A visual inspection method for crops grown in agricultural greenhouses

Abstract

The present invention relates to the technical field of greenhouse planting management, and in particular to a visual detection method for crops planted in agricultural greenhouses. Based on the premise that image processing meets the standards, the present invention performs image growth status recognition feedback analysis on the analysis image to understand the growth status of each planted crop, and then performs rational management through feedback information to improve the growth health of the planted crops. In a progressive manner, regional planting disease hazard identification, assessment and analysis are performed to carry out targeted management of the disease details of the planted crops, and at the same time, local or overall regional management is carried out on the target area to reduce the growth risk of the planted crops in the target area. Based on the growth health risk assessment management analysis under normal analysis of the planted crops, the growth health status of each planted crop can be intuitively understood, so that the planted crops can be managed rationally and targeted to improve the growth health of the planted crops.

Description

Technical Field

[0001] The present invention relates to the technical field of greenhouse planting management, and in particular to a visual detection method for crops planted in an agricultural greenhouse. Background Art

[0002] Smart agricultural greenhouses are systems that use modern technology to automate vegetable planting and management. By integrating IoT technology, artificial intelligence, sensors, actuators, and other equipment, they enable real-time collection, transmission, processing, and control of greenhouse planting environment parameters. Compared to traditional vegetable greenhouses, unmanned vegetable greenhouses improve agricultural production efficiency through intelligent management and automated operations.

[0003] However, existing technologies cannot perform growth detection on crops grown in agricultural greenhouses based on visual detection technology, thereby reducing the growth safety and management efficiency of crops grown in agricultural greenhouses. Furthermore, existing technologies cannot analyze the quality of visual image processing, which causes image processing to interfere with subsequent analysis and affects the accuracy of analysis results. Furthermore, existing technologies cannot identify and provide feedback on the growth health and diseases of crops grown in agricultural greenhouses, which is detrimental to the targeted management of crops.

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

[0005] The purpose of the present invention is to provide a visual inspection method for crops planted in agricultural greenhouses to solve the technical defects mentioned above. The present invention performs preliminary detection interference analysis from the perspective of image processing. On the one hand, it helps to understand whether the visual feature image processing is qualified to reduce the interference of image processing on subsequent analysis. On the other hand, it helps to reasonably optimize and adjust the image processing scheme to improve image processing efficiency. Image growth status recognition feedback analysis is performed through information feedback to understand the growth status of each planted crop, and then management is performed through feedback information to improve the growth health of the planted crops. Regional planting disease hazard identification, assessment and analysis are performed in a progressive manner to carry out targeted management of the disease details of the planted crops, and at the same time, local or overall regional management is carried out on the target area to reduce the growth risk of the planted crops in the target area. Growth health risk assessment management analysis based on normal analysis of the planted crops can intuitively understand the growth health of each planted crop, so as to carry out reasonable and targeted management of the planted crops to improve the growth health of the planted crops.

[0006] The purpose of the present invention can be achieved by the following technical solution: A method for visually detecting crops grown in an agricultural greenhouse comprises the following steps:

[0007] Step 1: Collect visual feature images of each crop in the target area, conduct image processing accuracy assessment and analysis on the visual feature images, and perform discriminant processing on the obtained image processing coefficients. If a qualified signal is obtained, proceed to step 3; if a feedback signal is obtained, proceed to step 2;

[0008] Step 2: Based on the image processing efficiency evaluation operation under the feedback signal, the obtained image processing demand index is discriminated and processed. If a control signal is obtained, feedback is output;

[0009] Step 3: Perform image growth status recognition and feedback analysis through information feedback, perform discrimination processing on the obtained growth deviation value, and obtain a normal signal or an alarm signal;

[0010] Step 4: Based on the regional crop disease hazard identification and assessment analysis under the alarm signal, the disease name corresponding to the target disease image is output as feedback, and the regional alarm signal or local alarm signal is output as feedback;

[0011] Step 5: Based on the normal signal premise, the planting environment information of the planted crops is evaluated and analyzed for growth health risk, the obtained growth health risk index is discriminated and processed, and the obtained growth health score SP is output as feedback.

[0012] Preferably, the image processing accuracy evaluation and analysis process is as follows:

[0013] The visual inspection period of agricultural greenhouse crops is collected and set as a time threshold, the agricultural greenhouse crop planting area is set as a target area, the visual feature image of each planted crop in the target area within the time threshold is obtained, and the visual feature image is preprocessed, and the preprocessed visual feature image is set as the analysis image;

[0014] The analyzed image is divided into g sub-region blocks, where g is a natural number greater than zero. Image quality information of each sub-region block is obtained, including image clarity and image brightness values. The number of values ​​corresponding to the obtained image quality information that are lower than a preset threshold or deviate from a preset range is set as the image quality interference value. The image quality interference value is subjected to discrimination processing. If the image quality interference value is zero, the corresponding sub-region block is determined to be qualified. The ratio between the corresponding number of qualified sub-region blocks and the total number of sub-region blocks is set as the image processing coefficient;

[0015] The image processing coefficients are discriminated and processed to generate a feedback signal or a qualified signal.

[0016] Preferably, the image processing efficiency evaluation operation process is as follows:

[0017] The time period between the moments when the next image analysis generates a qualified signal is obtained and set as the processing tracking period. The number of feedback signals generated in the processing tracking period is obtained and set as the image processing requirement index. The image processing requirement index is discriminated and processed to obtain a control signal.

[0018] Preferably, the image growth state recognition feedback analysis process is as follows:

[0019] Acquire analysis images of each planted crop in the target area within the time threshold, identify the planted crops in the analysis image, determine the plant name of the planted crops in the analysis image, obtain growth information of the planted crops in the analysis image, the growth information includes the planting date and growth cycle, obtain the time from the planting date of the planted crops in the analysis image to the current time, and set it as the extraction time;

[0020] Obtain the standard feature image of crops planted with the same extraction time, compare and analyze the analysis image with the standard feature image, set the difference value between the analysis image and the standard feature image as the growth deviation value, and perform discrimination processing on the growth deviation value to obtain a normal signal or an alarm signal.

[0021] Preferably, the process of identifying, evaluating and analyzing regional crop disease hazards is as follows:

[0022] Obtain a disease feature image set of crops planted in the analysis image within the time threshold, and perform one-to-one comparison and analysis between the analysis image and the disease feature image set, set the similarity between the analysis image and the disease feature image set as the disease identification value, obtain the maximum value among the disease identification values, and set the disease feature image corresponding to the maximum value among the disease identification values ​​as the target disease image, and extract the disease name corresponding to the target disease image.

[0023] Preferably, the total number of planted crops in the target area within the time threshold is obtained, and the number of planted crops corresponding to the alarm signal in the target area within the time threshold is also obtained. The ratio between the number of planted crops corresponding to the alarm signal in the target area and the total number is set as the regional planting risk ratio, and the regional planting risk ratio is discriminated and processed to generate a regional alarm signal or a local alarm signal.

[0024] Preferably, the growth health risk assessment management analysis process is as follows:

[0025] The planting environment information of the crops planted in the analysis image within the time threshold is obtained, and the planting environment information includes the planting obstacle value and the supply interference value. The number of planting obstacle values ​​and supply interference values ​​greater than or equal to the preset planting obstacle value threshold and the preset supply interference value threshold is obtained, and it is set as the growth health risk index. The growth health risk index is discriminated and processed, and the first-level health state, the second-level health state and the third-level health state are set as the growth health score SP, SP=1, 2, 3.

[0026] Preferably, the analysis process of the planting obstacle value is as follows: based on soil remote sensing technology, the electromagnetic spectrum signals reflected or emitted by the soil of the planted crops are collected at a long distance and processed into images that can be directly recognized, thereby obtaining soil data of the soil of the planted crops. The soil data includes moisture and humidity, and the number of parts of the corresponding values ​​of the soil data that deviate from the preset threshold value is greater than the preset critical value, and it is set as the planting obstacle value; the analysis process of the health interference value is as follows: based on infrared imaging technology, the water content and chlorophyll content of the vegetable leaves are detected, and the number of parts whose water content and chlorophyll content are lower than the preset threshold value is obtained, and it is set as the health interference value.

[0027] The beneficial effects of the present invention are as follows:

[0028] (1) The present invention conducts a preliminary detection interference analysis from the perspective of image processing. On the one hand, it helps to understand whether the visual feature image processing is qualified, so as to reduce the interference of image processing on subsequent analysis. On the other hand, it helps to reasonably optimize and adjust the image processing scheme to improve the image processing efficiency. The image growth status recognition feedback analysis is carried out through information feedback to understand the growth status of each planted crop, and then management is carried out through feedback information to improve the growth health of the planted crops.

[0029] (2) The present invention uses a progressive approach to identify, assess, and analyze regional crop disease hazards so as to conduct targeted management based on the details of crop disease, and simultaneously conduct local or overall regional management of the target area, thereby reducing the growth risk of crops in the target area. Based on the analysis of the growth health risk assessment and management of normal crop analysis, the growth health status of each crop can be intuitively understood, so that reasonable and targeted management of the crops can be carried out to improve the growth health of the crops. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] The present invention will be further described below with reference to the accompanying drawings;

[0031] Figure 1 It is a reference analysis diagram of the method of the present invention;

[0032] Figure 2 It is a reference diagram for local analysis of the present invention. DETAILED DESCRIPTION

[0033] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0034] Example 1:

[0035] See also Figures 1 to 2 As shown, the present invention is a visual inspection method for crops grown in agricultural greenhouses, comprising the following steps:

[0036] Step 1: Collect visual feature images of each crop in the target area, conduct image processing accuracy assessment and analysis on the visual feature images, and perform discriminant processing on the obtained image processing coefficients. If a qualified signal is obtained, proceed to step 3; if a feedback signal is obtained, proceed to step 2;

[0037] Step 2: Based on the image processing efficiency evaluation operation under the feedback signal, the obtained image processing demand index is discriminated and processed. If a control signal is obtained, feedback is output;

[0038] Step 3: Perform image growth status recognition and feedback analysis through information feedback, perform discrimination processing on the obtained growth deviation value, and obtain a normal signal or an alarm signal;

[0039] Step 4: Based on the regional crop disease hazard identification and assessment analysis under the alarm signal, the disease name corresponding to the target disease image is output as feedback, and the regional alarm signal or local alarm signal is output as feedback;

[0040] Step 5: Based on the normal signal premise, the planting environment information of the planted crops is evaluated and analyzed for growth health risk, the obtained growth health risk index is discriminated and processed, and the obtained growth health score SP is output as feedback;

[0041] Safety inspection of agricultural greenhouse crops is carried out from the perspective of image processing. That is, by collecting visual feature images of each crop in the target area and performing image processing accuracy evaluation and analysis on the visual feature images, the accuracy of the visual feature image processing is determined to determine whether the visual feature image processing is qualified. This also helps to optimize and adjust the image processing scheme to improve image processing efficiency. The specific image processing accuracy evaluation and analysis process is as follows:

[0042] The visual inspection period of agricultural greenhouse crops is collected and set as a time threshold, the agricultural greenhouse crop area is set as a target area, and visual feature images of each crop in the target area within the time threshold are obtained. The visual feature images are preprocessed, including cropping and scaling, to obtain the preprocessed visual feature images and set them as analysis images.

[0043] The analysis image is divided into g sub-region blocks, where g is a natural number greater than zero. The image quality information of each sub-region block is obtained, and the image quality information includes picture clarity, picture brightness value, etc. The number of values ​​corresponding to the obtained image quality information that are lower than a preset threshold or deviate from a preset range is set as the image quality interference value. The image quality interference value is subjected to discrimination processing. If the image quality interference value is equal to zero, the corresponding sub-region block is determined to be qualified. The ratio between the corresponding number of qualified sub-region blocks and the total number of sub-region blocks is obtained and set as the image processing coefficient. It should be noted that the larger the value of the image processing coefficient, the greater the risk of abnormal processing of the analysis image;

[0044] Perform discriminant processing on the image processing coefficients:

[0045] If the image processing coefficient is less than or equal to a preset threshold image processing coefficient threshold, a feedback signal is generated, and the analysis image is preprocessed again;

[0046] If the image processing coefficient is greater than a preset threshold image processing coefficient threshold, a qualified signal is generated;

[0047] When the feedback signal is generated, the image processing efficiency evaluation operation is performed. The specific image processing efficiency evaluation operation process is as follows:

[0048] The time period between the times when the next image analysis generates qualified signals is obtained and set as the processing tracking period. The number of feedback signals generated during the processing tracking period is obtained and set as the image processing requirement index. The image processing requirement index is then discriminated:

[0049] If the image processing demand index is less than the preset image processing demand index threshold, no signal is generated;

[0050] If the image processing demand index is greater than or equal to the preset image processing demand index threshold, a control signal is generated, and the qualified signal or the control signal is sent to the adjustment processing unit. After receiving the qualified signal or the control signal, the adjustment processing unit immediately performs the preset warning operation corresponding to the qualified signal or the control signal, so as to reasonably optimize and adjust the image processing solution to improve the image processing efficiency;

[0051] Once a qualified signal is generated, image growth status recognition and feedback analysis is performed through information feedback to reduce the planting risk of each crop and improve the growth health of the crop. The specific image growth status recognition and feedback analysis process is as follows:

[0052] Acquire analysis images of each planted crop in the target area within the time threshold, identify the planted crops in the analysis image, determine the plant names of the crops in the analysis image, which include Chinese cabbage, lettuce, corn, etc., obtain growth information of the crops in the analysis image, which includes planting date, growth cycle, etc., obtain the time duration from the planting date of the crops in the analysis image to the current moment, and set it as the extraction duration. It should be noted that the analysis is performed from the perspective of the extraction duration in order to determine the growth status of the crops, thereby improving the accuracy of the analysis results;

[0053] Obtain the standard feature image of crops grown for the same extraction time, compare and analyze the analysis image with the standard feature image, obtain the difference value between the analysis image and the standard feature image, set the difference value between the analysis image and the standard feature image as the growth deviation value, and perform discrimination processing on the growth deviation value:

[0054] If the growth deviation value falls within the preset growth deviation value range, a normal signal is generated;

[0055] If the growth deviation value does not fall within the preset growth deviation value threshold, an alarm signal is generated, and the normal signal or alarm signal is sent to the adjustment processing unit. After receiving the normal signal or alarm signal, the adjustment processing unit immediately performs the preset early warning operation corresponding to the normal signal or alarm signal, so as to manage the planted crops corresponding to the alarm signal, reduce the planting risks of each planted crop, and improve the growth health of the planted crops.

[0056] Example 2:

[0057] When an alarm signal is generated, the regional crop disease hazard identification, assessment and analysis based on the alarm signal is performed. The analysis image is compared with the disease feature image set to enable targeted management of the crop disease details. The specific regional crop disease hazard identification, assessment and analysis process is as follows:

[0058] Acquire a disease feature image set of the planted crops in the analysis image within the time threshold, perform a one-to-one comparison analysis on the analysis image and the disease feature image set, obtain a similarity between the analysis image and the disease feature image set, set the similarity between the analysis image and the disease feature image set as a disease identification value, obtain a maximum value among the disease identification values, set the disease feature image corresponding to the maximum value among the disease identification values ​​as a target disease image, extract a disease name corresponding to the target disease image, and send the disease name corresponding to the target disease image to an adjustment processing unit, which immediately displays the number of the corresponding planted crop and the corresponding disease name, so as to manage the planted crops in a targeted manner;

[0059] In the embodiment of the present invention, for example, if the plant is Chinese cabbage, a large number of cabbage images with five common diseases, including downy mildew images, soft rot images, bacterial angular spot images, viral disease images, and damping-off images, are collected to form a disease feature image set; for example, if the plant is potato, a large number of potato leaf images with five common diseases, including potato blight images, ring rot images, potato yellow leaf curl images, potato wilt images, and potato anthracnose images, are collected to form a disease feature image set;

[0060] The total number of crops planted in the target area within the time threshold is obtained, and the number of crops planted corresponding to the alarm signals in the target area within the time threshold is obtained. The ratio between the number of crops planted corresponding to the alarm signals in the target area and the total number is set as the regional planting risk ratio, and the regional planting risk ratio is discriminated:

[0061] If the regional planting risk ratio is greater than or equal to the preset regional planting risk ratio threshold, a regional alarm signal is generated;

[0062] If the regional planting risk ratio is less than a preset regional planting risk ratio threshold, a local alarm signal is generated, and the regional alarm signal or the local alarm signal is sent to the adjustment processing unit. After receiving the regional alarm signal or the local alarm signal, the adjustment processing unit immediately performs a preset early warning operation corresponding to the regional alarm signal or the local alarm signal, so as to perform local management or overall regional management on the target area, thereby reducing the growth risk of crops planted in the target area;

[0063] When a normal signal is generated, the planting environment information of the crops in the analysis image is collected and analyzed, and the growth health risk assessment management analysis of the planting environment information of the crops is performed to intuitively understand the growth health of each crop, so as to carry out reasonable and targeted management of the crops to improve the growth health of the crops. The specific growth health risk assessment management analysis process is as follows:

[0064] The planting environment information of the crops in the analyzed image within the time threshold is obtained. The planting environment information includes the planting obstacle value and the supply interference value. The number of planting obstacle values ​​and supply interference values ​​greater than or equal to the preset planting obstacle value threshold and the preset supply interference value threshold is obtained, and it is set as the growth health risk index, and the growth health risk index is discriminated:

[0065] If the growth health risk index = zero, it is judged to be a level one health state;

[0066] If the growth health risk index = 1, it is determined to be a secondary health state;

[0067] If the growth health risk index = 2, it is determined to be a level three health state, wherein the growth health risks corresponding to the level one health state, the level two health state, and the level three health state increase in sequence, and the level one health state, the level two health state, and the level three health state are set as the growth health score SP, SP = 1, 2, 3, that is, when the growth health score SP = 1, it indicates a level one health state, when the growth health score SP = 2, it indicates a level two health state, and when the growth health score SP = 3, it indicates a level three health state. The growth health score SP is sent to the adjustment processing unit. After receiving the growth health score SP, the adjustment processing unit immediately displays the preset warning text corresponding to the growth health score SP, so as to intuitively understand the growth health status of each planted crop, so as to carry out reasonable and targeted management of the planted crops and improve the growth health of the planted crops;

[0068] In an embodiment of the present invention, the analysis process of the planting obstacle value is as follows: based on soil remote sensing technology, electromagnetic spectrum signals reflected or emitted by the soil of the planted crops are collected from a distance and processed into a directly recognizable image, thereby obtaining soil data of the planted crops. The soil data includes moisture, humidity, etc. The number of portions of the soil data corresponding to the values ​​that deviate from a preset threshold value and exceed a preset critical value is obtained, and this number is set as the planting obstacle value. It should be noted that the larger the value of the planting obstacle value, the greater the risk of abnormal growth of the planted plants;

[0069] In the embodiment of the present invention, the analysis process of the health interference value is as follows: the water content and chlorophyll content of the vegetable leaves are detected based on infrared imaging technology, the number of leaves with water content and chlorophyll content below a preset threshold is obtained, and the number is set as the health interference value. It should be noted that the larger the value of the health interference value, the greater the risk to the healthy growth of the planted crops;

[0070] In summary, the present invention conducts a preliminary detection interference analysis from the perspective of image processing. On the one hand, it helps to understand whether the visual feature image processing is qualified to reduce the interference of image processing on subsequent analysis. On the other hand, it helps to reasonably optimize and adjust the image processing scheme to improve image processing efficiency. Image growth status recognition feedback analysis is performed through information feedback to understand the growth status of each planted crop, and then management is performed through feedback information to improve the growth health of the planted crops. Regional plant disease hazard identification, assessment and analysis are performed in a progressive manner to carry out targeted management of the disease details of the planted crops, and at the same time, local or overall regional management is carried out on the target area to reduce the growth risk of the planted crops in the target area. Growth health risk assessment management analysis based on normal analysis of the planted crops can intuitively understand the growth health of each planted crop, so as to carry out reasonable and targeted management of the planted crops to improve the growth health of the planted crops.

[0071] The threshold is set to facilitate comparison. The size of the threshold depends on the amount of sample data and the number of bases set by technicians in this field for each set of sample data; as long as it does not affect the proportional relationship between the parameter and the quantized value.

[0072] The above formulas are obtained by collecting a large amount of data and performing software simulation, and a formula close to the actual value is selected. The coefficients in the formula are set by those skilled in the art according to actual conditions. The above is only a preferred specific implementation method of the present invention, but the protection scope of the present invention is not limited to this. Any technician familiar with this technical field, within the technical scope disclosed by the present invention, can make equivalent replacements or changes based on the technical solution and inventive concept of the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A visual inspection method for crops grown in agricultural greenhouses, characterized in that: The following steps are involved: Step 1: Collect visual feature images of each crop in the target area, conduct image processing accuracy assessment and analysis on the visual feature images, and perform discriminant processing on the obtained image processing coefficients. If a qualified signal is obtained, proceed to step 3; if a feedback signal is obtained, proceed to step 2; Step 2: Based on the image processing efficiency evaluation operation under the feedback signal, the obtained image processing demand index is discriminated and processed. If a control signal is obtained, feedback is output; Step 3: Perform image growth status recognition and feedback analysis through information feedback, perform discrimination processing on the obtained growth deviation value, and obtain a normal signal or an alarm signal; Step 4: Based on the regional crop disease hazard identification and assessment analysis under the alarm signal, the disease name corresponding to the target disease image is output as feedback, and the regional alarm signal or local alarm signal is output as feedback; Step 5: Based on the normal signal premise, the planting environment information of the planted crops is evaluated and analyzed for growth health risk, the obtained growth health risk index is discriminated and processed, and the obtained growth health score SP is output as feedback; The image processing accuracy evaluation and analysis process is as follows: The visual inspection period of agricultural greenhouse crops is collected and set as a time threshold, the agricultural greenhouse crop planting area is set as a target area, the visual feature image of each planted crop in the target area within the time threshold is obtained, and the visual feature image is preprocessed, and the preprocessed visual feature image is set as the analysis image; The analyzed image is divided into g sub-region blocks, where g is a natural number greater than zero. Image quality information of each sub-region block is obtained, including image clarity and image brightness values. The number of values ​​corresponding to the obtained image quality information that are lower than a preset threshold or deviate from a preset range is set as the image quality interference value. The image quality interference value is subjected to discrimination processing. If the image quality interference value is zero, the corresponding sub-region block is determined to be qualified. The ratio between the corresponding number of qualified sub-region blocks and the total number of sub-region blocks is set as the image processing coefficient; Performing discrimination processing on the image processing coefficients to generate a feedback signal or a qualified signal; The image processing efficiency evaluation operation process is as follows: The time period between the moments when the next image analysis generates a qualified signal is obtained and set as the processing tracking period. The number of feedback signals generated in the processing tracking period is obtained and set as the image processing requirement index. The image processing requirement index is discriminated and processed to obtain a control signal.

2. A visual inspection method for agricultural greenhouse crops according to claim 1, characterized in that: The image growth state recognition feedback analysis process is as follows: Acquire analysis images of each planted crop in the target area within the time threshold, identify the planted crops in the analysis image, determine the plant name of the planted crops in the analysis image, obtain growth information of the planted crops in the analysis image, the growth information includes the planting date and growth cycle, obtain the time from the planting date of the planted crops in the analysis image to the current time, and set it as the extraction time; Obtain the standard feature image of crops planted with the same extraction time, compare and analyze the analysis image with the standard feature image, set the difference value between the analysis image and the standard feature image as the growth deviation value, and perform discrimination processing on the growth deviation value to obtain a normal signal or an alarm signal.

3. The visual inspection method for agricultural greenhouse crops according to claim 1, characterized in that: The process of identifying, evaluating and analyzing the damage caused by crop diseases in the region is as follows: Obtain a disease feature image set of crops planted in the analysis image within the time threshold, and perform one-to-one comparison and analysis between the analysis image and the disease feature image set, set the similarity between the analysis image and the disease feature image set as the disease identification value, obtain the maximum value among the disease identification values, and set the disease feature image corresponding to the maximum value among the disease identification values ​​as the target disease image, and extract the disease name corresponding to the target disease image.

4. A visual inspection method for agricultural greenhouse crops according to claim 3, characterized in that: The total number of crops planted in the target area within the time threshold is obtained, and the number of crops planted corresponding to the alarm signal in the target area within the time threshold is obtained at the same time. The ratio between the number of crops planted corresponding to the alarm signal in the target area and the total number is set as the regional planting risk ratio, and the regional planting risk ratio is discriminated and processed to generate a regional alarm signal or a local alarm signal.

5. The visual inspection method for agricultural greenhouse crops according to claim 1, characterized in that: The growth health risk assessment management analysis process is as follows: The planting environment information of the crops planted in the analysis image within the time threshold is obtained, and the planting environment information includes the planting obstacle value and the supply interference value. The number of planting obstacle values ​​and supply interference values ​​greater than or equal to the preset planting obstacle value threshold and the preset supply interference value threshold is obtained, and it is set as the growth health risk index. The growth health risk index is discriminated and processed, and the first-level health state, the second-level health state and the third-level health state are set as the growth health score SP, SP=1, 2, 3.

6. A visual inspection method for agricultural greenhouse crops according to claim 5, characterized in that: The analysis process of the planting obstacle value is as follows: based on soil remote sensing technology, the electromagnetic spectrum signals reflected or emitted by the soil of the planted crops are collected from a long distance and processed into images that can be directly recognized, thereby obtaining soil data of the soil of the planted crops. The soil data includes moisture and humidity. The number of parts of the corresponding values ​​of the soil data that deviate from the preset threshold value is greater than the preset critical value, and it is set as the planting obstacle value; the analysis process of the health interference value is as follows: based on infrared imaging technology, the water content and chlorophyll content of the vegetable leaves are detected, and the number of parts whose water content and chlorophyll content are lower than the preset threshold value is obtained, and it is set as the health interference value.

Images