Grain carrying vehicle grain loading state detection system and method based on visual identification

Abstract

The invention belongs to the technical field of grain loading supervision, and particularly relates to a grain transport vehicle grain loading state detection system and method based on visual identification, and the system comprises a visual collection module, an image processing and analysis module, a loading state evaluation module, an intelligent sampling auxiliary module and a cloud management and control module. Real-time pictures of grains in the vehicle are captured through the visual acquisition module, the image processing and analyzing module generates a three-dimensional distribution image of the grains based on the real-time pictures, the loading state evaluation module judges whether the grain loading state of the grain transporting vehicle is qualified or not according to the grain distribution image, the loading problem can be found and corrected in time, and the working efficiency is improved. The intelligent sampling auxiliary module automatically plans the sampling path and depth of the intelligent sampling machine according to the grain loading state information synchronized by the cloud management and control module, so that accurate and efficient automatic intelligent sampling is realized, and the accuracy of grain quality detection is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of grain loading supervision, and specifically to a grain loading status detection system and method for grain transport vehicles based on visual recognition. Background Art

[0002] The traditional management of grain loading in grain transport vehicles mainly relies on manual observation and empirical judgment, which has problems such as strong subjectivity, low efficiency, and large errors. With the development of Internet of Things, big data, and artificial intelligence technologies, automation and intelligence have become important directions for the upgrading of agricultural equipment;

[0003] However, there are still deficiencies in the real-time monitoring, accurate evaluation of the grain loading status, and the linkage with subsequent sampling processes. Moreover, it is unable to automatically adaptively adjust the subsequent sampling process and reasonably evaluate the sampling execution performance, and the application in intelligent sampling still needs to be deepened;

[0004] In view of the above technical deficiencies, a solution is proposed. Summary of the Invention

[0005] The purpose of the present invention is to provide a grain loading status detection system and method for grain transport vehicles based on visual recognition, which solves the problems that the existing technology still has deficiencies in the real-time monitoring, accurate evaluation of the grain loading status, and the linkage with subsequent sampling processes, and is unable to automatically adaptively adjust the subsequent sampling process and reasonably evaluate the sampling execution performance, and the application in intelligent sampling still needs to be deepened.

[0006] To achieve the above purpose, the present invention provides the following technical solutions:

[0007] A grain loading status detection system for grain transport vehicles based on visual recognition includes a visual acquisition module, an image processing and analysis module, a loading status evaluation module, an intelligent sampling assistance module, and a cloud control module; the visual acquisition module captures real-time images of the grain in the vehicle through a high-definition camera installed on the grain transport vehicle, supports multi-angle and high-resolution shooting, and comprehensively covers the grain loading area;

[0008] The image processing and analysis module receives the real-time images transmitted by the visual acquisition module, analyzes the images using image processing algorithms and machine learning models, generates a three-dimensional distribution image of the grain, and sends the image processing and analysis information to the loading status evaluation module;

[0009] The loading status evaluation module automatically evaluates the grain loading status based on a preset loading standard and according to the grain distribution image, determines whether the grain loading status of the grain transport vehicle is qualified, and sends the judgment information to the cloud control module; the intelligent sampling assistance module automatically plans the sampling path and depth of the intelligent sampler according to the grain loading status information synchronized by the cloud control module, so that the sampling points are reasonably distributed.

[0010] Furthermore, the image processing and analysis module is also used to judge the type, humidity and distribution uniformity of grains by analyzing image features, where the image features include color, texture and shape.

[0011] Furthermore, the intelligent sampling assistance module is communicatively connected to the sampling monitoring and transmission module and the sampling adjustment control module. The sampling monitoring and transmission module monitors the sampling process of the intelligent sampler, judges whether an adjustment warning signal is generated based on the monitoring information, and sends the adjustment warning signal to the sampling adjustment control module when the adjustment warning signal is generated; after receiving the adjustment warning signal, the sampling adjustment control module performs adaptive adjustment control on the intelligent sampler.

[0012] Furthermore, the specific judgment process for judging whether an adjustment warning signal is generated based on the monitoring information is as follows:

[0013] Monitor the sampling path and running speed of the intelligent sampler, and accordingly judge whether the sampling path of the intelligent sampler deviates and collect its running speed. When the sampling path of the intelligent sampler deviates or the running speed is not within the preset running speed range, an adjustment warning signal is generated.

[0014] Furthermore, the sampling adjustment control module is communicatively connected to the sampling execution evaluation module. The sampling execution evaluation module is used to set a monitoring period, analyze the sampling execution performance of the intelligent sampler within the monitoring period, generate a sampling execution qualified signal or a sampling execution warning signal through the analysis, and send the sampling execution qualified signal or the sampling execution warning signal to the cloud control module, and strengthen sampling supervision when the sampling execution warning signal is generated.

[0015] Furthermore, the specific analysis process of the sampling execution evaluation module includes:

[0016] Collect the number of times that the sampling path deviates within the monitoring period and mark it as the sampling path deviation frequency coefficient, and collect the number of times that the running speed is not within the preset running speed range within the monitoring period and mark it as the running speed deviation frequency coefficient;

[0017] Compare the sampling path deviation frequency coefficient and the running speed deviation frequency coefficient with the preset sampling path deviation frequency coefficient threshold and the preset running speed deviation frequency coefficient threshold respectively. If the sampling path deviation frequency coefficient or the running speed deviation frequency coefficient exceeds the corresponding preset threshold, a sampling execution warning signal is generated.

[0018] Furthermore, if both the sampling path deviation frequency coefficient and the running speed deviation frequency coefficient do not exceed the corresponding preset thresholds, collect the generation moment of the adjustment warning signal and mark it as the adjustment sending moment, and collect the moment when the corresponding adjustment control is completed and mark it as the adjustment end moment, and calculate the time difference between the adjustment end moment and the adjustment sending moment to obtain the adjustment interval value;

[0019] The adjustment interval value is numerically compared with a preset adjustment interval threshold. If the adjustment interval value exceeds the preset adjustment interval threshold, the corresponding adjustment interval value is marked as an adjustment specific value; and the average value of all adjustment interval values during the monitoring period is calculated to obtain an adjustment analysis value, and the number of adjustment specific values within the monitoring period is marked as an adjustment abnormal frequency value; the adjustment analysis value and the adjustment abnormal frequency value are numerically compared with a preset adjustment analysis threshold and a preset adjustment abnormal frequency threshold respectively. If the adjustment analysis value or the adjustment abnormal frequency value exceeds the corresponding preset threshold, a sampling execution warning signal is generated.

[0020] Furthermore, if both the adjustment analysis value and the adjustment abnormal frequency value do not exceed the corresponding preset thresholds, a sampling execution evaluation coefficient is obtained by numerically calculating the sampling path frequency deviation coefficient, the transportation speed frequency deviation coefficient, the adjustment analysis value, and the adjustment abnormal frequency value. The sampling execution evaluation coefficient is numerically compared with a preset sampling execution evaluation coefficient threshold. If the sampling execution evaluation coefficient exceeds the preset sampling execution evaluation coefficient threshold, a sampling execution warning signal is generated; if the sampling execution evaluation coefficient does not exceed the preset sampling execution evaluation coefficient threshold, a sampling execution qualified signal is generated.

[0021] Furthermore, the present invention also proposes a method for detecting the grain loading state of a grain transport vehicle based on visual recognition, including the following steps:

[0022] Step 1: The visual acquisition module captures the real-time image of the grain in the vehicle and sends the real-time image to the image processing and analysis module;

[0023] Step 2: The image processing and analysis module analyzes the image using image processing algorithms and machine learning models to generate a three-dimensional distribution image of the grain;

[0024] Step 3: The loading state evaluation module automatically evaluates the grain loading state based on a preset loading standard and according to the grain distribution image to determine whether the grain loading state of the grain transport vehicle is qualified;

[0025] Step 4: The cloud control module synchronizes the grain loading state information and shares it with the intelligent sampling assistance module;

[0026] Step 5: The intelligent sampling assistance module automatically plans the sampling path and depth of the intelligent sampler according to the grain loading state information.

[0027] Compared with the prior art, the beneficial effects of the present invention are:

[0028] 1. In the present invention, the visual acquisition module captures the real-time images of the grains in the vehicle, the image processing and analysis module generates the three-dimensional distribution images of the grains based on the real-time images, and the loading state evaluation module determines whether the grain loading state of the grain transport vehicle is qualified according to the grain distribution images, which can timely detect and correct the loading problems, avoid potential safety hazards such as overloading and uneven loading. Moreover, the intelligent sampling assistant module automatically plans the sampling path and depth of the intelligent sampler according to the grain loading state information synchronized by the cloud control module, realizing accurate and efficient automated intelligent sampling and improving the accuracy of grain quality detection.

[0029] 2. In the present invention, the sampling monitoring and transmission module monitors the sampling process of the intelligent sampler to determine whether an adjustment warning signal is generated. After the adjustment warning signal is generated, the sampling adjustment control module performs adaptive adjustment control on the intelligent sampler, with high automation and intelligence levels, reducing the operation difficulty and workload of the management personnel, ensuring the sampling accuracy and efficiency. Moreover, the sampling execution evaluation module analyzes the sampling execution performance of the intelligent sampler during the monitoring period, and strengthens the sampling supervision when the sampling execution warning signal is generated, ensuring the sampling execution accuracy and execution stability. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] For the convenience of those skilled in the art to understand, the present invention will be further described below with reference to the accompanying drawings.

[0031] Figure 1 It is the system block diagram of the first embodiment in the present invention.

[0032] Figure 2 It is the system block diagram of the second and third embodiments in the present invention.

[0033] Figure 3 It is the method flow chart of the fourth embodiment in the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0034] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0035] Embodiment 1: As Figure 1 shown, the grain loading state detection system of the grain transport vehicle based on visual recognition proposed by the present invention includes a visual acquisition module, an image processing and analysis module, a loading state evaluation module, an intelligent sampling assistant module, and a cloud control module; the visual acquisition module captures the real-time images of the grains in the vehicle through a high-definition camera installed on the grain transport vehicle, supporting multi-angle and high-resolution shooting, and comprehensively covering the grain loading area.

[0036] The image processing and analysis module receives the real-time images transmitted by the visual acquisition module, analyzes the images using image processing algorithms (such as edge detection, morphological processing, etc.) and machine learning models to generate a three-dimensional distribution image of the grains, and judges the type, humidity and distribution uniformity of the grains by analyzing the image features. The image features include color, texture and shape, and sends the image processing and analysis information to the loading state evaluation module.

[0037] The loading state evaluation module automatically evaluates the grain loading state based on preset loading standards (such as loading height, weight distribution, safety threshold, etc.) and according to the grain distribution image, judges whether the grain loading state of the grain transport vehicle is qualified, including whether there are problems such as overloading, partial loading or voids, and sends the judgment information to the cloud control module. Through automated monitoring and evaluation, loading problems can be discovered and corrected in a timely manner, avoiding safety hazards such as overloading and partial loading, and improving the transportation efficiency.

[0038] The cloud control module centrally stores the data, and can further use big data analysis technology to optimize the loading strategy, predict the loading efficiency, and provide remote monitoring and decision-making support for the manager. At the same time, the cloud control module is connected to the intelligent sampling auxiliary module through the API interface to realize the synchronous sharing of the loading state information.

[0039] The intelligent sampling auxiliary module automatically plans the sampling path and depth of the intelligent sampler according to the grain loading state information synchronized by the cloud control module, makes the sampling points reasonably distributed, and realizes accurate and efficient automated sampling by integrating machine vision and robotic arm technology, reducing human intervention, improving the accuracy and safety of sampling, and thus improving the accuracy of grain quality inspection.

[0040] Embodiment 2: As Figure 2 shown, the difference between this embodiment and Embodiment 1 is that the intelligent sampling auxiliary module is communicatively connected to the sampling monitoring and transmission module and the sampling adjustment and control module. The sampling monitoring and transmission module monitors the sampling process of the intelligent sampler (mainly monitors its running path and running speed), and judges whether an adjustment warning signal is generated based on the monitoring information. Specifically:

[0041] Monitor the sampling path and running speed of the intelligent sampler, and accordingly judge whether the sampling path of the intelligent sampler deviates and collect its running speed. When the sampling path of the intelligent sampler deviates or the running speed is not within the preset running speed range, an adjustment warning signal is generated.

[0042] And when generating an adjustment warning signal, it is sent to the sampling adjustment control module. After receiving the adjustment warning signal, the sampling adjustment control module performs adaptive adjustment control on the intelligent sampling machine, with high automation and intelligence levels, reducing the operation difficulty and workload of management personnel, and ensuring sampling accuracy and efficiency.

[0043] Embodiment 3: As Figure 2 shown, the difference between this embodiment and Embodiment 1 and Embodiment 2 is that the sampling adjustment control module is communicatively connected to the sampling execution evaluation module. The sampling execution evaluation module is used to set a monitoring period, analyze the sampling execution performance of the intelligent sampling machine within the monitoring period, and generate a sampling execution qualified signal or a sampling execution warning signal through the analysis;

[0044] And send the sampling execution qualified signal or the sampling execution warning signal to the cloud control module, strengthen sampling supervision when generating the sampling execution warning signal, and ensure sampling execution accuracy and execution stability; The specific analysis process of the sampling execution evaluation module is as follows:

[0045] Collect the number of times the sampling path deviates within the monitoring period and mark it as the sampling path deviation frequency coefficient, and collect the number of occurrences where the running speed is not within the preset running speed range within the monitoring period and mark it as the speed deviation frequency coefficient;

[0046] Compare the sampling path deviation frequency coefficient and the speed deviation frequency coefficient with the preset sampling path deviation frequency coefficient threshold and the preset speed deviation frequency coefficient threshold respectively. If the sampling path deviation frequency coefficient or the speed deviation frequency coefficient exceeds the corresponding preset threshold, it indicates that the sampling execution performance in the monitoring period is poor, and a sampling execution warning signal is generated;

[0047] If both the sampling path deviation frequency coefficient and the speed deviation frequency coefficient do not exceed the corresponding preset thresholds, then collect the generation moment of the adjustment warning signal and mark it as the adjustment sending moment, and collect the moment when the corresponding adjustment control is completed and mark it as the adjustment end moment. Calculate the time difference between the adjustment end moment and the adjustment sending moment to obtain the adjustment interval value; Among them, the larger the value of the adjustment interval value, the lower the adjustment efficiency for the corresponding adjustment process;

[0048] Compare the adjustment interval value with the preset adjustment interval threshold. If the adjustment interval value exceeds the preset adjustment interval threshold, it indicates that the adjustment efficiency for the corresponding adjustment process is low, and then mark the corresponding adjustment interval value as an adjustment special value; And calculate the average value of all adjustment interval values in the monitoring period to obtain an adjustment analysis value, and mark the number of adjustment special values in the monitoring period as an adjustment deviation frequency value;

[0049] The adjusted analysis value and the adjusted different-frequency value are respectively compared numerically with the preset adjusted analysis threshold and the preset adjusted different-frequency threshold. If the adjusted analysis value or the adjusted different-frequency value exceeds the corresponding preset threshold, it indicates that the sampling execution performance during the monitoring period is poor, and a sampling execution warning signal is generated.

[0050] Furthermore, if both the adjusted analysis value and the adjusted different-frequency value do not exceed the corresponding preset thresholds, the sampling path frequency deviation coefficient X, the transport speed frequency deviation coefficient L, the adjusted analysis value W, and the adjusted different-frequency value N are numerically calculated through the formula ZF = m×X + e×L + q×W + c×N to obtain the sampling execution evaluation coefficient ZF; where m, e, q, and c are preset proportionality coefficients greater than zero, and the larger the numerical value of the sampling execution evaluation coefficient ZF, the worse the comprehensive sampling execution performance during the monitoring period;

[0051] The sampling execution evaluation coefficient is compared numerically with the preset sampling execution evaluation coefficient threshold. If the sampling execution evaluation coefficient exceeds the preset sampling execution evaluation coefficient threshold, it indicates that the comprehensive sampling execution performance during the monitoring period is poor, and a sampling execution warning signal is generated; if the sampling execution evaluation coefficient does not exceed the preset sampling execution evaluation coefficient threshold, it indicates that the comprehensive sampling execution performance during the monitoring period is good, and a sampling execution qualified signal is generated.

[0052] Example 4: As Figure 3 shown, the difference between this embodiment and Embodiment 1, Embodiment 2, and Embodiment 3 is that the method for detecting the grain loading state of a grain transport vehicle based on visual recognition proposed by the present invention includes the following steps:

[0053] Step 1: The visual acquisition module captures the real-time image of the grain in the vehicle and sends the real-time image to the image processing and analysis module;

[0054] Step 2: The image processing and analysis module analyzes the image using image processing algorithms and machine learning models to generate a three-dimensional distribution image of the grain;

[0055] Step 3: The loading state evaluation module automatically evaluates the grain loading state based on the preset loading standard and according to the grain distribution image to determine whether the grain loading state of the grain transport vehicle is qualified;

[0056] Step 4: The cloud control module synchronizes the grain loading state information and shares it with the intelligent sampling assistance module;

[0057] Step 5: The intelligent sampling assistance module automatically plans the sampling path and depth of the intelligent sampler according to the grain loading state information.

[0058] Working principle of the present invention: During use, the visual acquisition module captures real-time images of the grains in the vehicle. The image processing and analysis module analyzes the images using image processing algorithms and machine learning models to generate a three-dimensional distribution image of the grains. The loading state evaluation module automatically evaluates the grain loading state based on the grain distribution image to determine whether the grain loading state of the grain transport vehicle is qualified, enabling timely detection and correction of loading problems, avoiding safety hazards such as overloading and uneven loading, improving the transportation efficiency. Moreover, through the intelligent sampling assistance module, according to the grain loading state information synchronized by the cloud control module, it automatically plans the sampling path and depth of the intelligent sampler, realizing precise and efficient automated intelligent sampling, improving the sampling accuracy and safety, and significantly enhancing the accuracy of grain quality inspection.

[0059] The above formulas are all dimensionless and take their numerical calculations. The formulas are obtained by collecting a large amount of data for software simulation to get a formula closest to the actual situation. The preset parameters in the formulas are set by those skilled in the art according to the actual situation. The preferred embodiments of the present invention disclosed above are only used to help illustrate the present invention. The preferred embodiments do not elaborate on all the details, nor do they limit the invention to only the specific implementation manners. Obviously, according to the content of this specification, many modifications and variations can be made. This specification selects and specifically describes these embodiments to better explain the principle and practical application of the present invention, so that those skilled in the relevant technical field can well understand and utilize the present invention. The present invention is only limited by the claims and their full scope and equivalents.

Claims

1. A grain truck loading status detection system based on visual recognition, characterized in that: It includes a visual acquisition module, an image processing and analysis module, a loading status assessment module, an intelligent sampling auxiliary module, and a cloud management and control module. The visual acquisition module captures real-time images of grains in the vehicle through a high-definition camera installed on the grain transport vehicle, supports multi-angle, high-resolution shooting, and fully covers the grain loading area. The image processing and analysis module receives the real-time images transmitted by the visual acquisition module, analyzes the images using image processing algorithms and machine learning models, generates a three-dimensional distribution image of the grains, and sends the image processing and analysis information to the loading status assessment module; The loading status assessment module automatically assesses the grain loading status based on the preset loading standards and the grain distribution image, determines whether the grain loading status of the grain transport vehicle is qualified, and sends the judgment information to the cloud management and control module; the intelligent sampling auxiliary module automatically plans the sampling path and depth of the intelligent sampling machine based on the grain loading status information synchronized by the cloud management and control module, so that the sampling points are reasonably distributed.

2. The grain loading status detection system for grain transport vehicles based on visual recognition according to claim 1 is characterized in that: The image processing and analysis module is also used to determine the type, moisture and distribution uniformity of the grain by analyzing image features, wherein the image features include color, texture and shape.

3. The grain loading status detection system for grain transport vehicles based on visual recognition according to claim 1 is characterized in that: The intelligent sampling auxiliary module is communicatively connected to the sampling monitoring and transmission module and the sampling adjustment control module. The sampling monitoring and transmission module monitors the sampling process of the intelligent sampling machine, determines whether to generate an adjustment warning signal based on the monitoring information, and sends the adjustment warning signal to the sampling adjustment control module when it is generated; after receiving the adjustment warning signal, the sampling adjustment control module performs adaptive adjustment control on the intelligent sampling machine.

4. The grain loading status detection system for grain transport vehicles based on visual recognition according to claim 3 is characterized in that: The specific judgment process for determining whether to generate an adjustment warning signal based on monitoring information is as follows: When the sampling path of the intelligent sampling machine deviates or the running speed is not within the preset running speed range, an adjustment warning signal is generated.

5. The grain loading status detection system for grain transport vehicles based on visual recognition according to claim 3 is characterized in that: The sampling adjustment control module is communicatively connected to the sampling execution evaluation module. The sampling execution evaluation module is used to set a monitoring period, analyze the sampling execution performance of the intelligent sampling machine within the monitoring period, and generate a sampling execution qualified signal or a sampling execution early warning signal through the analysis.

6. The grain loading status detection system for grain transport vehicles based on visual recognition according to claim 5 is characterized in that: The specific analysis process of the sampling execution evaluation module includes: The number of deviations of the sampling path during the monitoring period is collected and marked as the sampling path frequency deviation coefficient, and the number of occurrences during the monitoring period when the running speed is not within the preset running speed range is collected and marked as the speed frequency deviation coefficient; if the sampling path frequency deviation coefficient or the speed frequency deviation coefficient exceeds the corresponding preset threshold, a sampling execution warning signal is generated.

7. The grain loading status detection system for grain transport vehicles based on visual recognition according to claim 6 is characterized in that: If the sampling path frequency deviation coefficient and the speed frequency deviation coefficient do not exceed the corresponding preset threshold, all adjustment interval values ​​in the monitoring period are averaged to obtain the adjustment analysis value, and the number of adjustment specific values ​​in the monitoring period is marked as the adjustment frequency deviation value; If the adjusted analysis value or the adjusted frequency difference value exceeds the corresponding preset threshold, a sampling execution warning signal is generated.

8. The grain loading status detection system for grain transport vehicles based on visual recognition according to claim 7 is characterized in that: If neither the adjusted analysis value nor the adjusted frequency deviation value exceeds the corresponding preset threshold value, the sampling evaluation coefficient is obtained by numerically calculating the sampling path frequency deviation coefficient, the speed frequency deviation coefficient, the adjusted analysis value and the adjusted frequency deviation value. If the sampling evaluation coefficient exceeds the preset sampling evaluation coefficient threshold, a sampling execution warning signal is generated; otherwise, a sampling execution qualified signal is generated.

9. A method for detecting grain loading status of a grain transport vehicle based on visual recognition, characterized in that: The method adopts the grain loading status detection system for a grain transport vehicle based on visual recognition as described in any one of claims 1-8.

Images