Harvester stone removal device, method and harvester

Abstract

The present invention provides a harvester stone removal device, method and harvester. A harvester stone removal device includes: an elevator, a detection mechanism, a stone removal mechanism, a separation chute and a processor, wherein the detection mechanism and the stone removal mechanism are both installed at the output end of the elevator, the separation chute is located below the stone removal mechanism, and the detection mechanism and the stone removal mechanism are both connected to the processor. During the seed melon lifting and transporting stage, the detection mechanism can identify the mixing of stones early on, and then remove the identified stones, achieving the effect of active stone removal, thereby preventing stones from entering the machine and causing damage to the machine, achieving shutdown-free and maintenance-free operation, and thus greatly improving operating efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of stone removal for harvesters, and in particular to a stone removal device and method for harvesters, and a harvester. Background Art

[0002] Seed melon is an agricultural product with extremely regional characteristics and a very important economic crop, mainly used for its seeds. Similar seed-extracting economic crops include gourds, seed melons, pumpkins, etc.

[0003] At present, the harvesting of melons and other fruit-based agricultural products has been largely mechanized. A melon-seed harvester is a specialized harvesting machine used to harvest melon-seed agricultural products, including gourds, melon-seed, pumpkins, and similar seed-producing crops. When the harvester is operating, the pick-up and transport platform in front of the harvester picks up melon-seed agricultural products from the ground and transports them to a crusher to crush the fruits and remove the seeds. The crusher housing, melon-pressing roller, and roller blade work closely together to cut and crush the melon rind between the crusher housing and the melon-pressing roller, sending it to the subsequent threshing and sauce-spinning drum. The high-speed rotation of the threshing and sauce-spinning drum will throw the melon seeds out of the drum, thereby achieving the effect of separating the melon seeds from the melon rind. However, when picking up melons and other fruit products and loading them into the machine via the pick-up and conveyor platform, rocks from the ground are also carried along with the melons into the conveyor platform and into the machine. These hard rocks can damage the high-speed rotating rollers and melon-pressing rollers or get stuck between them, suffocating the machine. Damage to the melon-pressing rollers and rollers can lead to a decline in harvesting quality, and suffocating the machine can necessitate interruption of the harvesting process, requiring the machine to be shut down for maintenance, which is also time-consuming and labor-intensive. Frequent entry of rocks results in frequent machine downtime for maintenance, severely impacting operational efficiency. Therefore, the melon-seed harvesting industry urgently needs an automatic stone removal control system to automatically remove stones from the melon harvest process. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a harvester stone removal device, method and harvester in view of the deficiencies in the prior art.

[0005] The technical solution of the present invention to solve the above technical problems is as follows: A stone discharge device for a harvester, comprising: an elevator, a detection mechanism, a stone discharge mechanism, a separation chute and a processor, the detection mechanism and the stone discharge mechanism are both installed at the output end of the elevator, the separation chute is located below the stone discharge mechanism, and the detection mechanism and the stone discharge mechanism are both connected to the processor.

[0006] The beneficial effect of adopting the technical solution of the present invention is that stones can be identified early on by the detection mechanism during the seed melon lifting and transportation stage, and the identified stones can be removed, achieving the effect of active stone removal, thereby preventing stones from entering the machine and causing damage to the machine, eliminating the need for downtime and maintenance, and greatly improving work efficiency.

[0007] Furthermore, the detection mechanism includes: an x-ray source and an amorphous silicon flat panel detector, the x-ray source is located above the elevator, the amorphous silicon flat panel detector is located below the elevator, and the x-ray source and the amorphous silicon flat panel detector are arranged opposite to each other.

[0008] The beneficial effect of adopting the above-mentioned further technical solution is that during the stage of transporting the melon seeds, the presence of stones can be identified early by x-rays, and the identified stones can be removed, achieving the effect of active stone removal, thereby preventing stones from entering the machine and causing damage to the machine, eliminating downtime and maintenance, and greatly improving operating efficiency. The x-ray solution is selected to detect stones. The x-ray solution is particularly suitable for detecting situations where there is a difference in density between products and foreign objects. Obviously, the density of stones is much greater than that of fruits and vegetables, so they can be easily detected by x-rays. The x-ray solution includes an x-ray source and an amorphous silicon flat-panel detector. The amorphous silicon flat-panel detector can output the detection results in the form of a grayscale image. The color of the grayscale image is darker for stones with higher density. With the help of image morphological processing methods, a binary threshold segmentation algorithm is applied to the grayscale image to distinguish the position of the denser stones on the grayscale image. Then, combined with the expansion and corrosion morphological method, domain connectivity processing is performed to identify the outline and position of the stone.

[0009] Furthermore, the processor is provided with a binary threshold segmentation algorithm, and the processor is a 2 GHz CPU processor.

[0010] The beneficial effects of adopting the above-mentioned further technical solution are: using a binary threshold segmentation algorithm to directly identify the location of foreign objects, without the need to prepare a grayscale template in advance. The binary threshold segmentation algorithm can cover all actual operation scenarios, is faster, more accurate in recognition, more suitable for vehicle-mounted applications, and more suitable for non-stop operations. On a 2GHz CPU processor, the time spent processing an image is less than 0.002s. Compared with other algorithms such as image feature matching algorithms and deep model recognition algorithms, the computing speed is very fast and the accuracy rate is maintained at above 95%. The use of a binary threshold segmentation algorithm has a faster computing speed and low computing power requirements, making it more suitable for on-board online real-time applications without affecting operational efficiency.

[0011] Furthermore, the stone removal mechanism includes: multiple pushing components, multiple stone removal baffles and a rotating shaft, the tops of the multiple stone removal baffles are rotatably installed on the rotating shaft, the multiple pushing components are respectively connected to the multiple stone removal baffles, and the multiple pushing components are connected to the processor.

[0012] The beneficial effect of adopting this further technical solution is that after the stone and its location are identified, the subsequent stone removal action is controlled by an air valve to eject the pushing component, which pushes the stone removal baffle. The stone removal baffle strikes the stone, knocking it out of the machine. This reduces the loss rate during foreign object removal. The stone removal baffles are divided into multiple locations. When a stone appears in a corresponding position, only the stone removal baffle in that position is ejected, leaving the other stone removal baffles unchanged, preventing the removal of useful fruits and vegetables.

[0013] Furthermore, the pushing component is a pneumatic cylinder, an electric cylinder or a hydraulic cylinder.

[0014] The beneficial effect of adopting the above-mentioned further technical solution is that the stone-removing action is performed by a cylinder, which has a high speed and can ensure the working speed without affecting the working efficiency of the whole machine due to the stone removal. It is convenient for users to select the type of pushing component according to actual needs.

[0015] Furthermore, the separation chute is an inverted V-shaped structure, one side of the bottom of the separation chute is connected to a crusher box, and the other side of the bottom of the separation chute is connected to a stone collecting trough; the elevator is a scraper elevator, and the scraper elevator is provided with an elevator conveyor belt.

[0016] The beneficial effect of adopting the above-mentioned further technical solution is that a separation chute is provided at the entrance of the crusher housing, and stones that have changed their trajectory will fall into the stone collecting trough along the separation chute and will not enter the crusher housing, thus realizing active stone discharge. The stones are discharged into the stone collecting trough along the separation chute, rather than being discharged directly onto the ground, and can also be used to pick up and collect stones in the field, so that stones can be picked up and collected online at the same time. This allows the entire machine to not only have the function of stone discharge during the seed melon harvesting operation, but also the function of picking up field stones, achieving multiple uses for one machine; and as the seed melons are harvested and field stones are picked up each year, the number of field stones can be reduced year by year.

[0017] In addition, the present invention also provides a harvester, comprising a harvester stone removal device as described in any one of the above.

[0018] The beneficial effect of adopting the technical solution of the present invention is that stones can be identified early on by the detection mechanism during the seed melon lifting and transportation stage, and the identified stones can be removed, achieving the effect of active stone removal, thereby preventing stones from entering the machine and causing damage to the machine, eliminating the need for downtime and maintenance, and greatly improving work efficiency.

[0019] In addition, the present invention also provides a harvester stone discharge method, which is based on a harvester stone discharge device described in any one of the above items. The harvester stone discharge method includes: S1, a detection mechanism detects the transported objects on the elevator to generate a grayscale image; S2, a processor combines an image morphological algorithm to apply a binary threshold segmentation algorithm to the grayscale image to distinguish the positions of stones with higher density on the grayscale image; S3, the processor combines an expansion corrosion morphological method to perform domain connectivity processing to identify the outline and actual position of the stone; S4, after a delay time, the stone discharge mechanism actively discharges the stone.

[0020] The beneficial effects of the technical solution of the present invention are as follows: an X-ray detector is used to detect when a stone enters the machine. The detector will output the detection result in the form of a grayscale image. Fruits and stones of different densities have different brightness levels in the grayscale image. The shape of stones with higher density is much darker than the fruits in the grayscale image. Then, combined with the image morphology algorithm, a binary threshold segmentation algorithm is applied to the grayscale image to distinguish the position of the denser stones in the grayscale image. Then, combined with the expansion corrosion morphology method, domain connectivity processing is performed to identify the outline and position of the stone. In this way, the control system can sense the stone and its position. After the appropriate delay time, the stone removal baffle can just hit the moving stone, changing the stone's motion trajectory, thereby achieving the separation of the stone and fruit's motion trajectory. The use of a binary threshold segmentation algorithm has a faster computing speed and low computing power requirements. It is more suitable for on-board online real-time applications and will not affect work efficiency.

[0021] Furthermore, before step S1, it includes: adjusting the exposure time of the amorphous silicon flat panel detector and adjusting the segmentation threshold of the binary threshold segmentation algorithm until the amorphous silicon flat panel detector can distinguish between fruits and stones; step S4 includes: after the delay time, the stone discharge mechanism drives the pushing component corresponding to the actual position of the stone to pop out, push the stone discharge baffle to hit the stone, change the falling trajectory of the stone, and make the stone fall into the stone collecting trough along the separation chute, thereby realizing active stone discharge.

[0022] The benefits of adopting this further technical solution include: Before use, the X-ray detector sensitivity needs to be adjusted. First, the exposure time of the amorphous silicon flat-panel detector is changed; increasing the exposure time enhances image contrast. Second, the threshold of the binarization threshold segmentation algorithm is adjusted. Through parameter calibration, the balance between exposure time and segmentation threshold is adjusted until the detector can clearly distinguish between rocks and fruits. By adjusting the exposure time and segmentation threshold to set the sensitivity for different terrain conditions and fruit types, the maximum recognition accuracy is achieved for each terrain type, improving the rock rejection rate.

[0023] Furthermore, the delay time in step S4 is calculated by the following formula: T=Δt+t1, where T is the delay time, Δt is the image processing time, and t1 is the time it takes for the stone to leave the output end of the elevator and fall to the stone discharge baffle;

[0024] The time t1 when the stone leaves the elevator output end and falls to the stone discharge baffle is calculated by the following formula: Among them, t1 is the time it takes for the stone to leave the output end of the elevator and fall to the stone discharge baffle, h1 is the height from the output end of the elevator to the stone discharge baffle, and g is the gravitational acceleration of the stone.

[0025] The beneficial effect of this further technical solution is that when a stone reaches the detector position and is detected, the stone ejection mechanism, after a certain delay, controls the air valve to open, and compressed air pushes the cylinder to eject, causing the stone ejection baffle to rotate around the upper rotation axis. After the appropriate delay, the stone ejection baffle is precisely positioned to strike the incoming stone, changing its trajectory and thus separating the stone from the fruit. A formula is used to calculate the delay time and the time it takes for the stone to leave the elevator output and fall to the stone ejection baffle, improving accuracy.

[0026] Advantages of additional aspects of the present invention will be given in part in the following description and in part will be obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 A schematic diagram of the principle of a stone removal device for a harvester provided in an embodiment of the present invention.

[0028] Figure 2 This is one of the schematic flow charts of the stone removal method for a harvester provided in an embodiment of the present invention.

[0029] Figure 3 This is the second schematic flow chart of the stone removal method for a harvester provided in an embodiment of the present invention.

[0030] Explanation of the accompanying figures: 1. Elevator; 2. Detection mechanism; 3. Stone discharge mechanism; 4. Separation chute; 5. X-ray source; 6. Amorphous silicon flat panel detector; 7. Pushing component; 8. Stone discharge baffle; 9. Rotating shaft; 10. Crusher box. DETAILED DESCRIPTION

[0031] The principles and features of the present invention are described below with reference to the accompanying drawings. The embodiments given are only used to explain the present invention and are not used to limit the scope of the present invention.

[0032] like Figure 1As shown, a stone discharge device for a harvester includes: an elevator 1, a detection mechanism 2, a stone discharge mechanism 3, a separation chute 4 and a processor, the detection mechanism 2 and the stone discharge mechanism 3 are both installed at the output end of the elevator 1, the separation chute 4 is located below the stone discharge mechanism 3, and the detection mechanism 2 and the stone discharge mechanism 3 are both connected to the processor.

[0033] The beneficial effect of adopting the technical solution of the present invention is that stones can be identified early on by the detection mechanism during the seed melon lifting and transportation stage, and the identified stones can be removed, achieving the effect of active stone removal, thereby preventing stones from entering the machine and causing damage to the machine, eliminating the need for downtime and maintenance, and greatly improving work efficiency.

[0034] The present invention provides a stone removal device for a harvester, which can be a stone removal control system for a seed melon harvester. This device not only includes a detection mechanism for foreign objects such as stones, but also a foreign object removal mechanism. The control system's operating platform (the harvester's stone removal device) is based on a scraper-type elevator (the elevator). Fruits and stones picked up by a picker and cutter are conveyed via this elevator conveyor belt (the elevator's conveyor belt) into a crusher housing. An X-ray detector (an X-ray source and an amorphous silicon flat-panel detector) is located at the end of the elevator belt (the elevator's conveyor belt). This serves as the sensor for the control system (the harvester's stone removal device) and can detect any stones entering the machine. Upon reaching the end of the elevator, the fruit falls into the crusher housing located below. As the fruit falls, the air cylinder (the cylinder) of the actuator (the stone removal mechanism) pops out under the control of an air valve, pushing the stone removal baffle to rotate around its upper rotation axis, thereby striking the stone and changing its trajectory. A stone separation chute (separation chute) is provided at the entrance of the crusher box. The stones after changing their motion trajectory will fall into the stone collecting trough along the separation chute and will not enter the crusher box, thus realizing active stone discharge.

[0035] 1. The technical solution provided by the embodiment of the present invention can detect foreign objects such as stones in real time online and identify their center of mass position. Combined with the subsequent cylinder stone discharge actuator (stone discharge mechanism) structure, it can realize active stone discharge without stopping the machine, thereby protecting the machine from damage by stones.

[0036] 2. For stone detection, the grayscale value information of the X-Ray output grayscale image is directly used for binary threshold segmentation. Compared with other algorithms such as image feature matching algorithms and deep model recognition algorithms, the operation speed is very fast, with a processing time of less than 0.002s, and the work efficiency will not be reduced due to stone recognition.

[0037] 3. The air cylinder (pneumatic cylinder) is used to perform the stone removal operation. The cylinder has a high speed, which can ensure the operating speed and will not affect the operating efficiency of the entire machine due to stone removal.

[0038] 4. During the process of harvesting seed watermelons, the whole machine can not only remove stones, but also pick up stones in the field, so that one machine can be used for multiple purposes; and as the seed watermelons are harvested and stones are picked up in the field every year, the number of stones in the field can be reduced year by year.

[0039] 5. The stone removal function of the harvesting machine can be used not only on the seed watermelon harvesting machine, but also on other machines such as potato harvesters, potato harvesters, and tomato harvesters where stones may enter during operation and cause damage to the machine.

[0040] Furthermore, the detection mechanism 2 includes: an x-ray source 5 and an amorphous silicon flat panel detector 6, the x-ray source 5 is located above the elevator 1, and the amorphous silicon flat panel detector 6 is located below the elevator 1, and the x-ray source 5 and the amorphous silicon flat panel detector 6 are arranged opposite to each other.

[0041] The beneficial effect of adopting the above-mentioned further technical solution is that during the stage of transporting the melon seeds, the presence of stones can be identified early by x-rays, and the identified stones can be removed, achieving the effect of active stone removal, thereby preventing stones from entering the machine and causing damage to the machine, and achieving continuous operation without stopping the machine, thereby greatly improving operating efficiency. The x-ray solution is selected to detect stones. The x-ray solution is particularly suitable for detecting situations where there is a difference in density between products and foreign objects. It is obvious that the density of stones is much greater than that of fruits and vegetables, so they can be easily detected by x-rays. The x-ray solution includes an x-ray source and an amorphous silicon flat-panel detector. The amorphous silicon flat-panel detector can output the detection results in the form of a grayscale image. The color of the denser stones is darker in the grayscale image. With the help of image morphological processing methods, a binary threshold segmentation algorithm is applied to the grayscale image to distinguish the position of the denser stones on the grayscale image. Then, combined with the expansion and corrosion morphological method, domain connectivity processing is performed to identify the outline and position of the stones.

[0042] Furthermore, the processor is provided with a binary threshold segmentation algorithm, and the processor is a 2 GHz CPU processor.

[0043] The beneficial effects of adopting the above-mentioned further technical solution are: using a binary threshold segmentation algorithm to directly identify the location of foreign objects, without the need to prepare a grayscale template in advance. The binary threshold segmentation algorithm can cover all actual operation scenarios, is faster, more accurate in recognition, more suitable for vehicle-mounted applications, and more suitable for non-stop operations. On a 2GHz CPU processor, the time spent processing an image is less than 0.002s. Compared with other algorithms such as image feature matching algorithms and deep model recognition algorithms, the computing speed is very fast and the accuracy rate is maintained at above 95%. The use of a binary threshold segmentation algorithm has a faster computing speed and low computing power requirements, making it more suitable for on-board online real-time applications without affecting operational efficiency.

[0044] Furthermore, the stone-removing mechanism 3 includes: multiple pushing components 7, multiple stone-removing baffles 8 and a rotating shaft 9. The tops of the multiple stone-removing baffles 8 are rotatably mounted on the rotating shaft 9. The multiple pushing components 7 are respectively connected to the multiple stone-removing baffles 8, and the multiple pushing components 7 are all connected to the processor.

[0045] The beneficial effect of adopting this further technical solution is that after the stone and its location are identified, the subsequent stone removal action is controlled by an air valve to eject the pushing component, which pushes the stone removal baffle. The stone removal baffle strikes the stone, knocking it out of the machine. This reduces the loss rate during foreign object removal. The stone removal baffles are divided into multiple locations. When a stone appears in a corresponding position, only the stone removal baffle in that position is ejected, leaving the other stone removal baffles unchanged, preventing the removal of useful fruits and vegetables.

[0046] Furthermore, the pushing component 7 is a pneumatic cylinder, an electric cylinder or a hydraulic cylinder.

[0047] The beneficial effect of adopting the above-mentioned further technical solution is that the stone-removing action is performed by a cylinder, which has a high speed and can ensure the working speed without affecting the working efficiency of the whole machine due to the stone removal. It is convenient for users to select the type of pushing component according to actual needs.

[0048] In this embodiment of the present invention, a cylinder is used as a pushing component.

[0049] Furthermore, the separation chute 4 is an inverted V-shaped structure, one side of the bottom of the separation chute 4 is connected to the crusher box 10, and the other side of the bottom of the separation chute 4 is connected to the stone collecting trough; the elevator 1 is a scraper elevator, and the scraper elevator is provided with a lifting conveyor belt.

[0050] The beneficial effect of adopting the above-mentioned further technical solution is that a separation chute is provided at the entrance of the crusher housing, and stones that have changed their trajectory will fall into the stone collecting trough along the separation chute and will not enter the crusher housing, thus realizing active stone discharge. The stones are discharged into the stone collecting trough along the separation chute, rather than being discharged directly onto the ground, and can also be used to pick up and collect stones in the field, so that stones can be picked up and collected online at the same time. This allows the entire machine to not only have the function of stone discharge during the seed melon harvesting operation, but also the function of picking up field stones, achieving multiple uses for one machine; and as the seed melons are harvested and field stones are picked up each year, the number of field stones can be reduced year by year.

[0051] An embodiment of the present invention provides an active stone removal solution; stones are mixed into the machine when melon seeds and other fruits on the ground are picked up by a picking platform and brought into the machine. The melon seeds are then transported to the crusher box through an elevator. During the melon seed lifting and transportation stage, the presence of stones is identified early by X-Ray, and the identified stones are removed, achieving the effect of active stone removal, thereby preventing stones from entering the machine and causing damage to the machine, achieving shutdown-free and maintenance-free operation, and thus greatly improving operating efficiency.

[0052] The design of the active stone removal system for the seed melon harvester consists of two main parts: first, it must be able to reliably identify stones that have entered the machine, and second, it must be able to accurately remove the identified stones. Stone recognition is also called foreign object detection.

[0053] When melons and other fruits are harvested, the tumbling fruits are pushed into the machine by a conveyor (elevator). The fruits and stones are in various postures and may overlap with each other, with stones hidden on the back of the fruits. Therefore, the present invention selects an x-ray solution to detect stones. The x-ray solution is particularly suitable for detecting situations where there is a difference in density between the product (which can be fruits) and foreign objects (which can be stones). Obviously, the density of stones is much greater than that of fruits, so they can be easily detected by x-rays. The x-ray solution includes an x-ray source and an amorphous silicon flat panel detector. The detector (amorphous silicon flat panel detector 6) can output the detection results in the form of a grayscale image. The color of the denser stones on the grayscale image is darker. With the help of image morphological processing methods, a binary threshold segmentation algorithm is applied to the grayscale image to distinguish the position of the denser stones on the grayscale image. Then, combined with the expansion and corrosion morphological method, domain connectivity processing is performed to identify the outline and position of the stones. The embodiment of the present invention adopts a binary threshold segmentation algorithm to directly identify the position of foreign objects (stones) without the need to prepare a grayscale template in advance. The binary threshold segmentation algorithm can cover all actual operation scenarios, is faster, more accurate in recognition, more suitable for vehicle-mounted applications, and more suitable for non-stop operations.

[0054] After the stone and its position are identified, the subsequent stone discharge action is controlled by the air valve to eject the high-speed cylinder (cylinder), the cylinder pushes the stone discharge baffle, and the baffle (stone discharge baffle) hits the stone, and discharges the stone out of the machine by hitting the stone.

[0055] The embodiment of the present invention adopts an X-Ray-based foreign body detection solution to provide an output grayscale image. The image processing algorithm adopts a binary threshold segmentation algorithm, which has a faster computing speed and low computing power requirements. It is more suitable for on-board online real-time applications and will not affect operational efficiency.

[0056] While the present invention's technical solution uses X-ray detectors to identify the presence and location of rocks, optical color sorting can also be used. However, its application is limited, and in some work scenarios where color distinctions are unclear, the accuracy of color sorting is significantly reduced. The stone ejection action is then performed using a pneumatic cylinder ejection method, which can also be achieved using electric cylinders, springs, or other structures. However, compared to the pneumatic cylinder method used in the present invention, this is slower and reduces operational efficiency.

[0057] 1. During the harvesting process, the seed melon harvester proposed in the present invention uses an x-ray scheme to actively identify stones, and then uses a cylinder to push the stone removal baffle to eject the stones, thereby realizing the active stone removal function of the seed melon harvester.

[0058] 2. According to different terrain conditions and fruit types, the sensitivity can be set by adjusting the exposure time and segmentation threshold, so as to set the maximum recognition accuracy for different terrain conditions and improve the stone removal rate.

[0059] 3. The binary threshold segmentation algorithm can quickly identify stones mixed into the machine. The algorithm has a fast processing speed and will not reduce the operation speed. It is very suitable for online real-time detection.

[0060] 4. It also includes the use of air cylinders (cylinders) to perform the stone removal function, which is fast and can ensure the operating speed without affecting the operating efficiency of the entire machine.

[0061] 5. The stone-removal baffle structure reduces the loss rate during foreign object removal. Multiple stone-removal baffles are used. When a stone appears in a corresponding position, only the stone-removal baffle at that position is pushed out, while the other stone-removal baffles remain stationary, preventing the removal of useful fruits and vegetables. This embodiment of the present invention can use three stone-removal baffles instead of a single stone-removal baffle. If a single stone-removal baffle is used, the presence of a single stone will cause the entire operating area of ​​fruits and vegetables to be removed, resulting in greater loss of useful fruits and vegetables.

[0062] In addition, the present invention also provides a harvester, comprising a harvester stone removal device as described in any one of the above.

[0063] The beneficial effect of adopting the technical solution of the present invention is that the detection mechanism can identify the mixing of stones early in the seed melon lifting and transportation stage, and then remove the identified stones, achieving the effect of active stone removal, thereby preventing stones from entering the machine and causing damage to the machine, realizing shutdown-free and maintenance-free operation, thereby greatly improving operating efficiency.

[0064] like Figure 2 As shown, in addition, the present invention also provides a harvester stone discharge method, based on a harvester stone discharge device described in any one of the above, the harvester stone discharge method includes: S1, the detection mechanism detects the transported objects on the elevator and generates a grayscale image; S2, the processor combines the image morphology algorithm to apply a binary threshold segmentation algorithm to the grayscale image to distinguish the positions of stones with higher density on the grayscale image; S3, the processor combines the expansion corrosion morphology method to perform domain connectivity processing to identify the outline and actual position of the stone; S4, after the delay time, the stone discharge mechanism actively discharges the stone.

[0065] The beneficial effects of the technical solution of the present invention are as follows: an X-ray detector is used to detect when a stone enters the machine. The detector will output the detection result in the form of a grayscale image. Fruits and stones of different densities have different brightness levels in the grayscale image. The shape of stones with higher density is much darker than the fruits in the grayscale image. Then, combined with the image morphology algorithm, a binary threshold segmentation algorithm is applied to the grayscale image to distinguish the position of the denser stones in the grayscale image. Then, combined with the expansion corrosion morphology method, domain connectivity processing is performed to identify the outline and position of the stone. In this way, the control system can sense the stone and its position. After the appropriate delay time, the stone removal baffle can just hit the moving stone, changing the stone's motion trajectory, thereby achieving the separation of the stone and fruit's motion trajectory. The use of a binary threshold segmentation algorithm has a faster computing speed and low computing power requirements. It is more suitable for on-board online real-time applications and will not affect work efficiency.

[0066] Furthermore, before step S1, it includes: adjusting the exposure time of the amorphous silicon flat panel detector and adjusting the segmentation threshold of the binary threshold segmentation algorithm until the amorphous silicon flat panel detector can distinguish between fruits and stones; step S4 includes: after the delay time, the stone discharge mechanism drives the pushing component corresponding to the actual position of the stone to pop out, push the stone discharge baffle to hit the stone, change the falling trajectory of the stone, and make the stone fall into the stone collecting trough along the separation chute, thereby realizing active stone discharge.

[0067] The benefits of adopting this further technical solution include: Before use, the X-ray detector sensitivity needs to be adjusted. First, the exposure time of the amorphous silicon flat-panel detector is changed; increasing the exposure time enhances image contrast. Second, the threshold of the binarization threshold segmentation algorithm is adjusted. Through parameter calibration, the balance between exposure time and segmentation threshold is adjusted until the detector can clearly distinguish between rocks and fruits. By adjusting the exposure time and segmentation threshold to set the sensitivity for different terrain conditions and fruit types, the maximum recognition accuracy is achieved for each terrain type, improving the rock rejection rate.

[0068] Furthermore, the delay time in step S4 is calculated by the following formula:

[0069] T=Δt+t1,

[0070] Where T is the delay time, Δt is the image processing time, and t1 is the time it takes for the stone to leave the output end of the elevator and fall to the stone discharge baffle;

[0071] The time t1 when the stone leaves the elevator output end and falls to the stone discharge baffle is calculated by the following formula:

[0072]

[0073] Among them, t1 is the time it takes for the stone to leave the output end of the elevator and fall to the stone discharge baffle, h1 is the height from the output end of the elevator to the stone discharge baffle, and g is the gravitational acceleration of the stone.

[0074] The beneficial effect of this further technical solution is that when a stone reaches the detector position and is detected, the stone ejection mechanism, after a certain delay, controls the air valve to open, and compressed air pushes the cylinder to eject, causing the stone ejection baffle to rotate around the upper rotation axis. After the appropriate delay, the stone ejection baffle is precisely positioned to strike the incoming stone, changing its trajectory and thus separating the stone from the fruit. A formula is used to calculate the delay time and the time it takes for the stone to leave the elevator output and fall to the stone ejection baffle, improving accuracy.

[0075] The control system (a stone removal device for a harvester) provided by an embodiment of the present invention uses an X-ray detector (detection mechanism) to detect rocks entering the machine. The detector (detection mechanism) outputs the detection result as a grayscale image. Fruits and rocks of different densities appear different shades of light and dark in the grayscale image, with denser rocks appearing much darker than the fruits. A binary threshold segmentation algorithm, combined with an image morphology algorithm, is then applied to the grayscale image to distinguish the locations of denser rocks. A domain connectivity process, combined with the dilation-erosion morphology method, is then performed to identify the outline and location of the rocks. This allows the control system (a stone removal device for a harvester) to sense rocks and their locations. Before use, the sensitivity of the X-ray detector (detection mechanism) must be adjusted. First, the exposure time of the flat-panel detector (detection mechanism) must be changed; increasing the exposure time can enhance image contrast. Second, the threshold (segmentation threshold) of the binary threshold segmentation algorithm must be adjusted. Parameter calibration is used to adjust the balance between the exposure time and the segmentation threshold until the detector (detection mechanism) can clearly distinguish between rocks and fruits. When the stone reaches the position of the detector (detection mechanism) and is detected by the sensor (detection mechanism), the control system (harvester stone discharge device) controls the air valve to open after a certain delay, and the compressed air pushes the cylinder to pop out, pushing the stone discharge baffle to rotate around the upper rotating shaft.

[0076] T = Δt + t1, where Δt is the image processing time and t1 is the time it takes for the stone to leave the end of the elevator and fall to the stone discharge baffle.

[0077] On a 2GHz CPU processor, the time spent on processing an image is less than 0.002s. Compared with other algorithms such as image feature matching algorithms and deep model recognition algorithms, the computing speed is very fast, and the accuracy rate is maintained at above 95%. For the stone removal of fruit harvesters, even if fruits are occasionally identified as stones and discharged, the loss of one or two fruits is of no concern to the system.

[0078] After a proper delay of T, the stone-discharging baffle can just hit the moving stones, changing their trajectory, thus separating the stones from the fruits. The stones are discharged into the stone collection trough along the separation chute instead of directly onto the ground. This also serves to pick up and collect stones in the field, achieving the effect of picking up and collecting stones online.

[0079] like Figure 3As shown in the figure, 1. After the system is started, the detection system sensitivity is calibrated by adjusting the exposure time and segmentation threshold according to the types of fruits and stones in the field until the fruits and stones can be clearly distinguished; 2. The X-ray detector detects the transmission grayscale images of the fruits and stones on the lifting conveyor belt in real time, and determines whether there are stones with higher density and their locations through image morphology methods; 3. After detecting a stone, the control system drives the cylinder at the corresponding position to pop out, pushing the stone discharge baffle to hit the stone, changing its falling trajectory so that it falls into the stone collecting trough along the separation chute, realizing active stone discharge.

[0080] 1. Use X-Ray to identify stones during the seed melon harvest process. After identifying the stones, use a cylinder to push the stone removal baffle to remove the stones.

[0081] 2. Use air cylinder (pneumatic cylinder) as the pusher of the stone discharge actuator (stone discharge mechanism), which has high speed and can ensure that the operation of the whole machine does not need to reduce the operating speed. Other solutions include electric push-pull rod solution, electric cylinder solution and other actuators.

[0082] 3. Use multiple baffles (stone removal baffles) to remove stones and reduce the loss rate of fruits and vegetables. If a whole baffle is used, the appearance of a single stone will force the removal of fruits and vegetables across the entire width of the conveyor belt (elevator), resulting in unnecessary loss and waste.

[0083] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A stone removal method for a harvester, characterized in that: include: S1. The detection mechanism detects the transported objects on the elevator and generates a grayscale image; S2. The processor applies a binary threshold segmentation algorithm to the grayscale image in combination with an image morphology algorithm to distinguish the locations of denser stones on the grayscale image; S3, the processor combines the expansion and corrosion morphology method to perform domain connectivity processing and identify the outline and actual position of the stone; S4: After the delay time, the stone discharge mechanism will automatically discharge the stones; Before step S1, the process includes: adjusting the exposure time of the amorphous silicon flat panel detector and adjusting the segmentation threshold of the binary threshold segmentation algorithm until the amorphous silicon flat panel detector can distinguish between fruits and stones; Step S4 includes: after the delay time, the stone discharge mechanism drives the pushing component corresponding to the actual position of the stone to pop out, pushes the stone discharge baffle to hit the stone, changes the falling trajectory of the stone, and makes the stone fall into the stone collecting trough along the separation chute, realizing active stone discharge.

2. A stone removal method for a harvester according to claim 1, characterized in that: The delay time in step S4 is calculated by the following formula: ， in, is the delay time, is the image processing time, The time it takes for the stone to leave the output end of the elevator and fall to the stone discharge baffle; The time it takes for the stone to leave the output end of the elevator and fall to the stone discharge baffle Calculated by the following formula: ， in, It is the time it takes for the stone to leave the output end of the elevator and fall to the stone discharge baffle. It is the height from the output end of the elevator to the stone removal baffle. is the gravitational acceleration of the stone.

3. A stone removal device for a harvester, characterized in that: A harvester stone discharge method for implementing the above-mentioned claim 1 or 2 includes: an elevator, a detection mechanism, a stone discharge mechanism, a separation chute and a processor, the detection mechanism and the stone discharge mechanism are both installed at the output end of the elevator, the separation chute is located below the stone discharge mechanism, and the detection mechanism and the stone discharge mechanism are both connected to the processor.

4. A stone removal device for a harvester according to claim 3, characterized in that: The detection mechanism includes: an X-Ray source and an amorphous silicon flat panel detector. The X-Ray source is located above the elevator, and the amorphous silicon flat panel detector is located below the elevator. The X-Ray source and the amorphous silicon flat panel detector are arranged opposite to each other.

5. The stone removal device for a harvester according to claim 3, characterized in that: The processor is provided with a binary threshold segmentation algorithm, and the processor is a 2 GHz CPU processor.

6. The stone removal device for a harvester according to claim 3, characterized in that: The stone-removing mechanism includes: multiple pushing components, multiple stone-removing baffles and a rotating shaft. The tops of the multiple stone-removing baffles are rotatably mounted on the rotating shaft. The multiple pushing components are respectively connected to the multiple stone-removing baffles, and the multiple pushing components are all connected to the processor.

7. The stone removal device for a harvester according to claim 6, characterized in that: The pushing component is a pneumatic cylinder, an electric cylinder or a hydraulic cylinder.

8. The stone removal device for a harvester according to claim 3, characterized in that: The separation chute is an inverted V-shaped structure, one side of the bottom of the separation chute is connected to a crusher box, and the other side of the bottom of the separation chute is connected to a stone collecting trough; the elevator is a scraper elevator, and the scraper elevator is provided with an elevator conveyor belt.

9. A harvester, characterized in that: A stone removal device for a harvester comprising any one of claims 3 to 8.

Images