 23, the transplanting mechanical arm 24, the knocking mechanism 25, and the cutting handle spraying mechanism 26, and the operating screen 27 is arranged on the front of the protective shell 22.

[0028] Among them, the visual recognition system 23 is close to the feeding end of the detection device 2, and is used to identify the position of the durian stalk. The transplanting robot arm 24, the knocking mechanism 25, and the stalk cutting and spraying mechanism 26 are respectively located on both sides of the feeding line 1. The transplanting robot arm 24 grabs the stalk of the durian under the guidance of the visual recognition system 23, and then transfers it to the position of the knocking mechanism 25, and knocks the durian shell through the knocking mechanism 25. The transplanting robot arm 24 is provided with a sound collection and analysis system. The sound collection and analysis system determines the maturity of the fruit based on the spectrum analysis of the feedback sound. On the premise that the durian is sufficiently mature, the transplanting robot arm 24 sends the durian to the stalk cutting and spraying mechanism 26 to cut off the excess stalk and spray insect repellent. Finally, according to different fruit grades, it is transferred to different diversion conveying lines 3 to the next process.

[0029] The visual recognition system 23 includes a first 3D camera 231 and a second 3D camera 232. The first 3D camera 231 is located on one side of the feeding line 1 and is used to identify the 3D shape of the durian. The second 3D camera 232 is located above the feeding line 1 and is used to identify the position of the stalk of the durian.

[0030] The transplanting robot arm 24 is a six-axis robot arm that can realize the functions of circulation, transplanting and transportation of multiple workstations.

[0031] Combination Figure 3 As shown, the knocking mechanism 25 includes a motor seat 251, a rotating motor 252, a rotating wheel 253, a knocking plate 254 and a spring 255. The rotating motor 252 is arranged in the motor seat 251, and the motor shaft of the rotating motor 252 faces upward and is connected to the rotating wheel 253. The knocking plate 254 is pivotally arranged on the motor seat 251 through a pivot. The knocking plate 254 and the rotating wheel 253 are connected by a spring 255. The knocking plate 254 is driven by the rotating motor 252 to knock the durian, and the spring 255 has a buffering effect.

[0032] Combination Figure 4As shown, the sound collection and analysis system includes a probe and a microphone, a data collection module, a microphone and a data analysis module. The probe and the microphone, the data collection module, the microphone and the data analysis module are connected in sequence. The probe and the microphone are moved close to the durian knocking position by the transplanting mechanical arm 24, and a high level is sent to the data collection module. The data collection module triggers sampling, completes the electroacoustic signal collection of the microphone under the excitation of the sweep frequency meter, and transmits the data to the data analysis module. The data analysis module completes the electroacoustic signal analysis and outputs the judgment result according to the detection model.

[0033] The microphone adopts MPA201 microphone, which is a combination of 1 / 2 inch prepolarized free-field microphone and preamplifier.

[0034] The data acquisition module adopts a USB8812 data acquisition card, which is a high-precision data acquisition card designed for testing audio and vibration signals.

[0035] like Figure 5 As shown, the data analysis module is divided into a data layer, a model layer and a detection layer. The data layer includes a production database and a training database. The production database is connected to the training database through a data labeling module. The model layer collects and analyzes early data samples, including noise classification, feature extraction, and integrated learning to build a machine learning model, and then outputs the machine learning model to the detection layer. The detection layer includes an intelligent detection module. The production database transmits the data to be tested to the detection layer through the data navigation module, determines the maturity of the fruit according to the machine learning model, and transmits the detection results to the intelligent detection module, and generates a detection log and stores the detection log in the production database through the data navigation module.

[0036] Combination Figure 6 As shown, the handle cutting and spraying mechanism 26 includes handle cutting scissors 261 and a spraying nozzle 262. The redundant fruit handles are cut off by the handle cutting scissors 261. The spraying nozzle 262 is arranged above the handle cutting scissors 261 for spraying insect repellent.

[0037] The specific workflow is as follows:

[0038] First, the durian 10 is manually placed on a tray 20, and then the tray 20 is fed into the detection device 2 through the feeding line 1; the second 3D camera 232 identifies the position of the stalk of the durian 10, and the transplanting robot 24 identifies the stalk of the durian 10 under the guidance of the second 3D camera 232, and then grabs the stalk and hangs the durian 10 in front of the first 3D camera 231 to identify the number of pulp petals; then the durian 10 is transferred to the knocking mechanism 25, and the rotating motor 252 drives the knocking plate 25 4 The shell of the durian 10 is knocked, and the shell feeds back sounds with different frequency characteristics. The sound collection and analysis system performs spectrum analysis based on the fed-back sounds to determine the maturity of the fruit; on the premise that the durian 10 is sufficiently mature, the transplanting robot arm 24 sends the durian 10 to the stalk cutting and spraying mechanism 26, and the stalk cutting scissors 261 cut off the excess stalks, sprays insect repellent after cutting off the stalks, and then puts them back on the tray 20; the tray 20 flows to different diversion conveyor lines 3 to the next process according to different fruit grades.

[0039] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A durian grading system, characterized in that: The invention comprises a feeding line, a detection device and a plurality of shunt conveying lines, wherein the feeding line is connected with the plurality of shunt conveying lines through the detection device, the detection device comprises a base and a visual recognition system, a transplanting mechanical arm, a knocking mechanism and a handle cutting and spraying mechanism arranged on the base, the visual recognition system is close to the feeding end of the detection device and is used to identify the position of the fruit handle of the durian, the transplanting mechanical arm, the knocking mechanism and the handle cutting and spraying mechanism are respectively located on both sides of the feeding line, the transplanting mechanical arm grasps the fruit handle of the durian under the guidance of the visual recognition system, and then transfers to the position of the knocking mechanism, and knocks the durian shell through the knocking mechanism, the transplanting mechanical arm is provided with a sound collection and analysis system, and the sound collection and analysis system determines the maturity of the fruit according to the spectrum analysis of the feedback sound, and on the premise of determining that the durian is sufficiently mature, the transplanting mechanical arm then sends the durian to the handle cutting and spraying mechanism to cut off the excess fruit handle and spray insect repellent, and finally transfers to different shunt conveying lines to the next process according to different fruit grades.

2. A durian product grading system according to claim 1, characterized in that: The detection device also includes a protective shell and an operation screen. The protective shell cover is arranged on the base to cover the visual recognition system, the transplanting mechanical arm, the knocking mechanism, and the cutting handle spraying mechanism. The operation screen is arranged on the front of the protective shell.

3. A durian product grading system according to claim 1, characterized in that: Durian is placed on a tray for feeding.

4. A durian product grading system according to claim 1, characterized in that: The visual recognition system includes a first 3D camera and a second 3D camera. The first 3D camera is located on one side of the feeding line and is used to identify the 3D shape of the durian. The second 3D camera is located above the feeding line and is used to identify the position of the durian stalk.

5. A durian product grading system according to claim 1, characterized in that: The transplanting robot arm is a six-axis robot arm.

6. A durian product grading system according to claim 1, characterized in that: The striking mechanism includes a motor seat, a rotating motor, a rotating wheel, a striking plate and a spring. The rotating motor is arranged in the motor seat, the motor shaft of the rotating motor faces upward and is connected to the rotating wheel, the striking plate is pivotally arranged on the motor seat through a pivot, and the striking plate is connected to the rotating wheel through a spring.

7. A durian product grading system according to claim 1, characterized in that: The sound collection and analysis system includes a probe and a microphone, a data collection module, a microphone and a data analysis module. The probe and the microphone, the data collection module, the microphone and the data analysis module are connected in sequence. The probe and the microphone are moved close to the durian knocking position by a transplanting mechanical arm, and a high level is sent to the data collection module. The data collection module triggers sampling, completes the electroacoustic signal collection of the microphone under the excitation of a sweep frequency meter, and transmits the data to the data analysis module. The data analysis module completes the electroacoustic signal analysis and outputs a judgment result according to the detection model.

8. A durian product grading system according to claim 6, characterized in that: The microphone adopts MPA201 microphone, and the data acquisition module adopts USB8812 data acquisition card.

9. A durian product grading system according to claim 6, characterized in that: The data analysis module is divided into a data layer, a model layer and a detection layer. The data layer includes a production database and a training database. The production database is connected to the training database through a data marking module. The model layer collects and analyzes early data samples, including noise classification, feature extraction, and integrated learning to build a machine learning model, and then outputs the machine learning model to the detection layer. The detection layer includes an intelligent detection module. The production database transmits the data to be tested to the detection layer through a data navigation module, determines the maturity of the fruit according to the machine learning model, and transmits the detection results to the intelligent detection module, and generates a detection log and stores the detection log in the production database through the data navigation module.

10. A durian grading system according to claim 1, characterized in that: The handle cutting and spraying mechanism comprises handle cutting scissors and a spray head. The redundant fruit handles are cut off by the handle cutting scissors. The spray head is arranged above the handle cutting scissors and is used for spraying insect repellent.

Images