Intelligent detection system and method for impurity content of mechanically harvested sugarcane

Abstract

The application discloses a kind of intelligent detection system and method for the impurity rate of mechanized recovery of sugarcane, which comprises a weighing mechanism, a main conveying mechanism, a scraping mechanism, a visual recognition mechanism for obtaining the image of the material and a data processing unit for processing the obtained data;The weighing mechanism comprises a weighing and feeding frame, a weighing and feeding bucket and a weighing sensor, the weighing sensor is arranged on the weighing and feeding frame, and the weighing and feeding bucket is arranged on the weighing sensor, the bottom of the weighing and feeding bucket is provided with a switch structure;The first end of the main conveying mechanism is located below the weighing and feeding bucket;The scraping mechanism and the visual recognition mechanism are both arranged above the main conveying mechanism, and the scraping mechanism is located between the weighing mechanism and the visual recognition mechanism;The data processing unit is electrically connected with the weighing sensor and the visual recognition mechanism.The present application has the advantages of simple operation, high efficiency, low error rate and high detection accuracy.

Description

Technical Field

[0001] This invention relates to agricultural and sideline product processing technology, specifically to an intelligent detection system and method for the impurity content of mechanized sugarcane harvesting. Background Technology

[0002] Sugarcane is a perennial grass plant whose stalks are the primary resource. It is mainly cultivated in tropical and subtropical regions and is a vital economic pillar industry in many countries and regions. Sugarcane sugar production involves using sugarcane as raw material and processing it into products such as granulated sugar and raw sugar through steps including juice extraction, purification, evaporation, crystallization, separation, and drying. The characteristics of sugarcane and the chemical composition and properties of its juice have a significant impact on the sugar-making process and serve as the basis for selecting production methods and technological conditions. Before the formal sugar-making process begins, the impurity content of the sugarcane segments must be calculated to determine whether it meets quality standards.

[0003] Traditional methods for detecting impurity content in sugarcane segments mainly involve manual sampling, manually removing impurities, and weighing the samples to determine the impurity content. While this method can measure the impurity content, relying solely on human judgment leads to a high false positive rate, and the test results lack reliability. Furthermore, manual testing is slow and inefficient.

[0004] To address this, existing technologies have developed several impurity detection techniques based on visual recognition. For example, Chinese invention application CN117368195A discloses a graded intelligent detection device and method for impurity content in machine-harvested sugarcane. This detection device identifies impurities and sugarcane through a visual detection mechanism and transmits the images to an image processing module. The image processing module processes the images to obtain the volume of various materials. Combined with the density of various materials, the weight of each substance can be calculated, and then the impurity content can be calculated.

[0005] While the aforementioned detection device can automatically detect the impurity content of sugarcane segments and has advantages such as high efficiency and high detection accuracy, it also has the following problems:

[0006] The visual inspection mechanism identifies all materials (including sugarcane segments and various impurities) on the conveyor belt. The types and quantities of objects to be identified are numerous and vary in shape. This not only requires a large amount of data calculation, which is time-consuming, labor-intensive, and inefficient, but is also prone to errors and has poor detection accuracy, and needs to be improved. Summary of the Invention

[0007] The purpose of this invention is to overcome the above-mentioned problems and provide an intelligent detection system for impurity content in mechanized sugarcane harvesting. This intelligent detection system has the advantages of simple operation, high efficiency, low error rate, and high detection accuracy.

[0008] Another objective of this invention is to overcome the aforementioned problems and provide an intelligent method for detecting the impurity content of mechanized sugarcane harvesting.

[0009] The objective of this invention is achieved through the following technical solution:

[0010] A smart detection system for impurity content of mechanized sugarcane harvesting includes a weighing and feeding mechanism for weighing and feeding the harvested material, a main conveying mechanism for conveying the fed material, a leveling mechanism for leveling up excessively stacked material, a visual recognition mechanism for acquiring images of the material conveyed by the main conveying mechanism, and a data processing unit for processing the acquired data.

[0011] The weighing and dispensing mechanism includes a weighing and dispensing frame, a weighing and dispensing hopper, and a weighing sensor. The weighing sensor is mounted on the weighing and dispensing frame, and the weighing and dispensing hopper is mounted on the weighing sensor. A switch structure is provided at the bottom of the weighing and dispensing hopper.

[0012] The first end of the main conveying mechanism is located below the weighing and dispensing hopper;

[0013] Both the leveling mechanism and the visual recognition mechanism are located above the main conveying mechanism, with the leveling mechanism situated between the weighing and dispensing mechanism and the visual recognition mechanism.

[0014] The data processing unit is electrically connected to the weighing sensor and the visual recognition mechanism.

[0015] In a preferred embodiment of the present invention, a sugarcane leaf removal mechanism is further included for removing sugarcane leaves covering the sugarcane segment to be inspected. The sugarcane leaf removal mechanism includes a leaf-picking mechanism for pushing the sugarcane leaves and a secondary conveying mechanism for conveying the sugarcane leaves.

[0016] The leaf-picking mechanism is located between the leveling mechanism and the visual recognition mechanism. The leaf-picking mechanism includes a first leaf-picking mechanism and a second leaf-picking mechanism, which are arranged along the conveying direction of the main conveying mechanism.

[0017] Furthermore, the first leaf-picking mechanism includes a first leaf-picking frame, a first leaf-picking synchronization component, and a first leaf-picking drive mechanism. The first leaf-picking synchronization component is disposed on the first leaf-picking frame and includes a first leaf-picking synchronization belt and a first leaf-picking synchronization pulley.

[0018] The second leaf-picking mechanism includes a second leaf-picking frame, a second leaf-picking synchronization component, and a second leaf-picking drive mechanism. The second leaf-picking synchronization component is mounted on the second leaf-picking frame and includes a second leaf-picking synchronization belt and a second leaf-picking synchronization pulley.

[0019] In the horizontal projection perpendicular to the conveying direction of the main conveying mechanism, both the first leaf-picking synchronous belt and the second leaf-picking synchronous belt are triangular structures. Each triangular structure includes a leaf-picking edge parallel to the conveying plane of the main conveying mechanism located below. The two leaf-picking edges form a gathering space for gathering sugarcane leaves. The triangular structure corresponding to the second leaf-picking synchronous belt includes an inclined lifting edge for lifting sugarcane leaves. The top of the inclined lifting edge is connected to the first end of the secondary conveying mechanism, and the bottom of the inclined lifting edge extends into the gathering space.

[0020] Both the first and second leaf-pulling synchronous belts are equipped with elastically deformable paddles, and the first and second leaf-pulling synchronous belts move in opposite directions; in a top-view projection, the first and second leaf-pulling synchronous belts are not on the same straight line.

[0021] In operation, the first and second leaf-picking synchronous belts simultaneously and relative to each other push the sugarcane leaves into the gathering space and simultaneously lift the sugarcane leaves upwards.

[0022] In practice, since the harvested sugarcane segments contain a large number of sugarcane leaves, these leaves may cover the segments when they are transferred to the main conveyor mechanism. This can affect subsequent image recognition and lead to inaccurate detection results. To address this, the above structure, when the sugarcane segments are conveyed by the main conveyor mechanism, drives the first and second leaf-picking synchronous belts to move in opposite directions. This causes the leaf-picking blades on the first and second synchronous belts to move synchronously and in opposite directions above the sugarcane segments, simultaneously guiding the sugarcane leaves above the segments into the gathering space. (Since the leaf-picking blades are made of elastic material, they can adapt to the size of the sugarcane segments, allowing them to adhere precisely to the surface of the segments and guide the leaves without scratching them.) The leaves are then simultaneously lifted upwards, and the second leaf-picking synchronous belt transports them to the secondary conveyor mechanism, which then bypasses the visual recognition mechanism and returns the leaves to the main conveyor mechanism. In this way, the sugarcane leaves covering the sugarcane segments are actively removed, making the sugarcane segments clearly visible, which facilitates subsequent image recognition, greatly improves recognition accuracy, and thus improves the accuracy of the detection results.

[0023] Furthermore, the first leaf-pulling mechanism also includes a discharge stop bar, which is located directly above the retraction space; the leaf-pulling synchronous belt of the first leaf-pulling mechanism is provided with a downward discharge slot.

[0024] The paddles on the second leaf-pulling synchronous belt include synchronous paddles and relay paddles. In the forward direction, the relay paddles are located behind the synchronous paddles. The synchronous paddles on the second leaf-pulling synchronous belt correspond one-to-one with the paddles on the first leaf-pulling synchronous belt.

[0025] With the above structure, the sugarcane leaves are first lifted upwards synchronously in the gathering space by the paddles on the first and second leaf-pulling synchronous belts. When the paddle on the first leaf-pulling synchronous belt approaches the unloading stop, the unloading stop will cause the paddle on the first leaf-pulling synchronous belt to bend downwards around the downward unloading chute, thereby unloading the sugarcane leaves. At this time, the relay paddle on the second leaf-pulling synchronous belt is located in the gathering space, so it can receive the sugarcane leaves unloaded from the first leaf-pulling synchronous belt, that is, transfer the sugarcane leaves from the first leaf-pulling synchronous belt to the second leaf-pulling synchronous belt. The second leaf-pulling synchronous belt is specifically used to transfer the sugarcane leaves to the secondary conveying mechanism, which is more efficient.

[0026] Furthermore, the first leaf-picking drive mechanism includes a first leaf-picking drive motor and a power shaft. The first leaf-picking synchronization component is mounted on the power shaft, and one end of the power shaft is connected to the first leaf-picking drive motor. With this structure, under the drive of the first leaf-picking drive motor, the first leaf-picking synchronization belt can circulate around the triangle, thereby feeding the sugarcane leaves.

[0027] Furthermore, the second leaf-shifting drive mechanism includes two synchronous gears, one of which is connected to the power shaft, and the other is coaxially connected to the second leaf-shifting synchronization assembly. This allows the first leaf-shifting drive motor to share power, thus enabling the second leaf-shifting synchronous belt to drive the operation.

[0028] Furthermore, the secondary conveying mechanism is located above the main conveying mechanism. The secondary conveying mechanism includes a secondary conveying frame, a secondary conveying synchronization component, and a secondary conveying drive mechanism. The secondary conveying synchronization component includes a secondary conveying timing belt and a secondary conveying timing pulley.

[0029] The secondary conveying synchronization assembly consists of multiple sets, used to convey sugarcane leaves around the visual recognition mechanism and then back to the main conveying mechanism. These multiple sets of secondary conveying synchronization assemblies are connected end-to-end. One end of the foremost secondary conveying synchronization belt is coaxially connected to the second leaf-picking synchronization belt. Through this structure, the power of the first leaf-picking drive motor can be shared to drive the secondary conveying synchronization belts.

[0030] In a preferred embodiment of the present invention, the switching structure includes a switching gate and a switching cylinder, the switching cylinder being fixedly mounted on the weighing and dispensing hopper, and the telescopic rod of the switching cylinder being connected to the switching gate. With this structure, driven by the switching cylinder, the switching gate can automatically move closer to or further away from the bottom outlet of the weighing and dispensing hopper.

[0031] In a preferred embodiment of the present invention, the main conveying mechanism includes a main conveyor frame, a main conveyor belt, and a main conveying drive mechanism;

[0032] The main conveying drive mechanism includes a main conveying drive motor and two main conveying rollers rotatably connected to the main conveyor frame. The main conveying drive motor is connected to one of the main conveying rollers. The main conveyor belt is connected between the two main conveying rollers. With this structure, driven by the main conveying drive motor, the main conveyor belt can move in a cyclical manner, thereby conveying sugarcane segments and impurities.

[0033] In a preferred embodiment of the present invention, the leveling mechanism includes a leveling frame, a leveling mounting plate, and a leveling roller. The leveling mounting plate is disposed on the leveling frame, and the leveling roller is rotatably connected to the bottom of the leveling mounting plate. The leveling roller and the conveyor belt of the main conveying mechanism below form a limiting channel for the passage of sugarcane segments.

[0034] Furthermore, a vertical adjustment structure is provided between the leveling mounting plate and the leveling frame. This vertical adjustment structure includes a vertical adjustment rod, a guide rod, and a guide hole. The vertical adjustment rod is vertically positioned and connected to the leveling frame via a threaded structure. The guide hole is located on the side of the leveling frame. One end of the guide rod is fixed to the leveling mounting plate, and the other end extends laterally into the guide hole. This structure allows for adjustment of the leveling roller's height to accommodate sugarcane segments of different sizes.

[0035] In a preferred embodiment of the present invention, at least two sets of the leveling mechanism are provided and arranged along the conveying direction of the main conveying mechanism;

[0036] Along the conveying direction, the height of the rear limiting channel is lower than the height of the front limiting channel. This allows for the gradual leveling of excessively stacked sugarcane segments, improving sugarcane throughput and preventing blockages.

[0037] In a preferred embodiment of the present invention, the visual recognition mechanism includes a detection box and a camera, wherein the camera is disposed inside the detection box and above the main conveying mechanism.

[0038] A method for intelligent detection of impurity content in mechanized sugarcane harvesting includes the following steps:

[0039] The material to be tested is placed in the weighing hopper, and the weighing sensor weighs the material in the weighing hopper to obtain the total gross weight of all materials in this batch. The total gross weight is then transmitted to the data processing unit.

[0040] The material in the weighing and feeding hopper is lowered onto the main conveyor mechanism below, which then transports the material downwards.

[0041] When sugarcane segments pass under the leveling roller, the leveling mechanism scrapes down sugarcane segments that are stacked too high, spreading them out flat and improving the accuracy and precision of visual inspection.

[0042] When the sugarcane segment is conveyed down to the area below the vision recognition mechanism, the vision recognition mechanism acquires an image of the material to be detected on the main conveyor and transmits the image to the data processing unit for image processing.

[0043] The data processing unit specifically identifies sugarcane segments, calculates the volume of each segment, sums them to obtain the total volume of all sugarcane segments, combines the density of the sugarcane segments to obtain the total net weight of all sugarcane segments, subtracts the total net weight of the sugarcane segments from the total gross weight to obtain the total net weight of impurities, and finally divides the total net weight of impurities by the total gross weight to obtain the impurity content.

[0044] In a preferred embodiment of the present invention, the operation of identifying sugarcane segments and calculating the volume of each sugarcane segment by the data processing unit is as follows:

[0045] (1) Image preprocessing: Filter the received image (such as Gaussian filtering or median filtering) to suppress noise; perform illumination compensation to reduce the impact of uneven illumination; convert the image from RGB color space to HSV or HSL color space;

[0046] (2) Target region segmentation: Based on the typical color features of sugarcane segments (defined in RGB or HSV / HSL space), the image is segmented by thresholding or a pre-trained color classifier is used to generate a preliminary binary mask image, in which the region suspected to be a sugarcane segment is marked as the foreground (white) and the remaining region is the background (black).

[0047] Morphological opening operations (erosion followed by dilation) are performed on the initial binary mask image to remove small noise points and break small non-cane connections;

[0048] A morphological closing operation (dilation followed by erosion) is performed on the preliminary binary mask image to fill the small cavities inside the sugarcane segment and connect adjacent fracture areas belonging to the same sugarcane segment.

[0049] On the optimized binary mask image, a contour search algorithm is used to extract the outer boundary contours of all connected regions (candidate objects).

[0050] (3) Sugarcane segment discrimination: For each extracted candidate object contour: calculate its minimum bounding rectangle or fitted ellipse;

[0051] Calculate the key geometric features of the candidate object: aspect ratio (length / width) and outline area;

[0052] Based on preset rules (e.g., aspect ratio greater than a set threshold T1, contour area within a set range [A_min, A_max]), determine whether the candidate object is a sugarcane segment;

[0053] (4) Calculation of the three-dimensional dimensions and volume of sugarcane segments:

[0054] For each outline identified as a sugarcane segment: using the region corresponding to the outline in the depth map, combined with camera intrinsic parameters, calculate and generate the 3D point cloud of the sugarcane segment;

[0055] Perform principal axis analysis (such as PCA) on the 3D point cloud to determine its principal direction (length direction);

[0056] The extension length of the point cloud along the main direction is calculated to obtain the true physical length of the sugarcane segment;

[0057] On multiple cross sections perpendicular to the main direction, the average diameter of the point cloud distribution is calculated to obtain the average physical diameter of the sugarcane segment.

[0058] Treat the sugarcane section as an approximation of a cylinder and calculate its volume.

[0059] Compared with the prior art, the present invention has the following advantages:

[0060] 1. The intelligent impurity detection system of the present invention first obtains the total gross weight of all collected materials, then identifies only sugarcane segments after acquiring the image and calculates their volume, and then calculates the mass of the sugarcane segments by combining the density. Subsequently, the weight of impurities and the impurity content can be obtained. The detection steps are very simple and fast, which is conducive to improving detection efficiency.

[0061] 2. Since only sugarcane segments are identified, and the top-view projection of sugarcane segments is rectangular, the features are relatively obvious and easy to identify, which can reduce the error rate and improve the detection accuracy. Attached Figure Description

[0062] Figures 1-2 These are three-dimensional structural diagrams from two different perspectives of the intelligent detection system for impurity content in mechanized sugarcane harvesting according to the present invention.

[0063] Figure 3 This is a three-dimensional structural diagram of the leaf-removing mechanism of the sugarcane leaf removal mechanism of the present invention.

[0064] Figures 4-6 This is a front view of the leaf-picking mechanism of the present invention under different working states. Detailed Implementation

[0065] To enable those skilled in the art to fully understand the technical solutions of the present invention, the present invention will be further described below in conjunction with embodiments and accompanying drawings, but the embodiments of the present invention are not limited thereto.

[0066] Example 1

[0067] The intelligent detection system for impurity content of mechanized sugarcane harvesting in this embodiment includes a weighing and feeding mechanism for weighing and feeding the harvested material, a main conveying mechanism for conveying the fed material, a leveling mechanism for leveling up excessively stacked material, a visual recognition mechanism for acquiring images of the material conveyed by the main conveying mechanism, and a data processing unit for processing the acquired data.

[0068] Combination Figures 1-2 The weighing and dispensing mechanism includes a weighing and dispensing frame 1, a weighing and dispensing hopper 2, and a weighing sensor 3. The weighing sensor 3 is mounted on the weighing and dispensing frame 1, and the weighing and dispensing hopper 2 is mounted on the weighing sensor 3. The bottom of the weighing and dispensing hopper 2 is equipped with a switch structure. The head of the main conveying mechanism is located below the weighing and dispensing hopper 2. The leveling mechanism and the visual recognition mechanism are both located above the main conveying mechanism, with the leveling mechanism located between the weighing and dispensing mechanism and the visual recognition mechanism. The data processing unit is electrically connected to the weighing sensor 3 and the visual recognition mechanism.

[0069] Combination Figure 2 The switching structure includes a switching gate 4 and a switching cylinder 5. The switching cylinder 5 is fixedly mounted on the weighing and dispensing hopper 2, and its extension rod is connected to the switching gate 4. Through this structure, driven by the switching cylinder 5, the switching gate 4 can automatically move closer to or further away from the bottom outlet of the weighing and dispensing hopper 2.

[0070] Combination Figures 1-2 The main conveying mechanism includes a main conveyor frame 6, a main conveyor belt 7, and a main conveying drive mechanism. The main conveying drive mechanism includes a main conveying drive motor 8 and two main conveying rollers rotatably connected to the main conveyor frame 6. The main conveying drive motor 8 is connected to one of the main conveying rollers. The main conveyor belt 7 is connected between the two main conveying rollers. With this structure, driven by the main conveying drive motor 8, the main conveyor belt 7 can move in a cyclical manner, thereby conveying sugarcane segments and impurities.

[0071] Combination Figures 1-2 The leveling mechanism includes a leveling frame 9, a leveling mounting plate 10, and a leveling roller 11. The leveling mounting plate 10 is mounted on the leveling frame 9, and the leveling roller 11 is rotatably connected to the bottom of the leveling mounting plate 10. The leveling roller 11 and the main conveyor belt 7 of the main conveying mechanism below form a limiting channel for the passage of sugarcane segments.

[0072] Furthermore, the leveling mechanism has at least two sets arranged along the conveying direction of the main conveying mechanism; along the conveying direction, the height of the rear limiting channel is less than the height of the front limiting channel. In this way, excessively stacked sugarcane segments can be leveled step by step, improving the sugarcane's throughput and preventing blockages.

[0073] Combination Figures 1-2 The visual recognition mechanism includes a detection box 12 and a camera, the camera being disposed inside the detection box 12 and located above the main conveying mechanism.

[0074] This embodiment also includes a sugarcane leaf removal mechanism for removing sugarcane leaves covering the sugarcane segment to be inspected. The sugarcane leaf removal mechanism includes a leaf-picking mechanism for picking the sugarcane leaves and a secondary conveying mechanism for conveying the sugarcane leaves. The leaf-picking mechanism is located between the leveling mechanism and the visual recognition mechanism. The leaf-picking mechanism includes a first leaf-picking mechanism and a second leaf-picking mechanism, which are arranged along the conveying direction of the main conveying mechanism.

[0075] Combination Figures 1-3 The first leaf-picking mechanism includes a first leaf-picking frame 13, a first leaf-picking synchronization component, and a first leaf-picking drive mechanism. The first leaf-picking synchronization component is mounted on the first leaf-picking frame 13 and includes a first leaf-picking synchronization belt 14 and a first leaf-picking synchronization pulley. The second leaf-picking mechanism includes a second leaf-picking frame 16, a second leaf-picking synchronization component, and a second leaf-picking drive mechanism. The second leaf-picking synchronization component is mounted on the second leaf-picking frame 16 and includes a second leaf-picking synchronization belt 17 and a second leaf-picking synchronization pulley. In a horizontal projection perpendicular to the conveying direction of the main conveying mechanism, both the first leaf-picking synchronization belt 14 and the second leaf-picking synchronization belt 17 have a triangular structure. Each of these triangular structures includes a leaf-picking edge parallel to the conveying plane of the main conveying mechanism located below. The two leaf-picking edges form a gathering space 19 for gathering sugarcane leaves; the triangular structure corresponding to the second leaf-picking timing belt 17 includes an inclined lifting edge for lifting the sugarcane leaves, the top of which is connected to the first end of the secondary conveying mechanism, and the bottom of which extends into the gathering space 19; both the first leaf-picking timing belt 14 and the second leaf-picking timing belt 17 are provided with elastically deformable paddles 20, and the first leaf-picking timing belt 14 and the second leaf-picking timing belt 17 move in opposite directions; in a top-view projection, the first leaf-picking timing belt 14 and the second leaf-picking timing belt 17 are not on the same straight line; in the working state, the paddles 20 of the first leaf-picking timing belt 14 and the second leaf-picking timing belt 17 synchronously and relatively push the sugarcane leaves into the gathering space 19, and synchronously lift the sugarcane leaves upward.

[0076] In practice, since the harvested sugarcane segments contain a large number of sugarcane leaves, when the sugarcane segments are transferred to the main conveyor mechanism, sugarcane leaves may cover them, which will affect subsequent image recognition and lead to inaccurate detection results. Therefore, the above structure, through which the sugarcane segments are conveyed on the main conveyor mechanism, drives the first leaf-picking synchronous belt 14 and the second leaf-picking synchronous belt 17 to move in opposite directions via the first leaf-picking driving mechanism and the second leaf-picking synchronous belt 17. This causes the picks 20 on the first leaf-picking synchronous belt 14 and the second leaf-picking synchronous belt 17 to move synchronously in opposite directions above the sugarcane segments, simultaneously pushing the sugarcane leaves above the sugarcane segments into the gathering space 19 (since the picks 20 are made of elastically deformable material, they can adaptively deform according to the size of the sugarcane segments, thus fitting snugly against the surface of the sugarcane segments to push the sugarcane leaves without scratching them), and then synchronously lifting the sugarcane leaves upwards. Figure 4-5 Then, the second leaf-picking synchronous belt 17 transports the sugarcane leaves to the secondary conveying mechanism, which then bypasses the visual recognition mechanism and transports the sugarcane leaves back to the main conveying mechanism. In this way, the sugarcane leaves covering the sugarcane segments are actively removed, making the sugarcane segments clearly visible, which facilitates subsequent image recognition, greatly improves recognition accuracy, and thus improves the accuracy of the detection results.

[0077] Combination Figures 3-6 The first leaf-pulling mechanism also includes a discharge stop bar 21, which is located directly above the retracting space 19; the leaf-pulling timing belt 14 has a downward discharge slot on its leaf-pulling blade 20; the leaf-pulling timing belt 17 has a timing blade 201 and a relay blade 202 on its blade 20, and in the forward direction, the relay blade 202 is located behind the timing blade 201. The timing blade 201 on the second leaf-pulling timing belt 17 corresponds one-to-one with the blade 20 on the first leaf-pulling timing belt 14.

[0078] With the above structure, the sugarcane leaves are first lifted upwards synchronously in the gathering space 19 by the paddle 20 on the first leaf-picking synchronous belt 14 and the synchronous paddle 201 on the second leaf-picking synchronous belt 17. When the paddle 20 on the first leaf-picking synchronous belt 14 approaches the unloading stop bar 21, the unloading stop bar 21 will cause the paddle 20 on the first leaf-picking synchronous belt 14 to bend downwards around the downward unloading chute, thereby unloading the sugarcane leaves. At this time, the relay paddle 202 on the second leaf-picking synchronous belt 17 is located in the gathering space 19, so it can receive the sugarcane leaves unloaded from the first leaf-picking synchronous belt 14, that is, transfer the sugarcane leaves on the first leaf-picking synchronous belt 14 to the second leaf-picking synchronous belt 17. Figure 6 The second leaf-picking synchronous belt 17 is specifically used to transfer sugarcane leaves to the secondary conveying mechanism, which is more efficient.

[0079] Furthermore, the first leaf-picking drive mechanism includes a first leaf-picking drive motor 22 and a power shaft. The first leaf-picking synchronization component is mounted on the power shaft, and one end of the power shaft is connected to the first leaf-picking drive motor 22. With the above structure, under the drive of the first leaf-picking drive motor 22, the first leaf-picking synchronization belt 14 can move cyclically around the triangle, thereby feeding the sugarcane leaves.

[0080] Furthermore, the second leaf-shifting drive mechanism includes two synchronous gears 23, one of which is connected to the power shaft, and the other is coaxially connected to the second leaf-shifting synchronization assembly. This allows the first leaf-shifting drive motor 22 to share power, thus enabling the second leaf-shifting synchronization belt 17 to be driven.

[0081] Combination Figures 1-2 The secondary conveying mechanism is located above the main conveying mechanism. This secondary conveying mechanism includes a secondary conveyor frame 24, a secondary conveying synchronization assembly, and a secondary conveying drive mechanism. The secondary conveying synchronization assembly includes a secondary conveying synchronization belt 25 and a secondary conveying synchronization belt 25 wheel. Multiple sets of the secondary conveying synchronization assembly are provided to convey sugarcane leaves around the visual recognition mechanism and then back to the main conveying mechanism. These multiple sets of secondary conveying synchronization assemblies are connected end-to-end. One end of the foremost secondary conveying synchronization belt 25 is coaxially connected to the second leaf-picking synchronization belt 17. Through this structure, the power of the first leaf-picking drive motor 22 can be shared to drive the secondary conveying synchronization belt 25.

[0082] Example 2

[0083] Combination Figures 1-2 The intelligent detection method for impurity content in mechanized sugarcane harvesting according to this embodiment includes the following steps:

[0084] The material to be tested is placed into the weighing hopper 2, and the weighing sensor 3 weighs the material in the weighing hopper 2 to obtain the total gross weight of all materials in this batch, and then transmits the total gross weight to the data processing unit.

[0085] The material in the weighing and feeding hopper 2 is lowered onto the main conveyor mechanism below, which then transports the material downwards.

[0086] When sugarcane segments pass under the leveling roller 11, the leveling mechanism scrapes down the excessively stacked sugarcane segments, spreading them out flat and improving the accuracy and precision of visual inspection.

[0087] When the sugarcane segments are conveyed down to the area below the vision recognition mechanism, the vision recognition mechanism acquires an image of the material to be detected on the main conveyor and transmits the image to the data processing unit for image processing.

[0088] The data processing unit specifically identifies sugarcane segments, calculates the volume of each segment, sums them to obtain the total volume of all sugarcane segments, combines the density of the sugarcane segments to obtain the total net weight of all sugarcane segments, subtracts the total net weight of the sugarcane segments from the total gross weight to obtain the total net weight of impurities, and finally divides the total net weight of impurities by the total gross weight to obtain the impurity content.

[0089] Furthermore, the operation of identifying sugarcane segments and calculating the volume of each segment using the data processing unit is as follows:

[0090] (1) Image preprocessing: Filter the received image (such as Gaussian filtering or median filtering) to suppress noise; perform illumination compensation to reduce the impact of uneven illumination; convert the image from RGB color space to HSV or HSL color space.

[0091] (2) Target region segmentation: Based on the typical color features of sugarcane segments (defined in RGB or HSV / HSL space), threshold segmentation is performed on the image or a pre-trained color classifier is used to generate a preliminary binary mask image, in which the region suspected to be a sugarcane segment is marked as the foreground (white) and the remaining region is the background (black).

[0092] A morphological opening operation (erosion followed by dilation) is performed on the initial binary mask image to remove small noise points and break fine non-cane connections.

[0093] A morphological closing operation (dilation followed by erosion) is performed on the initial binary mask image to fill the small cavities inside the sugarcane segment and connect adjacent fracture areas belonging to the same sugarcane segment.

[0094] On the optimized binary mask image, a contour search algorithm is used to extract the outer boundary contours of all connected regions (candidate objects).

[0095] (3) Sugarcane segment discrimination: For each extracted candidate object contour: calculate its minimum bounding rectangle or fitted ellipse.

[0096] Calculate the key geometric features of the candidate object: aspect ratio (length / width) and contour area.

[0097] Based on preset rules (e.g., aspect ratio greater than a set threshold T1, contour area within a set range [A_min, A_max]), determine whether the candidate object is a sugarcane segment.

[0098] (4) Calculation of the three-dimensional dimensions and volume of sugarcane segments:

[0099] For each outline identified as a sugarcane segment: using the corresponding region in the depth map and combining the camera intrinsic parameters, the 3D point cloud of the sugarcane segment is calculated and generated.

[0100] Perform principal axis analysis (such as PCA) on the 3D point cloud to determine its principal direction (length direction).

[0101] The extension length of the point cloud along the main direction is calculated to obtain the true physical length of the sugarcane segment.

[0102] The average diameter of the point cloud distribution is calculated on multiple cross sections perpendicular to the main direction to obtain the average physical diameter of the sugarcane segment.

[0103] Treat the sugarcane section as an approximation of a cylinder and calculate its volume.

[0104] The above are preferred embodiments of the present invention, but the embodiments of the present invention are not limited to the above content. Any changes, modifications, substitutions, combinations, or simplifications made without departing from the spirit and principle of the present invention shall be considered equivalent substitutions and shall be included within the protection scope of the present invention.

Claims

1. An intelligent detection system for the impurity content of mechanized sugarcane harvesting, characterized in that, It includes a weighing and dispensing mechanism for weighing and dispensing the harvested materials, a main conveying mechanism for conveying the dispensed materials, a leveling mechanism for leveling up materials that are stacked too high, a visual recognition mechanism for acquiring images of the materials conveyed by the main conveying mechanism, and a data processing unit for processing the acquired data. The weighing and dispensing mechanism includes a weighing and dispensing frame, a weighing and dispensing hopper, and a weighing sensor. The weighing sensor is mounted on the weighing and dispensing frame, and the weighing and dispensing hopper is mounted on the weighing sensor. A switch structure is provided at the bottom of the weighing and dispensing hopper. The first end of the main conveying mechanism is located below the weighing and dispensing hopper; Both the leveling mechanism and the visual recognition mechanism are located above the main conveying mechanism, with the leveling mechanism situated between the weighing and dispensing mechanism and the visual recognition mechanism. The data processing unit is electrically connected to the weighing sensor and the vision recognition mechanism. It also includes a sugarcane leaf removal mechanism for removing sugarcane leaves covering the sugarcane segment to be inspected, the sugarcane leaf removal mechanism including a leaf-picking mechanism for picking sugarcane leaves and a secondary conveying mechanism for conveying sugarcane leaves. The leaf-picking mechanism is located between the leveling mechanism and the visual recognition mechanism. The leaf-picking mechanism includes a first leaf-picking mechanism and a second leaf-picking mechanism, which are arranged along the conveying direction of the main conveying mechanism. The first leaf-picking mechanism includes a first leaf-picking frame, a first leaf-picking synchronization component, and a first leaf-picking drive mechanism. The first leaf-picking synchronization component is disposed on the first leaf-picking frame and includes a first leaf-picking synchronization belt and a first leaf-picking synchronization pulley. The second leaf-picking mechanism includes a second leaf-picking frame, a second leaf-picking synchronization component, and a second leaf-picking drive mechanism. The second leaf-picking synchronization component is mounted on the second leaf-picking frame and includes a second leaf-picking synchronization belt and a second leaf-picking synchronization pulley. In the horizontal projection perpendicular to the conveying direction of the main conveying mechanism, both the first leaf-picking synchronous belt and the second leaf-picking synchronous belt are triangular structures. Each triangular structure includes a leaf-picking edge parallel to the conveying plane of the main conveying mechanism located below. The two leaf-picking edges form a gathering space for gathering sugarcane leaves. The triangular structure corresponding to the second leaf-picking synchronous belt includes an inclined lifting edge for lifting sugarcane leaves. The top of the inclined lifting edge is connected to the first end of the secondary conveying mechanism, and the bottom of the inclined lifting edge extends into the gathering space. Both the first and second leaf-pulling synchronous belts are equipped with elastically deformable paddles, and the first and second leaf-pulling synchronous belts move in opposite directions; in a top-view projection, the first and second leaf-pulling synchronous belts are not on the same straight line. In operation, the first and second leaf-picking synchronous belts simultaneously and relative to each other push the sugarcane leaves into the gathering space and simultaneously lift the sugarcane leaves upwards.

2. The intelligent detection system for impurity content of mechanized sugarcane harvesting according to claim 1, characterized in that, The first leaf-pulling mechanism also includes a discharge stop bar, which is located directly above the retraction space; the leaf-pulling timing belt of the first leaf-pulling mechanism is provided with a downward discharge slot. The paddles on the second leaf-pulling synchronous belt include synchronous paddles and relay paddles. In the forward direction, the relay paddles are located behind the synchronous paddles. The synchronous paddles on the second leaf-pulling synchronous belt correspond one-to-one with the paddles on the first leaf-pulling synchronous belt.

3. The intelligent detection system for impurity content of mechanized sugarcane harvesting according to claim 1, characterized in that, The first leaf-shifting drive mechanism includes a first leaf-shifting drive motor and a power shaft. The first leaf-shifting synchronization component is disposed on the power shaft, and one end of the power shaft is connected to the first leaf-shifting drive motor.

4. The intelligent detection system for impurity content of mechanized sugarcane harvesting according to claim 3, characterized in that, The second leaf-shifting drive mechanism includes two synchronizing gears, one of which is connected to the power shaft, and the other is coaxially connected to the second leaf-shifting synchronization assembly.

5. The intelligent detection system for impurity content of mechanized sugarcane harvesting according to claim 4, characterized in that, The secondary conveying mechanism is located above the main conveying mechanism. The secondary conveying mechanism includes a secondary conveying frame, a secondary conveying synchronization component, and a secondary conveying drive mechanism. The secondary conveying synchronization component includes a secondary conveying timing belt and a secondary conveying timing pulley. The secondary conveying synchronization component is provided in multiple sets, which are used to convey sugarcane leaves around the visual recognition mechanism and then back to the main conveying mechanism. The multiple sets of secondary conveying synchronization components are connected end to end; one end of the secondary conveying synchronization belt located at the front is coaxially connected to the second leaf-picking synchronization belt.

6. The intelligent detection system for impurity content of mechanized sugarcane harvesting according to claim 1, characterized in that, The switch structure includes a switch gate and a switch cylinder. The switch cylinder is fixedly mounted on the weighing and feeding hopper, and the telescopic rod of the switch cylinder is connected to the switch gate.

7. A method for intelligently detecting impurity content in a mechanized sugarcane harvesting system according to any one of claims 1-6, characterized in that, Includes the following steps: The material to be tested is placed in the weighing hopper, and the weighing sensor weighs the material in the weighing hopper to obtain the total gross weight of all materials in this batch. The total gross weight is then transmitted to the data processing unit. The material in the weighing and feeding hopper is lowered onto the main conveyor mechanism below, which then transports the material downwards. When sugarcane segments pass under the leveling roller, the leveling mechanism scrapes down sugarcane segments that are stacked too high, spreading them out flat and improving the accuracy and precision of visual inspection. When the sugarcane segment is conveyed down to the area below the vision recognition mechanism, the vision recognition mechanism acquires an image of the material to be detected on the main conveyor and transmits the image to the data processing unit for image processing. The data processing unit specifically identifies sugarcane segments, calculates the volume of each segment, sums them to obtain the total volume of all sugarcane segments, combines the density of the sugarcane segments to obtain the total net weight of all sugarcane segments, subtracts the total net weight of the sugarcane segments from the total gross weight to obtain the total net weight of impurities, and finally divides the total net weight of impurities by the total gross weight to obtain the impurity content.

8. The intelligent impurity detection method according to claim 7, characterized in that, The operation of identifying sugarcane segments and calculating the volume of each segment using the data processing unit is as follows: (1) Image preprocessing: Filter the received image to suppress noise; perform illumination compensation to reduce the impact of uneven illumination; convert the image from RGB color space to HSV or HSL color space; (2) Target region segmentation: Based on the typical color features of sugarcane segments, threshold segmentation is performed on the image or a pre-trained color classifier is used to generate a preliminary binary mask image, in which the region suspected to be a sugarcane segment is marked as the foreground and the remaining region is the background; Morphological opening operations are performed on the initial binary mask image to remove small noise points and break small non-cane connections; Morphological closing operations are performed on the preliminary binary mask image to fill small cavities inside the sugarcane segments and connect adjacent fracture regions belonging to the same sugarcane segment. On the optimized binary mask image, a contour lookup algorithm is used to extract the outer boundary contours of all connected regions; (3) Sugarcane segment discrimination: For each extracted candidate object contour: calculate its minimum bounding rectangle or fitted ellipse; Calculate the key geometric features of the candidate object: aspect ratio and outline area; Determine whether the candidate is a sugarcane segment based on preset rules; (4) Calculation of the three-dimensional dimensions and volume of sugarcane segments: For each outline identified as a sugarcane segment: using the region corresponding to the outline in the depth map, combined with camera intrinsic parameters, calculate and generate the 3D point cloud of the sugarcane segment; Perform principal axis analysis on the 3D point cloud to determine its principal direction; The extension length of the point cloud along the main direction is calculated to obtain the true physical length of the sugarcane segment; On multiple cross sections perpendicular to the main direction, the average diameter of the point cloud distribution is calculated to obtain the average physical diameter of the sugarcane segment. Treat the sugarcane section as an approximation of a cylinder and calculate its volume.

Images