An automatic soil compactness identification system

Abstract

The application relates to the technical field of soil compaction degree, in particular to an automatic soil compaction degree identification system, which comprises a collection module, a data processing module and a compaction degree prediction module.The collection module is used for collecting soil samples, obtaining soil test sample images under different illumination intensities through the collected soil samples, and obtaining spectral data of the soil samples; the data processing module comprises image preprocessing, particle identification and classification, and water content estimation, the image preprocessing is used for obtaining soil characteristics based on different illumination conditions, and the water content of the soil is estimated through the spectral data of the soil samples; and the compaction degree prediction module generates a compaction degree prediction model through a machine learning algorithm according to the soil characteristics of each soil test sample under different illumination intensities and the water content of the soil test sample, so that the detection accuracy of the soil compaction degree can be improved.

Description

Technical Field

[0001] This application relates to the technical field of soil compaction, and in particular to an automatic soil compaction identification system. Background Technology

[0002] Soil compaction is one of the important indicators for assessing the stability of foundations in civil engineering projects such as roads and buildings. During construction, ensuring that the soil achieves sufficient compaction is crucial for the safety and durability of subsequent structures.

[0003] Currently, in order to solve the problems existing in the traditional manual measurement methods of soil compaction, the industry usually adopts physical methods, such as soil nuclear density meters and ring cutter methods. These methods mainly estimate the soil compaction by directly measuring the soil mass, volume and moisture content.

[0004] However, the existing technical methods mentioned above are easily affected by changes in light conditions during soil compaction testing at construction sites. As a result, traditional physical and chemical methods are difficult to accurately characterize parameters such as soil particle composition, gradation, and moisture content, which leads to inaccurate and unreliable soil compaction measurement results. Summary of the Invention

[0005] To improve the accuracy of soil compaction detection, this application provides an automatic soil compaction identification system.

[0006] This application provides an automatic soil compaction degree identification system, which adopts the following technical solution: An automatic soil compaction degree identification system includes: Acquisition Module: The acquisition module is used to collect soil samples, obtain soil test sample images under different light intensities through the collected soil samples, and obtain spectral data of the soil samples through the collected soil samples; Data processing module: The data processing module includes image preprocessing, particle recognition and classification, and moisture content estimation. Image preprocessing is used to acquire multiple soil features based on different lighting conditions. Particle recognition and classification is used to identify and classify particles of different sizes and shapes in soil test sample images. Soil moisture content is estimated through the spectral data of soil samples. Compaction Prediction Module: Based on the soil characteristics of each soil test sample under different light intensities and the moisture content of the soil test sample, a compaction prediction model is generated through machine learning algorithms.

[0007] By adopting the above technical solution, soil test sample images and spectral data under different light intensities are obtained through the acquisition module. Then, the data of the soil test sample images are processed, and the water content is estimated through the spectral data. The compaction degree prediction model is obtained through data processing, and the compaction degree of the soil samples in the field is predicted through the compaction degree prediction model. The size, distribution characteristics, and shape characteristics of particles in the soil test samples under different light conditions are obtained, thereby improving the accuracy and reliability of soil feature extraction and improving the accuracy of compaction degree prediction. Compared with the existing technology, this application can quickly and accurately assess soil compaction degree.

[0008] Preferably, the acquisition module includes a sample acquisition mechanism for collecting soil samples.

[0009] By adopting the above technical solution, soil samples can be easily collected by the sample collection mechanism, which facilitates the subsequent acquisition of soil sample images and spectral data.

[0010] Preferably, the acquisition module further includes a data acquisition mechanism, which includes an image acquisition component located at the output end of the sample acquisition mechanism. The image acquisition component is used to capture soil sample images.

[0011] By adopting the above technical solution, after the sample collection agency collects soil samples, the image acquisition component can capture images of the collected soil samples.

[0012] Preferably, the data acquisition mechanism further includes a spectral data acquisition component, which is used to acquire spectral data of the soil sample.

[0013] By adopting the above technical solution, after the sample collection agency collects soil samples, the spectral data acquisition component can capture the spectral data of the collected soil samples, and the moisture content can be estimated by the acquired spectral data.

[0014] Preferably, the sample collection mechanism includes a lifting drive component and a sampling component, the sampling component being disposed at the output end of the lifting drive component, and the sampling component being used for collecting soil samples.

[0015] By adopting the above technical solution, the lifting drive component can drive the sampling component to move downward, so that the sampling component can collect soil samples. After the sampling component completes the collection of soil samples, the lifting drive component can drive the sampling component to move upward, which facilitates the subsequent analysis of the collected soil samples.

[0016] Preferably, the image acquisition component includes a light source and a camera, wherein the light source is used to provide illumination for the soil sample, and the camera is used to acquire images of the soil sample.

[0017] By adopting the above technical solution, the light source provides illumination to the soil sample, and after the soil is under specific illumination, the image of the soil sample is extracted by the camera.

[0018] Preferably, the sample collection mechanism includes a collection component for collecting the collected soil samples and transporting the soil samples to the image acquisition component.

[0019] By adopting the above technical solution, after the sampling component completes the collection of soil samples, the collected soil samples are stored in the collection component. After the collection component collects multiple soil samples, the multiple soil samples are transferred to the image acquisition component, and the image acquisition component acquires images of multiple soil samples.

[0020] Preferably, the spectral data acquisition component includes a near-infrared spectrometer, which is used to acquire spectral data from soil samples.

[0021] By adopting the above technical solution, the near-infrared spectrometer can collect spectral data of soil samples, and then the water content can be estimated through the spectral data.

[0022] Preferably, image preprocessing includes noise reduction, image enhancement, and normalization.

[0023] By adopting the above technical solutions, denoising the image training set can reduce the interference of noise on soil feature analysis and improve the quality of soil test sample images; enhancing the features in the image training set can make the subsequent identification of particle features in the soil more accurate; and converting the soil test sample images of the image training set data into a uniform format or range can minimize the scale differences between different images.

[0024] Preferably, the compaction degree prediction module includes model training and validation, which includes acquiring soil images, soil and corresponding soil compaction degrees collected on site, establishing a validation set, and validating the validation set and the corresponding on-site soil through the compaction degree prediction model.

[0025] By employing the above technical solution, a compaction degree prediction model is used to predict the compaction degree of soil images and soil samples from the field. This includes predicting and analyzing the soil characteristics of several test areas under different light intensities, as well as predicting and analyzing the moisture content of soil samples. After obtaining the predicted soil characteristic data and moisture content data, the data is used to perform a detailed analysis of the field soil under different light conditions using the compaction degree prediction model, thereby optimizing the parameters of the compaction degree prediction model.

[0026] In summary, this application includes at least one of the following beneficial technical effects: 1. By acquiring soil test sample images and spectral data under different light intensities through the acquisition module, the soil test sample image data is then processed to obtain moisture content estimation through spectral data. Through data processing, a compaction degree prediction model is obtained. The compaction degree prediction model is used to predict the compaction degree of soil samples in the field, and the particle size, distribution characteristics, and shape characteristics in soil test samples under different light conditions are obtained. This improves the accuracy and reliability of soil feature extraction and enhances the accuracy of compaction degree prediction. Compared with the prior art, this application can quickly and accurately assess soil compaction degree.

[0027] 2. After the sampling component completes the collection of soil samples, it stores the collected soil samples in the collection component. When the collection component has collected multiple soil samples, it transfers the multiple soil samples to the image acquisition component, which then acquires images of the multiple soil samples. Attached Figure Description

[0028] Figure 1 This is a schematic diagram of the overall structure of an embodiment of this application.

[0029] Figure 2 This is a partial structural diagram of an embodiment of this application.

[0030] Figure 3 This is a schematic diagram of the sampling component in an embodiment of this application.

[0031] Figure 4 This is a bottom view of the data acquisition component in an embodiment of this application.

[0032] Explanation of reference numerals in the attached drawings: 1. Sample acquisition mechanism; 11. Frame; 12. Lifting drive assembly; 121. Lead screw; 122. Moving block; 13. Guide rod; 14. Rotation drive assembly; 141. Rotary motor; 142. Transmission wheel; 143. Belt; 144. Driven wheel; 145. Rotating shaft; 15. Sampling assembly; 151. Connecting seat; 152. Ring cutter; 153. Clamping block; 154. Clamping slot; 16. Collection assembly; 161. Support frame; 162. Rotary motor; 163. Turntable; 164. Base; 165. Mounting seat; 2. Data acquisition mechanism; 21. Image acquisition assembly; 211. Telescopic cylinder; 212. Mounting plate; 213. Covering cylinder; 214. Illuminating element; 215. Camera; 22. Spectral data acquisition assembly; 221. Near-infrared spectrometer. Detailed Implementation

[0033] The following is in conjunction with the appendix Figure 1-4 This application will be described in further detail.

[0034] This application discloses an automatic soil compaction degree identification system.

[0035] Reference Figure 1 and Figure 2 An automatic soil compaction identification system includes a data acquisition module, which includes a sample acquisition mechanism 1. The sample acquisition mechanism 1 includes a frame 11 and a lifting drive assembly 12. The lifting drive assembly 12 includes a lifting drive component, a lead screw 121, and a moving block 122. In this embodiment, the lifting drive component is a lifting motor, which is installed inside the frame 11 and located at the top of the frame 11. The lead screw 121 is vertically rotatably connected to the frame 11, with its top inserted into the frame 11. The top of the lead screw 121 is assembled with the output end of the lifting motor. The moving block 122 is sleeved on the outer wall of the lead screw 121, and the moving block 122 is threadedly connected to the lead screw 121. When the lifting motor is started, it drives the lead screw 121 to rotate. With the cooperation of the lead screw 121 and the moving block 122, the moving block 122 moves up and down.

[0036] A guide rod 13 is fixed inside the frame 11. The guide rod 13 is vertically installed inside the frame 11 and passes through the moving block 122. The number of guide rods 13 depends on the specific situation. In this embodiment, there are four guide rods 13. The four guide rods 13 are distributed in a rectangle at the four corners of the moving block 122. The guide rods 13 can guide the movement of the moving block 122 and also bear part of the load of the moving block 122, reducing the swaying and instability of the moving block 122 during movement.

[0037] The sample collection mechanism 1 also includes a rotary drive assembly 14. The optional drive assembly includes a rotary drive component, a transmission wheel 142, a belt 143, a driven wheel 144, and a rotating shaft 145. In this embodiment, the rotary drive component is a rotary motor 141, which is mounted on the top surface of the movable plate. The transmission wheel 142 is assembled with the output end of the motor. The rotating shaft 145 moves through the top surface of the movable block 122. The driven wheel 144 is fixed to the top of the rotating shaft 145. One end of the belt 143 is sleeved on the outer wall of the transmission wheel 142, and the other end of the belt 143 is sleeved on the outer wall of the driven wheel 144. When the rotary motor 141 is started, it drives the transmission wheel 142 to rotate. With the cooperation of the transmission wheel 142, the belt 143, and the driven wheel 144, the driven wheel 144 is driven to rotate, thereby driving the rotating shaft 145 to rotate.

[0038] Reference Figure 2 and Figure 3 The sample collection mechanism 1 also includes a sampling component 15, which includes a connecting seat 151 and a ring cutter 152. The connecting seat 151 has a flow cavity inside, which is a bottom-through cavity. The ring cutter 152 has a storage cavity inside, which is a top-to-bottom-through cavity. The top surface of the connecting seat 151 abuts against the bottom surface of the moving block 122. The bottom of the rotating shaft 145 moves through the top surface of the moving block 122, and the bottom of the rotating shaft 145 is assembled with the top surface of the connecting seat 151. The rotating shaft 145 rotates. The connecting seat 151 is rotated. Multiple locking blocks 153 are fixedly installed on the bottom surface of the connecting seat 151. The locking blocks 153 are L-shaped. The top surface of the ring cutter 152 has multiple slots 154, which are L-shaped to match the locking blocks 153. When the connecting seat 151 and the ring cutter 152 are installed, the locking blocks 153 are inserted into the corresponding slots 154, and then the ring cutter 152 is rotated to achieve relative locking between the ring cutter 152 and the connecting seat 151.

[0039] A flow channel is formed through the bottom surface of the frame 11, aligned with the ring cutter 152. The diameter of the flow channel is larger than that of the ring cutter 152, allowing the ring cutter 152 to pass through the flow channel for soil sampling. When soil sampling is required, the lifting motor is activated, causing the moving block 122 to move downwards. The moving block 122 drives the sampling assembly 15 downwards. When the bottom surface of the ring cutter 152 abuts against the top surface of the soil, the rotary motor 141 is activated. The rotary motor 141 drives the connecting seat 151 to rotate in the same direction as the rotation direction locked between the ring cutter 152 and the connecting seat 151. The rotation of the connecting seat 151 drives the ring cutter 152 to rotate, thereby... The soil is broken up, and the moving block 122 moves downward to apply downward pressure to the ring cutter 152, thereby inserting the soil sample device into the storage cavity. The top part of the soil sample that extends beyond the storage cavity enters the flow cavity. The inner diameter of the connecting seat 151 is larger than the inner diameter of the ring cutter 152. Therefore, there is a gap between the soil sample entering the flow cavity and the inner wall of the connecting seat 151. After the soil sample collection is completed, the rotation direction of the lifting motor is changed, thereby driving the sampling component 15 to move upward. When the bottom surface of the ring cutter 152 is far away from the top surface of the soil, the rotating motor 141 stops working, and the ring cutter 152 continues to move upward to return to its original position, thereby completing the collection of the soil sample.

[0040] Reference Figure 2 The sample collection mechanism 1 includes a collection component 16, which includes a moving drive unit and a support frame 161. The structure of the moving drive unit is consistent with the structure of the lifting drive component 12. In this embodiment, it will not be described in detail. The moving drive unit is located on one side of the top of the frame 11, and the long side of the moving drive unit is parallel to the long side of the frame 11. One end of the support frame 161 is connected to the moving drive component, and the other end of the support frame 161 is slidably connected to the inner wall of the frame 11. When the moving drive unit is activated, the moving drive unit can drive the support frame 161 to move along the long side of the frame 11.

[0041] The collecting component 16 also includes a rotation drive, a turntable 163, and a base 164. In this embodiment, the rotation drive is a rotation motor 162. A mounting base 165 is fixed to the bottom of the support frame 161. The rotation motor 162 is installed in the mounting base 165. The middle of the top surface of the turntable 163 is rotatably connected to the bottom surface of the mounting base 165. The motor output end is assembled with the top surface of the turntable 163. The base 164 is fixed to the top surface of the turntable 163. The number of bases 164 depends on the specific situation. In this embodiment, there are four bases 164. The four bases 164 are evenly distributed along the circumference of the top surface of the turntable 163. A connecting groove is opened on the top surface of the base 164. The shape of the connecting groove is adapted to the shape of the bottom of the ring cutter 152.

[0042] Reference Figure 1 and Figure 2After sample collection is completed and the ring cutter 152 returns to its original position, the moving drive unit is activated. The moving unit drives the support frame 161 to move closer to the ring cutter 152 until the base 164 is located at the bottom of the ring cutter 152. Then, the lifting motor is activated to insert the bottom of the ring cutter 152 into the connecting slot. Next, the rotary motor 141 is activated, which drives the connecting seat 151 to rotate, releasing the lock between the connecting seat 151 and the ring cutter 152. Then, the lifting motor is activated again to move the connecting seat 151 upward, thereby placing the ring cutter 152 containing the soil sample on top of the base 164. Correspondingly, the tops of the other three bases 164 are respectively fitted with ring cutters 152, and the soil sample is placed on top of one base 164. Afterwards, start the rotary motor 162, which drives the turntable 163 to rotate, rotating another empty ring cutter 152 to the bottom of the connecting seat 151. Then, start the lifting motor, which drives the connecting seat 151 to move downwards, inserting the locking block 153 into the slot 154 on the top surface of the empty ring cutter 152. Then, the connecting ring moves the empty ring cutter 152 upwards. Then, start the rotary motor 141, which rotates the ring cutter 152 to achieve relative locking between the ring cutter 152 and the connecting seat 151. Then, start the moving drive unit, which drives the turntable 163 to move away from the connecting ring, so that the next ring cutter 152 can continue to collect the next soil sample. Repeat the previous steps until four samples are collected and placed on the top of the corresponding base 164.

[0043] Reference Figure 2 and Figure 4 The acquisition module also includes a data acquisition mechanism 2, which includes an image acquisition component 21. The image acquisition component 21 includes a telescopic drive, a mounting plate 212, and a cover cylinder 213. In this embodiment, the telescopic drive is a telescopic cylinder 211, which is installed on the top surface of the frame 11. The output end of the telescopic cylinder 211 passes through the top surface of the frame 11 and is assembled with the top of the mounting plate 212. The cover cylinder 213 is installed on the bottom surface of the mounting plate 212. The number of cover cylinders 213 is adapted to the number of bases 164. In this embodiment, there are four cover cylinders 213, which are evenly distributed along the bottom surface of the mounting plate 212. Each cover cylinder 213 corresponds to one base 164.

[0044] The mounting plate 212 has a mounting groove near the connecting ring. When the moving drive unit drives the support frame 161 to move away from the connecting ring, the support frame 161 moves along the mounting groove to the middle of the mounting plate 212, so that the four cover cylinders 213 are respectively located on the top of their corresponding bases 164. Then, the telescopic cylinder 211 is activated, and the telescopic cylinder 211 drives the cover cylinders 213 to move downward. The bottom of the cover cylinder 213 is fixed with a connecting block, and the connecting blocks are respectively inserted into the slots 154, so that the cover cylinders 213 cover the top of the ring cutter 152.

[0045] The image acquisition component 21 also includes a light source 214 and a camera 215. The light source 214 is installed at the top end of the cover tube 213, and the light source 214 is aimed at the soil sample. The camera 215 is installed in the middle of the top of the cover tube 213. The lighting conditions inside different cover tubes are different, so that images of the soil sample can be captured under different conditions.

[0046] The data acquisition unit 2 also includes a spectral data acquisition component 22, which includes a near-infrared spectrometer 221. The near-infrared spectrometer 221 is installed at the top inside the cover tube 213. The near-infrared spectrometer 221 is used to acquire spectral data of soil samples to estimate moisture content.

[0047] This application also includes a data processing module, which includes image preprocessing, particle identification and classification, and moisture content estimation.

[0048] After acquiring multiple images through the acquisition module, an image training set is formed. The multiple image training sets are preprocessed to obtain several soil characteristics based on different lighting conditions.

[0049] Image preprocessing includes noise reduction, image enhancement, and normalization.

[0050] Denoising the image training set reduces noise interference with soil feature analysis and improves the quality of soil test sample images. Enhancing features in the training set makes subsequent identification of soil particle features more accurate. Converting the soil test sample images in the training set to a uniform format or range minimizes scale differences between images. Denoising can be performed using algorithms such as Gaussian filtering or median filtering. Image enhancement can be achieved by adjusting contrast. Standardization is achieved through data normalization to ensure image data consistency. Furthermore, abnormal data points in the training set, such as outliers or obviously erroneous data, are cleaned. Through data cleaning, the accuracy and reliability of soil test samples can be improved. Using a convolutional neural network, particles of different sizes, distribution characteristics, and shape characteristics are identified and classified from each soil test sample in the preprocessed image training set.

[0051] In this process, for soil particles, convolution operations are used to extract local features in the image training set of soil test samples. Convolution kernels are designed to identify particle edges, thereby helping to identify particle size and shape. Since the edges of soil particles usually have a large gray-level difference from the background, the ReLU activation function introduces non-linearity and learns a complex function mapping to distinguish particles from the background. Pooling operations are then performed, where max pooling helps to retain the largest features of the particles, while average pooling can smooth noise and improve the model's generalization ability. Finally, the gradient of the loss function with respect to the network parameters is calculated using the backpropagation algorithm, and the parameters are updated using the gradient descent method. In this embodiment, the loss function can be designed as the classification error. By optimizing the error, particles can be identified and classified more accurately.

[0052] Based on the particle size, distribution characteristics, and shape characteristics of each soil test sample in the image training set, a soil feature model is established to obtain several soil features based on different lighting conditions.

[0053] In this process, a portion of the soil test sample images are further divided into an image test set. The image test set is used to test and optimize the soil feature model generated based on the image training set. The model parameters are optimized through multiple iterations to improve the accuracy of the soil feature model prediction.

[0054] In this embodiment, preprocessing the soil test sample images in the image training set can improve the quality and consistency of the images, which helps the subsequent convolutional neural network to analyze the soil test sample images of each type and under different lighting conditions. The convolutional neural network identifies and classifies the features of particles, providing the size, distribution characteristics, and shape characteristics of particles in the soil test samples under different lighting conditions for subsequent compaction degree prediction, thereby improving the accuracy and reliability of soil feature extraction and enhancing the accuracy of compaction degree prediction.

[0055] Near-infrared spectrometer 221 was used to detect each training set in the sample training set, so that the samples covered a wide range of soil moisture content, so that the moisture content model could accurately predict different moisture content levels.

[0056] Based on spectral data, the moisture content of each soil test sample in the sample training set is obtained through machine learning, and a moisture content model is established to obtain the moisture content of each soil test sample.

[0057] In this embodiment, a sample training set is constructed by collecting spectral data from soil test samples. Machine learning algorithms are used to analyze and calculate the correlation between spectral data features and soil moisture content. Based on the moisture content data of a large number of test samples, a moisture content model capable of accurately predicting soil moisture content is constructed. Using this moisture content model, the soil moisture content can be predicted after obtaining the spectral data of the soil in the field, providing moisture content-based data for subsequent compaction degree prediction.

[0058] This application also includes a compaction degree prediction module, which includes feature extraction, compaction degree prediction model, model training and validation. Based on the soil characteristics of each soil test sample under different light intensities and the moisture content of the soil test sample, a compaction degree prediction model is generated through machine learning algorithms.

[0059] Feature extraction includes particle size distribution, shape features, and moisture content.

[0060] Soil is composed of particles of different sizes, including sand, silt, and clay. The size and distribution of these particles, as well as soil moisture, can affect the compactness of soil structure and soil plasticity.

[0061] Moisture can fill the gaps between soil particles, increasing the soil's cohesion and thus affecting its compaction.

[0062] Soil porosity reflects the proportion of void volume to total volume in soil. Porosity is closely related to soil particle size, distribution, and moisture content. The size of porosity directly affects soil compaction; the smaller the porosity, the denser the soil and the higher the compaction.

[0063] Soil particle size, distribution characteristics, shape characteristics, and moisture content also affect soil density. Soil density can be divided into dry density and wet density. Dry density refers to the mass of the solid portion of soil per unit volume, while wet density includes the mass of both soil solids and water. An increase in soil density usually means an increase in soil compaction.

[0064] Knowing the size, distribution characteristics, shape characteristics, and moisture content of soil particles in the soil test sample, and combining this with soil physics, we start from soil particle structure, soil plasticity, soil moisture content, soil porosity, and soil density, and select machine learning algorithms such as random forest and gradient booster to generate a compaction rate prediction model based on soil characteristics and moisture content.

[0065] Model training and validation: Based on the light intensity in different areas of the soil image, the soil image is divided into several test areas.

[0066] In this process, several soil images are converted to grayscale. By detecting the grayscale values ​​of the soil images, each soil image is divided into test areas with different grayscale values.

[0067] The soil compaction degree of each test area was obtained by using a compaction degree prediction model to analyze several test areas and the corresponding field soil in those areas.

[0068] In this embodiment, a compaction degree prediction model is used to predict the compaction degree of soil images and soil samples from the field. This includes predicting and analyzing the soil characteristics of several test areas with different light intensities, as well as predicting and analyzing the moisture content of soil samples. After obtaining the predicted soil characteristic data and moisture content data, the data is used to perform a detailed analysis of the field soil under different light conditions using the compaction degree prediction model.

[0069] The above are all preferred embodiments of this application. These embodiments are merely explanations of this application and are not intended to limit the scope of protection of this application. Therefore, all equivalent changes made in accordance with the structure, shape, and principle of this application should be covered within the scope of protection of this application.

Claims

1. An automatic soil compaction degree identification system, characterized in that, include: Acquisition Module: The acquisition module is used to collect soil samples, obtain soil test sample images under different light intensities through the collected soil samples, and obtain spectral data of the soil samples through the collected soil samples; Data processing module: The data processing module includes image preprocessing, particle recognition and classification, and moisture content estimation. Image preprocessing is used to acquire multiple soil features based on different lighting conditions. Particle recognition and classification is used to identify and classify particles of different sizes and shapes in soil test sample images. Soil moisture content is estimated through the spectral data of soil samples. Compaction Prediction Module: Based on the soil characteristics of each soil test sample under different light intensities and the moisture content of the soil test sample, a compaction prediction model is generated through machine learning algorithms.

2. The soil compaction degree automatic identification system according to claim 1, characterized in that, The acquisition module includes a sample acquisition mechanism (1), which is used for collecting soil samples.

3. The soil compaction degree automatic identification system according to claim 2, characterized in that, The acquisition module also includes a data acquisition mechanism (2), which includes an image acquisition component (21). The image acquisition component (21) is located at the output end of the sample acquisition mechanism (1) and is used to capture soil sample images.

4. The soil compaction degree automatic identification system according to claim 3, characterized in that, The data acquisition mechanism (2) also includes a spectral data acquisition component (22), which is used to acquire spectral data of soil samples.

5. The soil compaction degree automatic identification system according to claim 2, characterized in that, The sample collection mechanism (1) includes a lifting drive component (12) and a sampling component (15). The sampling component (15) is located at the output end of the lifting drive component (12) and is used for collecting soil samples.

6. The soil compaction degree automatic identification system according to claim 3, characterized in that, The image acquisition component (21) includes a light source (214) and a camera (215). The light source (214) is used to provide illumination for the soil sample, and the camera (215) is used to acquire images of the soil sample.

7. The soil compaction degree automatic identification system according to claim 3, characterized in that, The sample collection mechanism (1) includes a collection component (16) for collecting the collected soil samples and transporting the soil samples to the image acquisition component (21).

8. The soil compaction degree automatic identification system according to claim 4, characterized in that, The spectral data acquisition component (22) includes a near-infrared spectrometer (221), which is used to acquire spectral data of soil samples.

9. The soil compaction degree automatic identification system according to claim 1, characterized in that, Image preprocessing includes noise reduction, image enhancement, and normalization.

10. The soil compaction degree automatic identification system according to claim 1, characterized in that, The compaction degree prediction module includes model training and validation. Model training and validation includes acquiring soil images, soil and corresponding soil compaction degrees collected on site, and establishing a validation set; the validation set and the corresponding on-site soil are validated using the compaction degree prediction model.

Images