CN118536854A - Method and device for intelligently detecting and screening quality of pasture seeds - Google Patents

Abstract

The present invention belongs to the field of biological agricultural technology, and discloses a method and device for intelligent detection and screening of forage seed quality, including: an image acquisition unit, a physical property detection unit, a data processing and analysis unit, an automatic screening unit, and a user interaction unit. The data analysis and processing unit is arranged in the upper left corner of the device, the image acquisition unit is arranged on the right side of the data analysis and processing unit, the user interaction unit is arranged below the data analysis and processing unit, the physical property detection unit is arranged in the middle of the device, and an automatic screening unit is arranged on the right side of the physical property detection unit. The design of the present invention takes into account the real-time, accuracy, stability and user-friendliness of the system, and aims to provide users with an efficient and reliable intelligent detection and screening solution for forage seed quality.

Description

A method and device for intelligent detection and screening of forage grass seed quality

Technical Field

The invention belongs to the technical field of biological agriculture, and in particular relates to a method and a device for intelligent detection and screening of forage grass seed quality.

Background Art

Forage grass refers to plants cultivated as livestock feed. In a broad sense, forage grass includes green fodder and crops. The best conditions for forage grass are vigorous growth, tender grass quality, high yield per unit area, strong regeneration ability, multiple harvests a year, good palatability for livestock, rich in crude protein, phosphorus, calcium and appropriate crude fiber, etc., which cannot be replaced by grain and other feeds. Most of the forage grass seeds in my country are native plants, and some are plants introduced from abroad and have been planted for many years. In order to ensure the planting quality of forage grass and obtain higher production capacity benefits, various technical treatments should be adopted after the seeds are harvested to remove impurities, improve purity, and maintain high vitality of the seeds. The seeds should be tested and screened before sowing to ensure that the seeds are pure, full, uniform, moderately moist, vigorous and free of pests and diseases. Traditional forage seed testing and screening, due to the lack of professional high-performance mechanical equipment, is generally done manually with the help of simple tools such as blowers, dustpans, sieves, basins, or visual inspection, manual picking, and empirical evaluation. The screened seeds are often of poor quality, low purity, uneven size, and uneven seed vitality, and cannot meet the technical standards of GB/2930.8-2017 "Forage Seed Inspection Procedure Moisture Determination".

Through the above analysis, the problems and defects existing in the prior art are: there is still a lack of methods and devices on the market that can perform high-precision detection and screening of forage seed quality.

Summary of the invention

In view of the problems existing in the prior art, the present invention provides a method and device for intelligent detection and screening of forage grass seed quality.

The present invention is implemented as follows: a method for intelligent detection and screening of forage grass seed quality, comprising the following steps:

Image acquisition: Use a high-resolution camera to capture images of forage seeds, and turn on the LED lighting to provide sufficient lighting conditions for the camera to ensure that the captured images are clear.

Physical property testing: The collected forage seeds are placed on a vibration platform, which is used to simulate the actual movement state of the seeds and observe their vibration response. At the same time, a fan is used to generate airflow to screen the seeds and remove lightweight and unqualified seeds. Finally, a weight sensor is used to measure the weight of the seeds as an important parameter for judging seed quality.

Data processing and analysis: The collected images and physical property data are transmitted to the data processing and analysis unit, and the CPU is used to identify and analyze the images to determine the shape, size, color and other characteristics of the seeds; at the same time, the physical property data is combined to comprehensively evaluate the quality of the seeds.

Automated screening: Based on the results of data processing and analysis, the seeds are automatically screened using a sorting robot arm. Qualified seeds are transported to the designated location via a conveyor belt, while unqualified seeds are removed.

User interaction: Screening results and data analysis reports are displayed on the touch screen for users to view and operate. Users can adjust screening parameters as needed to achieve personalized seed quality testing and screening.

This method uses intelligent detection and screening technology to achieve rapid and accurate detection and screening of forage grass seed quality, improves screening efficiency and quality, and reduces the difficulty and cost of manual operation.

The present invention also provides an intelligent detection and screening method for forage seed quality, and the data processing and analysis steps include: using a pre-trained machine learning model to extract and classify features of the collected images, and identify features such as seed shape, size, and color; at the same time, combined with physical property data, the quality of the seeds is quantitatively evaluated through an algorithm model, and a corresponding quality evaluation report is generated; wherein the machine learning model includes at least a convolutional neural network model, which is used to extract feature information of the seeds from the image; the algorithm model is based on statistical principles, combined with seed shape, physical properties and historical data, to construct a prediction model to accurately predict the quality of the seeds.

The present invention also provides a forage grass seed quality intelligent detection and screening device, comprising:

Image acquisition unit, physical property detection unit, data processing and analysis unit, automatic screening unit, user interaction unit, the data analysis and processing unit is arranged at the upper left corner of the device, the image acquisition unit is arranged on the right side of the data analysis and processing unit, the user interaction unit is arranged below the data analysis and processing unit, the physical property detection unit is arranged in the middle of the device, and the automatic screening unit is arranged on the right side of the physical property detection unit.

Furthermore, the image acquisition unit includes a high-resolution camera and an LED lighting lamp. The high-resolution camera is adsorbed on the left side wall of the device, and the LED lighting lamp is set on the high-resolution camera and distributed around the lens of the high-resolution camera.

Furthermore, the physical property detection unit includes a vibration platform, a fan, and a weight sensor. The weight sensor is arranged at the bottom of the middle part of the device, the vibration platform is arranged on the top of the weight sensor, and the fan is arranged on the upper part of the weight sensor.

Furthermore, the data processing and analysis unit includes a CPU and a storage unit. The storage unit is arranged at the upper left corner of the device, and the CPU is located at the lower right of the storage unit.

Furthermore, the automated screening unit includes a sorting robot arm and a conveyor belt. The sorting robot arm is located on the right side of the device. A conveyor belt is provided at the bottom of the sorting robot arm and is connected to the physical property detection unit.

Furthermore, the user interaction unit includes a touch screen display, which is arranged below the data processing and analysis unit.

The present invention provides an image acquisition module for forage seed quality detection, comprising at least one high-resolution camera capable of acquiring clear forage seed images under a variety of lighting conditions; at least one group of LED lighting lamps designed with adjustable brightness and lighting angles to adapt to the lighting requirements of different environments and seed types, ensuring that the camera can capture high-quality images; and a control unit for adjusting camera settings and the lighting intensity of the LED lighting lamps to optimize the image acquisition process.

The present invention provides a physical property detection module for forage seeds, comprising a vibration platform with adjustable frequency for simulating motion states of different intensities to detect the stability and response of seeds in an actual environment; a wind screening device equipped with a fan with adjustable speed and a direction control mechanism to achieve accurate wind screening of seeds; a weight sensor group with high-precision measurement function, capable of measuring the weight of single or batch seeds, and automatically transmitting data to an analysis unit to provide key data support for seed quality assessment.

The present invention provides a data processing and analysis unit for forage seed quality detection, which integrates a high-performance CPU and special image processing software, can efficiently process and analyze the collected image data, identify the size, shape, color and other characteristics of the seeds, and compare them with historical data; combine the physical property measurement results, use the preset algorithm model and scoring mechanism to classify the seeds by quality level; provide a data interface, support data export and integration with other systems.

The present invention provides a forage seed screening and user interaction integrated module, which integrates a high-precision sorting robot arm, has multi-axis control and fine motion simulation capabilities, and accurately sorts and places seeds according to analysis results; the conveyor belt system is designed with speed control and sorting area indication to ensure the efficiency and accuracy of seed sorting; the touch screen display provides an intuitive user interface to display real-time data analysis results, screening progress and historical records, and users can input operating instructions, set parameters and query results through the touch screen, thereby realizing the convenience and practicality of interaction.

In combination with the above technical solutions and the technical problems solved, the advantages and positive effects of the technical solutions to be protected by the present invention are as follows:

First, the design of the present invention takes into account the real-time, accuracy, stability and user-friendliness of the system, aiming to provide users with an efficient and reliable forage seed quality detection and screening solution.

The technical problems in the prior art targeted by the present invention mainly focus on the automation and accuracy of seed detection and sorting. The traditional seed detection and sorting process often relies on manual operation, which is not only inefficient but also easily affected by subjective judgment, resulting in inconsistent and inaccurate results. In addition, traditional methods are difficult to capture the tiny features and subtle differences of seeds, thus limiting the accuracy and reliability of sorting.

Second, the present invention solves these problems and achieves significant technical progress through the following aspects:

1. Automation and efficiency: By integrating high-resolution cameras, LED lighting, vibration platforms, fans and other equipment, the present invention realizes the automation of seed detection and sorting processes, significantly improving operational efficiency. The automation process reduces manual intervention and increases processing speed, allowing seed detection and sorting to be completed in a short time.

2. Accuracy and reliability: By capturing seed images with a high-resolution camera and analyzing them with advanced image processing and machine learning algorithms, the present invention can accurately identify minute features of seeds, such as surface texture and color changes. In addition, combined with the results of physical property testing, the system can more comprehensively assess seed quality, thereby improving the accuracy and reliability of sorting.

3. Data integration and intelligent analysis: The data processing and analysis unit of the present invention conducts comprehensive analysis of image data and physical property data, and uses a fusion network to extract multi-level features to achieve more sophisticated seed evaluation. In addition, the system's intelligent analysis function also supports learning and adaptation, and can continuously optimize the detection and sorting process to meet the technical standards of GB/2930.8-2017 "Forage Seed Inspection Procedure Moisture Determination".

4. User interaction and network access: Through the touch screen display and network interface, the present invention provides an easy-to-use user interaction interface, allowing users to easily set parameters, monitor system status and access analysis results. The network function also supports remote access and control, increasing the flexibility and accessibility of the system.

The present invention effectively solves the problems existing in traditional seed detection and screening methods through automated, integrated and intelligent design, achieves significant technological progress, and provides a more efficient and accurate solution for seed pretreatment before sowing.

Third, the technical problems in the prior art solved by the present invention and the significant technical progress achieved are as follows:

Existing technical problems solved:

1. Lack of accuracy: Traditional forage seed quality testing can only be based on experience, based on the appearance of the seeds or simple physical properties, and the accuracy of the assessment of the intrinsic quality of the seeds is poor.

2. Inefficiency: Traditional forage seed screening often relies on manual operation, which leads to slow progress and low efficiency. Affected by human factors, it is easy to cause the quality of the screened seeds to fail to meet the standards.

3. Lack of automation and intelligence: Existing detection and screening equipment often lacks automation and intelligent functions, and cannot achieve fast, accurate and effective detection and screening of seeds.

Technological progress achieved:

1. Improved accuracy: Using image processing and machine learning algorithms to extract and classify features of seed images, combined with physical property data, it is possible to achieve a comprehensive assessment of seed quality and improve detection accuracy.

2. High efficiency of production: The present invention realizes rapid and accurate detection and screening of forage seeds by introducing high-resolution cameras, physical property detection units and automated screening units, greatly improving the production efficiency and speed of processing enterprises.

3. Automation and intelligence: Through the automated screening unit and the user interaction unit, automated screening and personalized operation of seeds are realized, which reduces the errors caused by manual operation and improves the level of intelligent production.

4. Scalability and flexibility: The design of the present invention allows adjustment and optimization of relevant parameters and models according to actual requirements to meet the needs of detection and screening of different types of forage seeds, and has strong scalability and flexibility.

The present invention solves the problems existing in the existing forage seed quality detection and screening methods by introducing advanced image processing, machine learning and automation technologies, and realizes efficient, accurate, automated and intelligent detection and screening of forage seeds, providing strong support for improving the seed quality of forage and promoting the high-quality development of animal husbandry.

BRIEF DESCRIPTION OF THE DRAWINGS

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required to be used in the embodiments of the present invention. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

FIG1 is an overall diagram of a device for intelligently monitoring and screening for forage seed quality provided by an embodiment of the present invention;

FIG2 is a partial diagram of an image acquisition unit provided in an embodiment of the present invention;

FIG3 is a partial diagram of a physical property detection unit provided in an embodiment of the present invention;

FIG4 is a partial diagram of an automated screening unit provided in an embodiment of the present invention;

FIG5 is a partial diagram of a user interaction interface provided by an embodiment of the present invention;

In the figure: 1. High-resolution camera; 2. LED lighting; 3. Vibration platform; 4. Fan; 5. Weight sensor; 6. CPU; 7. Storage unit; 8. Sorting robot arm; 9. Conveyor belt; 10. Touch screen display.

DETAILED DESCRIPTION

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

The following are two application examples of the intelligent detection and screening method for forage seed quality:

Application Example 1:

In this embodiment, the intelligent detection and screening method for forage seed quality uses a high-resolution camera and LED lighting for image acquisition. The seed images captured by the camera are subjected to feature extraction and classification by a pre-trained convolutional neural network model. The model can accurately identify the shape, size, color and other features of the seeds and output the corresponding recognition results.

Meanwhile, in the physical property detection step, a vibration platform and a fan are used to simulate the movement of seeds and perform wind screening. A weight sensor measures the weight of the seeds to provide data support for subsequent quality assessment.

In the data processing and analysis stage, the image recognition results and physical property data are combined to quantitatively evaluate the quality of seeds through an algorithm model based on statistical principles. The algorithm model takes into account multiple characteristic parameters of seeds and builds a prediction model based on historical data, which can accurately predict the quality grade of seeds.

Finally, based on the quality assessment results, the automated screening unit uses a sorting robot to screen the seeds, delivers qualified seeds to a designated location, and discards unqualified seeds. Users can view the screening results and data analysis reports on the touch screen display and adjust the screening parameters as needed.

Application Example 2:

In this embodiment, the forage seed quality intelligent detection and screening method combines deep learning and image processing technology. First, a high-resolution camera and LED lighting are used to collect high-quality images of seeds.

Then, the deep learning algorithm is used to learn and classify the features of the image. The convolutional neural network model is used to extract multi-level features of the image, and the model is trained to enable it to accurately identify various features of seeds.

For physical property inspection, a vibrating platform and fans are used to remove lightweight and substandard seeds, while a weight sensor measures the weight of the seeds.

In the data processing and analysis stage, the image features and physical property data are combined to comprehensively evaluate the seed quality through an integrated learning algorithm. The algorithm combines the outputs of multiple weak classifiers and improves the accuracy and stability of classification through voting or weighting.

Finally, based on the results of data processing and analysis, the automated screening unit uses a robotic arm and a conveyor belt to achieve automated screening and sorting of seeds. Users can view the screening results and data analysis reports through the user interaction unit and adjust relevant parameters according to actual requirements to meet the needs of quality testing and screening of different types of seeds.

These two embodiments demonstrate the specific application of the intelligent detection and screening method for forage seed quality in different technical implementations and application scenarios, reflecting the practicality and innovation of the present invention in the field of forage seed quality detection.

The following are two specific embodiments of the intelligent detection and screening system for forage seed quality:

Embodiment 1:

In a large forage grass planting base, a set of intelligent forage grass seed quality detection and screening system was applied. The system uses high-resolution cameras and LED lighting for image acquisition, ensuring that clear images of forage grass seeds can be obtained under various lighting conditions. In the physical property detection module, the coordinated work of the vibration platform and the fan effectively simulates the actual movement state of the seeds, and removes lightweight and unqualified seeds through wind screening.

After receiving the image and physical property data, the data processing and analysis module uses advanced image recognition algorithms and central processing units to accurately analyze and evaluate the seeds. Based on this data, the automated screening module automatically transports qualified seeds to the designated storage area through a sorting robot arm, while rejecting unqualified seeds.

The user interaction module displays the screening results and detailed data analysis reports to the operator through a touch screen display. The operator can adjust relevant parameters based on the information in the report to meet the detection and screening needs of different types of forage seeds.

Embodiment 2:

In order to improve the accuracy and work efficiency of seed quality detection in a forage seed production enterprise, an intelligent forage seed quality detection and screening system was introduced. The system uses a combination of high-resolution cameras and LED lighting to ensure that high-quality images can be collected in dark or strong light environments.

In the physical property detection module, the precise control of the vibration platform and fan enables the system to accurately simulate the vibration response and wind screening process of the seeds. The weight sensor accurately measures the weight of each seed, providing an important basis for quality assessment.

The data processing and analysis module uses powerful computing power and advanced algorithms to conduct in-depth analysis of the collected images and physical property data. By identifying characteristics such as seed morphology, size, color, and comprehensive evaluation of physical properties, the system can accurately determine the quality of seeds.

The automated screening module automatically sorts and screens the seeds based on the analysis results through the cooperation of the sorting robot arm and the conveyor belt. Qualified seeds are transported to the designated packaging area, while unqualified seeds are removed and collected for further processing.

The user interaction module provides a friendly operation interface and report display function, allowing operators to easily view screening results and data analysis reports, and adjust parameters according to actual requirements to meet the detection and screening needs of different batches or types of forage seeds.

These two embodiments demonstrate the different scenarios and uses of the intelligent detection and screening system for forage seed quality in practical applications. Through automation and intelligent means, the accuracy and production efficiency of seed quality detection are improved, providing strong support for accelerating the modernization of seed production and processing enterprises.

The image acquisition module for forage seed quality detection in the embodiment of the present invention includes at least one high-resolution camera, which can obtain clear forage seed images under various lighting conditions; at least one group of LED lighting lamps, which are designed with adjustable brightness and lighting angle to adapt to the lighting requirements of different environments and seed types, ensuring that the camera can capture high-quality images; and a control unit, which is used to adjust the camera settings and the lighting intensity of the LED lighting lamps to optimize the image acquisition process.

The physical property detection module of the forage seeds in the embodiment of the present invention includes a vibration platform with adjustable frequency, which is used to simulate motion states of different intensities and detect the stability and response of seeds in actual environments; a wind screening device, equipped with an adjustable speed fan and a direction control mechanism, to achieve accurate wind screening of seeds; a weight sensor group, which has a high-precision measurement function and can measure the weight of single or batch seeds. The data is automatically transmitted to the analysis unit, providing key data support for seed quality evaluation.

The data processing and analysis unit for forage seed quality detection in the embodiment of the present invention integrates a high-performance CPU and specialized image processing software, can efficiently process and analyze the collected image data, identify the size, shape, color and other characteristics of the seeds, and compare them with historical data; classify the seeds by quality level in combination with the physical property measurement results, using a preset algorithm model and scoring mechanism; provide a data interface to support data export and integration with other systems.

The user interaction integrated module for forage seed screening in the embodiment of the present invention integrates a high-precision sorting robot arm, has multi-axis control and fine motion simulation capabilities, and accurately sorts and places seeds according to analysis results; the conveyor belt system is designed with speed control and sorting area indication to ensure the efficiency and accuracy of seed sorting; the touch screen display provides an intuitive user interface to display real-time data analysis results, screening progress and historical records. Users can input operating instructions, set parameters and query results through the touch screen, thereby realizing the convenience and practicality of interaction.

In view of the problems existing in the prior art, the present invention provides a forage seed quality intelligent detection and screening device, which is described in detail below in conjunction with the accompanying drawings.

As shown in FIG1 , the forage seed quality intelligent detection and screening device provided by the embodiment of the present invention comprises:

Image acquisition unit, physical property detection unit, data processing and analysis unit, automatic screening unit, user interaction unit, the data analysis and processing unit is arranged at the upper left corner of the device, the image acquisition unit is arranged on the right side of the data analysis and processing unit, the user interaction unit is arranged below the data analysis and processing unit, the physical property detection unit is arranged in the middle of the device, and the automatic screening unit is arranged on the right side of the physical property detection unit.

As shown in FIG2 , the image acquisition unit specifically includes:

High-resolution camera 1 and LED lighting lamp 2. The high-resolution camera 1 is adsorbed on the left side wall of the device. The LED lighting lamp 2 is set on the high-resolution camera 1 and distributed around the lens of the high-resolution camera 1.

High-resolution camera1: At least one high-definition camera to capture high-quality images of seeds. The camera should have macro shooting capabilities to clearly record subtle features of seeds, such as surface texture, color changes, etc.

LED lighting 2: To ensure image quality, the system is equipped with a uniform LED lighting 2 to ensure that the seeds are evenly illuminated during the photo shooting process and reduce the impact of shadows and reflections.

As shown in FIG3 , the physical property detection unit specifically includes:

A vibration platform 3, a fan 4, and a weight sensor 5, wherein the weight sensor 5 is arranged at the bottom middle end of the device, the vibration platform 3 is arranged at the top of the weight sensor 5, and the fan 4 is arranged at the upper part of the weight sensor 5.

Vibrating platform 3: An electric vibrating platform 3 is used to separate light impurities and seeds by vibration.

Wind separation device: comprising a fan 4 and an adjustable air duct, which separates light impurities and seeds according to the adjustment of wind speed and wind direction.

Weight sensor 5: used to measure the weight of seeds and assist in determining seed quality.

The data processing and analysis unit specifically includes:

CPU6 and storage unit 7. The storage unit 7 is arranged at the upper left corner of the device, and the CPU6 is located at the lower right of the storage unit 7.

Central Processing Unit (CPU)/Graphics Processing Unit (GPU): Powerful computing units used to run image processing and machine learning algorithms to analyze image data and physical property data in real time.

Storage unit 7: A high-speed storage device for storing image data, physical property data and analysis results.

Analysis software: Software that integrates image processing and machine learning algorithms to automatically identify seeds in images and perform quality assessment.

As shown in Figure 4, the automated screening unit specifically includes:

A sorting robot arm 8 and a conveyor belt 9. The sorting robot arm 8 is located on the right side of the device. A conveyor belt 9 is provided at the bottom of the sorting robot arm 8 and is connected to the physical property detection unit.

Sorting robot arm 8: A precision-controlled robot arm that automatically picks up and sorts seeds based on the results of the data processing and analysis unit.

Conveyor belt 9: used to transport seeds to different collection areas and store them according to their quality.

As shown in FIG5 , the user interaction unit specifically includes:

The touch screen display 10 is arranged below the data processing and analysis unit.

Touch screen display 10: used to display system status, test results and operation menus. Users can perform operation settings and query results through the touch screen.

Operating system: An operating system used to manage hardware devices, run analysis software, and provide a user interface, supporting fast response and multitasking.

Network interface: supports remote access to the system via LAN or the Internet for parameter setting, status monitoring and viewing of data analysis results.

The detailed working principle of this automatic seed detection and sorting system is as follows:

1) Image acquisition unit

A high-resolution camera 1 is fixed to the left wall of the device to obtain high-definition images of seeds. The camera has a macro shooting function and can capture the tiny features of seeds, such as surface texture and color changes. LED lighting 2 is arranged around the lens of the camera 1 to provide uniform lighting, ensure image quality, and reduce the impact of shadows and reflections on image quality.

2) Physical property detection unit

The vibration platform 3 generates vibrations through the motor to separate light impurities from heavy seeds. The fan 4 generates airflow, and cooperates with the adjustable air duct to further separate light impurities from seeds according to the adjustment of wind speed and wind direction.

The weight sensor 5 is located below the vibration platform and is used to measure the weight of the seeds after separation to assist in determining the quality grade of the seeds.

3) Data processing and analysis unit

CPU6 and storage unit 7 together constitute the brain and memory of the system, where the CPU cooperates with GPU for high-speed computing and is responsible for processing image data and physical property data. Storage unit 7 is responsible for saving data and analysis results obtained from seeds. The analysis software runs on the CPU/GPU, uses image processing and machine learning algorithms to analyze data, automatically identifies seeds in images, and performs quality assessment based on their physical properties and image features.

4) Automated screening unit

The sorting robot arm 8 automatically selects and sorts the seeds according to the information provided by the data processing and analysis unit, and places them in different categories. The conveyor belt 9 transports the sorted seeds to a designated collection area or container for subsequent processing or packaging.

5) User interaction unit

The touch screen display 10 provides an interface for the user to display the system status, test results and operation menu. The operating system manages the hardware devices, runs the analysis software, and provides a user-friendly interactive interface.

The network interface allows users to remotely access the system through the LAN or the Internet to set parameters, monitor system status, and view analysis results.

The signal and data processing system provided by the present invention is a key part of the forage seed quality intelligent detection and screening method, and is responsible for processing the data from the image acquisition unit and the physical property detection unit, and performing corresponding analysis.

The main task of the signal and data processing system is to receive, process and analyze the data obtained from the image acquisition unit and the physical property detection unit. Through efficient data processing algorithms, the system can accurately identify and analyze the quality characteristics of seeds, providing a reliable basis for automated screening.

The high-definition image data acquired by the image acquisition unit is first transmitted to the signal and data processing system. The system uses the built-in image processing algorithm to pre-process the image, including denoising, contrast enhancement and other operations to improve image quality. Subsequently, the system extracts the characteristic information of the seed shape, size, color, etc. through image recognition technology.

At the same time, the vibration response, wind screening results and weight data provided by the physical property detection unit are also received by the system. After necessary calibration and standardization, these data are used to evaluate the physical properties of seeds.

After receiving the image and physical property data, the signal and data processing system uses advanced algorithms to conduct in-depth analysis of the data. By extracting and comparing image features, the system can determine the purity, quality and variety characteristics of the seeds. At the same time, combined with the physical property data, the system can further evaluate the vitality, growth potential and stress resistance of the seeds.

In addition, the system has the ability to learn and can continuously optimize the analysis algorithm based on historical data and user feedback to improve the accuracy of detection and the efficiency and speed of screening.

After analysis and evaluation, the signal and data processing system will generate screening results and data analysis reports. These results are displayed to operators through the user interaction unit for viewing and operation. Operators can adjust screening parameters or view detailed analysis reports as needed to better understand seed quality and make corresponding decisions.

At the same time, the system also supports data export function, users can export screening results and data analysis reports into Excel, PDF and other formats to facilitate further data processing and analysis.

The signal and data processing system is the core component of the intelligent detection and screening method for forage seed quality. It realizes rapid and accurate detection and screening of forage seed quality through efficient data processing and analysis technology.

The working principle of the forage seed quality intelligent detection and screening device provided by the embodiment of the present invention is as follows:

First, the image acquisition unit starts working. The high-resolution camera 1 captures high-quality images of forage seeds under the uniform illumination of the LED lighting 2. These images are not only high-definition, but also rich in information such as color and texture, providing a rich data source for subsequent analysis. The macro shooting function of the camera can clearly record the subtle features of the seeds, such as surface texture and color changes, which are crucial for subsequent seed quality assessment.

Next, the physical property detection unit starts to operate. The vibration platform 3 starts to vibrate, and the light impurities and seeds are separated by vibration. At the same time, the fan 4 further separates the light impurities and seeds according to the preset wind speed and wind direction. During this process, the weight sensor 5 measures the weight of the seeds, and these data will serve as an important basis for evaluating the quality of the seeds.

Then, the data processing and analysis unit starts working. Powerful computing units such as CPU6 and GPU start running, which will run image processing and machine learning algorithms to analyze image data and physical property data in real time. Storage unit 7 is responsible for storing these data and analysis results for subsequent use. The analysis software integrates image processing and machine learning algorithms, which can automatically identify seeds in the image and perform quality assessment based on the image and physical property data.

After evaluation by the data processing and analysis unit, the automated screening unit begins to work. According to the preset quality standards, the automated screening unit screens the seeds, removes the seeds that do not meet the quality requirements, and leaves only the seeds that meet the quality requirements.

Finally, the user interaction unit provides an interface for users to interact with the device. Through this interface, users can view information such as seed quality assessment results and screening results, and can also set parameters and control operations of the device.

The working principle of the entire device integrates complex processes in multiple fields such as modern image processing, computer vision, machine learning, and data analysis, realizing intelligent detection and screening of forage seed quality, greatly improving the accuracy of seed quality detection and the efficiency of screening.

The system realizes automatic detection and classification of seed quality by comprehensively using multiple modules such as image acquisition, physical property detection, data processing, automatic sorting and user interaction, which greatly improves the accuracy of detection and the efficiency of screening, and is suitable for large-scale seed processing and sorting operations.

The specific implementation examples of the present invention are as follows:

The intelligent detection and screening device for forage grass seed quality of the present invention is deployed in a large-scale forage grass seed processing plant, which needs to process tens of thousands of forage grass seeds every day to ensure that the quality of the seeds meets market demand.

First, workers pour the forage seeds to be tested onto the vibration platform of the device. The vibration platform starts working, simulating the actual movement of the seeds, and the fan starts to generate airflow to screen the seeds. In this process, light and unqualified seeds are quickly removed, while heavier qualified seeds continue to remain on the vibration platform.

At the same time, a high-resolution camera starts working to capture images of the seeds on the vibrating platform. LED lighting provides uniform lighting conditions to ensure clear images without shadows. The images captured by the camera are transmitted to the data processing and analysis unit in real time.

In the data processing and analysis unit, the pre-trained machine learning model begins to extract features and classify the images. The model uses deep learning algorithms to identify features such as seed shape, size, color, etc., and outputs the results to the analysis software. At the same time, the seed weight data measured by the weight sensor is also transmitted to the analysis software.

The analysis software combines image features and physical property data to quantitatively evaluate seed quality through an algorithm model. The model takes into account multiple parameters, such as seed size uniformity, color consistency, and weight distribution, and builds a prediction model based on historical data and statistical principles to accurately predict seed quality.

Based on the quality assessment results, the automated screening unit starts working. The sorting robot automatically screens the seeds according to the instructions, and transports the qualified seeds to the designated location through the conveyor belt for subsequent processing or packaging; the unqualified seeds are removed and collected in another container for subsequent processing or analysis.

During the whole process, users can view the screening results and data analysis reports in real time through the touch screen display. If they need to adjust the screening parameters or perform other operations, users can operate directly on the display to achieve personalized seed quality detection and screening.

Through the implementation of the present invention, the forage seed processing plant can greatly improve the production efficiency of seed detection and screening, reduce the errors caused by manual operation, improve the accuracy, and provide strong support for the modernization of seed production and processing enterprises.

The specific implementation process of the analysis software combining image features and physical property data to quantitatively evaluate seed quality is as follows:

1. Data Preprocessing

1. Image feature extraction: Using image processing technology, the shape, size, color and other features of seeds are extracted from the seed images taken by high-resolution cameras. These features can be obtained through edge detection, color segmentation, shape recognition and other algorithms.

2. Physical property data integration: The seed weight data measured by the weight sensor and the motion state data provided by the vibration platform and fan are integrated to form a complete physical property data set.

2. Feature Fusion

The image features and physical property data are combined to form a comprehensive feature vector. This process can be achieved through feature concatenation, feature selection or feature transformation, so that subsequent models can better use these features for quality assessment.

3. Model training and prediction

1. Model training: Based on historical data (including seed images and physical property data of known quality), an algorithm model is trained. This model can be a machine learning model (such as support vector machine, decision tree, etc.) or a deep learning model (such as convolutional neural network, recurrent neural network, etc.). During the training process, the model will learn how to identify patterns related to seed quality from the feature vector.

2. Model optimization: By adjusting the parameters and structure of the model, and using cross-validation, regularization and other techniques, the performance of the model is optimized so that it can more accurately predict the quality of seeds.

3. Quality prediction: For the newly input seed images and physical property data, after preprocessing and feature fusion, they are input into the trained model. The model will output a prediction value based on these features, which represents the quality grade or score of the seed.

4. Automated Screening

1. Instruction generation: Based on the quality assessment results predicted by the model, the analysis software will generate corresponding screening instructions. These instructions include which seeds need to be retained and which seeds need to be removed.

2. Robotic arm control: After receiving the command, the automated screening unit will control the sorting robot to work. The robot grabs and sorts the seeds according to the command through precise motion control.

3. Seed transportation: Qualified seeds are placed on the conveyor belt by the robot arm and transported to the designated location for subsequent processing or packaging. Unqualified seeds are placed in another container by the robot arm for subsequent processing or analysis.

During the whole process, the automated screening unit interacts with the analysis software in real time to ensure the accuracy and efficiency of the screening. At the same time, the system can also adjust and optimize parameters according to actual conditions to meet the detection and screening needs of different types of forage seeds.

Through the specific implementation of the above steps, the present invention can achieve accurate prediction and automated screening of forage grass seed quality, and improve the efficiency, speed and accuracy of seed quality detection and screening.

In the description of the present invention, unless otherwise specified, "plurality" means two or more than two; the orientations or positional relationships indicated by the terms "upper", "lower", "left", "right", "inner", "outer", "front end", "rear end", "head", "tail", etc. are based on the orientations or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operate in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first", "second", "third", etc. are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

The above description is only a specific implementation mode of the present invention, but the protection scope of the present invention is not limited thereto. Any modifications, equivalent substitutions and improvements made by any technician familiar with the technical field within the technical scope disclosed by the present invention and within the spirit and principle of the present invention should be covered within the protection scope of the present invention.

Claims (10)

1. The intelligent detecting and screening method for the quality of pasture seeds is characterized by comprising the following steps:

and (3) image acquisition: the high-resolution camera is utilized to collect images of pasture seeds, and meanwhile, the LED illuminating lamp is turned on to provide enough illumination conditions for the camera, so that the collected images are ensured to be clear;

And (3) physical characteristic detection: placing the collected pasture seeds on a vibration platform, simulating the actual motion state of the seeds through the vibration platform, and observing the vibration response of the seeds; simultaneously, air flow is generated by a fan, and wind power screening is carried out on seeds to remove light and unqualified seeds; finally, measuring the weight of the seeds by a weight sensor, wherein the weight sensor is used as an important parameter for judging the quality of the seeds;

Data processing and analysis: the collected image and physical characteristic data are transmitted to a data processing and analyzing unit, and the CPU is utilized to identify and analyze the image and judge the shape, size, color and other characteristics of the seeds; meanwhile, comprehensively evaluating the quality of seeds by combining physical characteristic data;

automated screening: according to the data processing and analysis results, the sorting mechanical arm is utilized to automatically screen the seeds, the seeds with qualified quality are conveyed to a designated position through the conveying belt, and meanwhile unqualified seeds are removed;

User interaction: displaying screening results and data analysis reports through a touch screen display for users to check and operate; the user can adjust the relevant parameters according to the actual requirements so as to adapt to the quality detection and screening requirements of different kinds of pasture seeds.

2. The intelligent detection and screening method for quality of grass seeds as defined in claim 1, wherein said data processing and analyzing step comprises:

Extracting and classifying the characteristics of the acquired images by utilizing a pre-trained machine learning model, and identifying the characteristics of the shape, the size, the color and the like of seeds; simultaneously, combining physical characteristic data, quantitatively evaluating the quality of seeds through an algorithm model, and generating a corresponding quality evaluation report; the machine learning model at least comprises a convolutional neural network model, and is used for extracting characteristic information of seeds from the images; the algorithm model is based on a statistical principle, combines the morphology, physical characteristics and historical data of the seeds, constructs a prediction model, and accurately predicts the quality of the seeds.

3. An intelligent detection and screening system for grass seed quality based on the method of claim 1, comprising:

the image acquisition module is used for acquiring images of pasture seeds by using the high-resolution camera and is provided with an LED illuminating lamp to provide enough illumination conditions so as to ensure the definition of image acquisition;

the physical characteristic detection module is provided with a vibration platform, a fan and a weight sensor for measuring the weight of seeds, wherein the vibration platform is used for simulating the actual motion state of the seeds to observe the vibration response of the seeds, and the fan is used for generating air flow to perform wind power screening to remove light and unqualified seeds;

The data processing and analyzing module is connected with the image acquisition module and the physical characteristic detection module and is used for receiving the image and the physical characteristic data, identifying and analyzing the image by utilizing a Central Processing Unit (CPU), judging the shape, the size, the color and other characteristics of the seeds, and comprehensively evaluating the quality of the seeds by combining the physical characteristic data;

the automatic screening module is used for automatically screening the seeds according to the result of the data processing and analyzing module, conveying the seeds with qualified quality to a designated position through a conveying belt, and rejecting unqualified seeds;

The user interaction module comprises a touch screen display and is used for displaying screening results and data analysis reports for viewing and operation of a user, and the user can adjust screening parameters through the module to realize personalized seed quality detection and screening;

The system realizes intelligent detection and screening of pasture seed quality by integrating functional modules such as image acquisition, physical characteristic detection, data processing and analysis, automatic screening, user interaction and the like, improves screening efficiency and quality, and reduces manual operation difficulty and cost.

4. A system as claimed in claim 3, comprising:

The image acquisition module for detecting the quality of the pasture seeds comprises at least one high-resolution camera, so that clear pasture seed images can be obtained under various illumination conditions;

At least one group of LED illuminating lamps with adjustable brightness and illumination angle are designed to adapt to the illumination requirements of different environments and seed types, so that the camera can capture high-quality images; the control unit is used for adjusting camera setting and illumination intensity of the LED illuminating lamp and optimizing an image acquisition process;

The physical characteristic detection module of the pasture seeds is adopted, and comprises a vibration platform with adjustable frequency, which is used for simulating motion states with different intensities so as to detect the stability and response of the seeds in an actual environment;

A wind power screening device provided with a speed-adjustable fan and a direction control mechanism to realize accurate wind power screening of seeds; the weight sensor group has a high-precision measurement function, can carry out weight measurement on single or batch seeds, automatically transmits data to the analysis unit, and provides key data support for seed quality evaluation.

5. The system of claim 2, wherein the data processing and analyzing unit for detecting the quality of the pasture seeds is used for integrating a high-performance CPU and special image processing software, so that the collected image data can be processed and analyzed efficiently, and the characteristics of the seeds such as size, shape, color and the like are identified and compared with the historical data; combining with a physical characteristic measurement result, classifying the quality level of the seeds by using a preset algorithm model and a scoring mechanism; a data interface is provided to support data export and integration with other systems.

6. The system as recited in claim 2, comprising:

The pasture seed screening and user interaction integrated module is integrated with a high-precision sorting mechanical arm, so that the pasture seed sorting mechanical arm has multi-axis control and fine action simulation capabilities, and seeds are accurately sorted and placed according to analysis results;

The conveying belt system is designed with speed control and sorting area indication, so that the high efficiency and accuracy of seed sorting are ensured; the touch screen display provides an intuitive user interface, shows real-time data analysis results, screens progress and historical records, and a user can input operation instructions, set parameters and inquire results through the touch screen, so that convenience and practicability of interaction are realized.

7. The intelligent forage grass seed quality detection and screening device based on the method of claim 1, wherein the image acquisition unit, the physical characteristic detection unit, the data processing and analysis unit, the automatic screening unit and the user interaction unit are arranged at the upper left corner of the device, the image acquisition unit is arranged at the right side of the data analysis and processing unit, the user interaction unit is arranged below the data analysis and processing unit, the physical characteristic detection unit is arranged in the middle of the device, and the automatic screening unit is arranged at the right side of the physical characteristic detection unit.

8. The intelligent forage grass seed quality detection and screening device as claimed in claim 7, wherein the image acquisition unit comprises a high-resolution camera and an LED illuminating lamp, the high-resolution camera is adsorbed on the left side wall of the device, the LED illuminating lamp is arranged on the high-resolution camera and distributed around the lens of the high-resolution camera; the physical characteristic detection unit comprises a vibration platform, a fan and a weight sensor, wherein the weight sensor is arranged at the bottom end of the middle part of the device, the vibration platform is arranged at the top of the weight sensor, and the fan is arranged at the upper part of the weight sensor; the data processing and analyzing unit comprises a CPU and a storage unit, wherein the storage unit is arranged at the upper left corner of the device, and the CPU is positioned at the lower right part of the storage unit.

9. The intelligent forage grass seed quality detecting and screening device according to claim 7, wherein the automatic screening unit comprises a sorting mechanical arm and a conveying belt, the sorting mechanical arm is positioned on the right side of the device, and the conveying belt is arranged at the bottom of the sorting mechanical arm and is connected with the physical characteristic detecting unit.

10. The intelligent detection and screening apparatus for quality of pasture seeds as defined in claim 7, wherein the user interaction unit comprises a touch screen display disposed below the data processing and analysis unit.