CN111066442B - Targeted variable fertilization method and device for corn and application - Google Patents
Targeted variable fertilization method and device for corn and application Download PDFInfo
- Publication number
- CN111066442B CN111066442B CN201911229580.0A CN201911229580A CN111066442B CN 111066442 B CN111066442 B CN 111066442B CN 201911229580 A CN201911229580 A CN 201911229580A CN 111066442 B CN111066442 B CN 111066442B
- Authority
- CN
- China
- Prior art keywords
- corn
- fertilizer spraying
- image information
- image acquisition
- fertilizer
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- A—HUMAN NECESSITIES
- A01—AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
- A01C—PLANTING; SOWING; FERTILISING
- A01C23/00—Distributing devices specially adapted for liquid manure or other fertilising liquid, including ammonia, e.g. transport tanks or sprinkling wagons
- A01C23/04—Distributing under pressure; Distributing mud; Adaptation of watering systems for fertilising-liquids
- A01C23/047—Spraying of liquid fertilisers
-
- A—HUMAN NECESSITIES
- A01—AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
- A01C—PLANTING; SOWING; FERTILISING
- A01C23/00—Distributing devices specially adapted for liquid manure or other fertilising liquid, including ammonia, e.g. transport tanks or sprinkling wagons
- A01C23/007—Metering or regulating systems
Abstract
The invention relates to the technical field of agricultural fertilization, and particularly discloses a targeted variable fertilization method and device for corn and application. The device includes: the device comprises an image acquisition device, a control device and a fertilizer spraying device; the image acquisition device acquires image information of an area to be fertilized in real time when the fertilization device advances, and the control device respectively controls the fertilization device according to the image information; the fertilizer spraying device comprises a flat fan-shaped nozzle, the spraying shape of the flat fan-shaped nozzle is fan-shaped, and the plane of the fan-shaped nozzle is vertical to the advancing direction of the fertilizer spraying device; the image acquisition device and the fertilizer spraying device are arranged on the same horizontal plane, and the vertical projection of the sector on the plane of the area to be fertilized is positioned in the middle position of the visual field distance of the image acquisition device parallel to the advancing direction; the image acquisition device and the fertilizer spraying device responsible for fertilizing in the field of view area of the image acquisition device are arranged in a specific manner. The invention also provides a fertilizing method and application applying the device. The invention has the advantages of precise fertilization, improved fertilizer utilization rate and reduced fertilizer pollution.
Description
Technical Field
The invention relates to the technical field of agricultural fertilization, in particular to a targeted variable fertilization method and device for corn and application.
Background
The variable fertilization technology is one of the core contents of precision agriculture, and finely and accurately adjusts the fertilization amount of crops according to the space difference of the crop growth environment and the fertilizer requirement rule of the crops so as to obtain the highest economic benefit and ecological benefit with the least fertilizer investment. The current variable fertilization technology mainly comprises two types: a variable fertilization technology based on a prescription chart and an online real-time variable fertilization technology based on a sensor.
The variable fertilization based on the prescription chart mainly comprises two links of generation of a fertilization prescription chart and variable fertilization operation. The generation of the prescription chart refers to that firstly, the nutrient contents of different areas in the field are measured by a soil sampling and chemical analysis method. And then, combining an agricultural geographic information technology to generate a soil nutrient spatial variation graph to guide variable fertilization operation. However, the method has the main defects that the soil testing formula is high in cost and long in period, and the variability of the field nutrition distribution cannot be reflected in real time.
With the development of spectral analysis technology and information technology, the real-time detection of the nutrient distribution of the farmland through a spectral sensor is realized, and the hardware cost is lower and lower. This also promotes the development of on-line real-time variable fertilization techniques. Aiming at the defects of the variable fertilization method based on the prescription chart, the sensor-based online real-time variable fertilization technology is developed and researched. Firstly, acquiring spectral information of field crops and soil through a spectral sensor, and obtaining variability of field nutrition distribution by combining a spectral analysis technology; then the system controller processes the spectrum data, and a fertilization decision is made by combining the fertilization decision model and a control instruction is sent to a fertilization mechanism; and finally, the fertilizing mechanism is controlled by the fertilizing mechanism according to the control instruction to perform variable rate fertilizing operation. For example, chinese patent CN201510163619.9 proposes a method for diagnosing growth of leaf vegetables by a spectrum sensor and performing variable fertilization according to the diagnosis result.
The online real-time variable fertilization method based on the spectral sensor can diagnose the crop nutrition state in a certain area in advance through the spectral sensor, determine the growth requirement of crops through the nutrition state, and control the fertilization mechanism to perform variable fertilization operation according to the crop requirement. However, as shown in fig. 1, the operation method is still a continuous operation method, and does not distinguish the plant type in the area and whether the plant type is blank land, and although the fertilizer input can be reduced to a certain extent compared with the conventional uniform fertilization method without nutrition judgment, the application position of the fertilizer cannot be accurately ensured, and the targeted fertilization cannot be performed according to the real-time characteristics of each plant.
Chinese patent CN201410037916.4 is directed to the application of orchard fertilization, and provides a device and a method for targeted variable fertilization of fruit tree roots in order to ensure that the fruit tree roots can effectively absorb fertilizer nutrients. The device and the method predict the distribution position of the root system by detecting the three-dimensional structure of the branches of the fruit tree, thereby carrying out targeted fertilization on the roots. However, compared with the planting mode of fruit trees, the plant form of wide-row crops (such as corn) in the field is small, the distribution is denser, and the mode cannot be applied to top dressing of the crown of the crops such as corn.
Therefore, it is necessary to provide a new method, device and application for targeted variable fertilization of corn to solve the above problems.
Disclosure of Invention
In order to solve the problems in the prior art, the invention aims to provide a corn targeting variable fertilizer applying device which can reduce the waste of fertilizer in a non-fertilizing area, improve the utilization rate of fertilizer resources and avoid omission of fertilizing objects.
In order to achieve the purpose of the invention, the technical scheme of the invention is as follows:
a corn targeted variable fertilizing device comprises: the device comprises an image acquisition device, a control device and a fertilizer spraying device;
the image acquisition device acquires image information of an area to be fertilized in real time when the fertilization device moves, and the control device respectively controls the fertilization device according to the image information;
the fertilizer spraying device comprises a flat fan-shaped nozzle, the spraying shape of the flat fan-shaped nozzle is fan-shaped, the spraying angle of the flat fan-shaped nozzle is theta, and the vertical projection of the plane where the fan-shaped nozzle is located on the plane of the area to be fertilized is perpendicular to the advancing direction of the fertilizer spraying device;
the image acquisition device and the fertilizer spraying device are arranged on the same horizontal plane, and the vertical projection of the fan shape on the plane of the area to be fertilized is positioned in the middle position of the field distance of the image acquisition device parallel to the advancing direction;
the arrangement of the image acquisition device and the fertilizer spraying device responsible for fertilizing in the field of view area of the image acquisition device satisfies the following formulas (1) to (5):
d1≈2×x1 (3)
d2≈n×x2 (4)
wherein d is1Distance of field of view parallel to the direction of travel for the image acquisition device, d2A field distance, x, perpendicular to the direction of travel for the image acquisition device1For planting corn at a plant spacing of x2For planting row spacing of corn, theta1Angle of view, θ, of said image capturing device parallel to said direction of travel2The field angle of the image acquisition device perpendicular to the advancing direction, h is the height of the image acquisition device and the fertilizer spraying device from the ground, n is the number of the fertilizer spraying devices (n is a positive integer, preferably n is 2-6), and theta are1、θ2In degrees, h in centimeters, d1、d2、x1、x2In centimeters.
In order to facilitate accurate positioning of the spraying position, the invention uses flat fan nozzles and ensures that the fan plane is perpendicular to the travelling direction of the fertilizing device when mounted.
The fertilizing device comprises at least one image acquisition device and at least one fertilizer spraying device, the number of the fertilizer spraying devices is determined according to actual fertilizing requirements (each fertilizer spraying device can correspond to a row of corn plants), the number of the image acquisition devices is determined according to the relation between the spraying angle of each fertilizer spraying device and the field angle of each image acquisition device, and therefore the image information of the area to be fertilized corresponding to the fan-shaped spraying coverage area of each fertilizer spraying device is preferably acquired by the image acquisition devices. For example, when the field angle of the image acquisition devices is 2 times of the spraying angle of the fertilizer spraying devices, that is, each image acquisition device can simultaneously collect the image information of the areas to be fertilized corresponding to two fertilizer spraying devices, at this time, the number of the fertilizer spraying devices on the fertilizer spraying devices can be 2 times of that of the image acquisition devices.
According to the fertilizing device, the specific relative position relationship between the image acquisition device and the fertilizer spraying device is set, so that the image acquisition device can scan each target in the advancing direction of the fertilizing mechanism, the full coverage of the target in the area to be fertilized is ensured, the spraying surface of the fertilizer spraying device can cover the detection area of the image acquisition device, and the overlapping is reduced as much as possible. Thereby realized spraying the accurate matching of position and target position and not having missed the target, improved fertilizer resource utilization.
In the fertilizing device, the image information comprises color image information, spectral image information and depth image information; and/or the image acquisition device is an RGB-D camera.
In the fertilizing device, the control device comprises a corn detection and positioning module, the corn detection and positioning module detects whether corn exists in the area to be fertilized according to the color image information, and when the corn is detected, the corn is positioned.
The invention can train a target detection network (such as Yolov3) through a deep learning image processing technology. The trained network is then deployed to the control device for real-time detection. The corn plant target real-time detection positioning method based on deep learning can accurately position the target in real time, and provides a foundation for guaranteeing the accuracy of fertilization.
In the fertilizing device, the control device further comprises a nutrition demand diagnosis module and a fertilizer spraying device control module, the nutrition demand diagnosis module predicts the nutrition demand of the corn according to the spectral image information and the depth image information of the corn, and the fertilizer spraying device control module controls the fertilizer spraying amount of the fertilizer spraying device according to the nutrition demand.
Preferably, each fertilizer spraying device is independently controlled by a switch electromagnetic valve.
The invention can control the fertilizing amount in a targeted way according to the real-time characteristics of each crop, is beneficial to the growth of the crop and improves the utilization rate of the fertilizer.
In the fertilizer applying device of the present invention, the control device further includes a response distance determining module and a response position determining module, the response distance determining module determines the response distance of the fertilizer applying device according to the following equations (6) to (7),
D＝v0×(t+m) (6)
D≤x1/100 (7)
wherein D is the response distance, v0The advancing speed of the fertilizing device is represented by t, the response delay time of the fertilizer spraying device is represented by m, the processing time of the control device for predicting the nutritional requirement of the corn is represented by m, v is represented by m, and0the unit of (d) is meter/second, the unit of t is second, and the unit of m is second;
and the response position judging module judges whether to send a fertilizer spraying instruction to the fertilizer spraying device control module according to the position of the fertilizer spraying device, the position information of the corn obtained from the corn detecting and positioning module and the response distance.
In the application of on-line real-time target fertilization, delay conditions caused by image processing, instruction transmission, control response of a fertilizer spraying device and the like exist. Therefore, the fertilization position can be controlled more accurately by setting the response distance for the fertilization action.
The invention also aims to provide a fertilizing method adopting the fertilizing device, which comprises the following steps:
(1) when the fertilizing device advances, the image acquisition device acquires the image information of the area to be fertilized in real time;
(2) and the control device gives control instructions after analyzing the image information in real time so as to respectively control the fertilizer spraying devices.
In the method of the present invention, the step (2) specifically includes:
s1: and a corn detection positioning module is used for acquiring color image information in the image information, detecting whether corn exists in the area to be fertilized or not according to the color image information, positioning the corn and judging the area to be fertilized as a fertilization area when the corn is detected, and judging the area to be fertilized as a non-fertilization area when the corn is not detected.
The method of the present invention further includes, after step S1:
s2: after the nutrition demand of the corns in the fertilizing area is predicted by the nutrition demand diagnosis module, the fertilizer spraying device control module controls the fertilizer spraying amount of the fertilizer spraying device according to the nutrition demand, and the specific mode is as follows:
s21: segmenting all the corns at a pixel level from the color image information corresponding to the fertilization area to obtain segmented areas;
s22: extracting spectral image information of the segmented areas, wherein the spectral image information comprises infrared spectrum information and near infrared spectrum information, and calculating vegetation indexes (spectral parameters) of the corns respectively according to the spectral image information;
s23: extracting depth image information of the segmented regions to respectively obtain spatial morphological parameters of the corns;
s24: predicting nutritional requirements of each of the corn in real time according to the vegetation index and the spatial morphological parameters of each of the corn;
s25: and the fertilizer spraying device control module respectively controls the fertilizer spraying amount of the fertilizer spraying device corresponding to each corn according to the nutrition requirement of each corn.
The method of the present invention further includes, after step S2:
s3: and the response position judging module judges whether to send a fertilizer spraying instruction to the fertilizer spraying device control module according to the position information of the corns, the position of the fertilizer spraying device and the response distance determined by the response distance determining module.
In the case of corn crops, the row spacing is generally 30-40cm, the row spacing is generally 50-60cm, and when in topdressing in the seedling stage, besides obvious soil gaps, position areas such as weeds and the like which do not need to be fertilized exist among the crop plants. The invention can effectively identify and distinguish weeds, soil background and the like, and close the fertilizing device in the position areas which do not need to be fertilized, thereby further reducing the fertilizer input, improving the resource utilization rate and promoting the protection of the soil environment.
The invention relates to a crop nutrition demand real-time diagnosis method based on the fusion of spectral parameters and spatial morphological parameters, which can establish a corn plant nutrition diagnosis model by utilizing the obtained spectral parameters and spatial morphological parameters and is used for predicting the nutrition demand of corn plants in real time. In the advancing process of the fertilizing device, the variable operation of the fertilizing device is dynamically and variably controlled through the real-time prediction of the nutrition requirement of the corn plants.
The invention also provides application of the fertilizing device or method in corn planting.
The fertilizing device or method provided by the invention is also suitable for other wide-row crops similar to corn planting modes.
The invention has the beneficial effects that:
the method utilizes the RGB-D camera to simultaneously acquire the color image and the depth image below the nozzle, utilizes the depth learning and image processing technology to detect the corn plant in real time, diagnoses the nutrition state of the corn plant through the spectral information of the red light channel and the near infrared channel and the three-dimensional shape information fused with the depth information, provides more reliable data support for fertilization decision, simultaneously realizes targeted variable fertilization, avoids spraying fertilizer at the soil background between the non-crop plant and the plant, ensures more accurate fertilization, improves the utilization rate of fertilizer resources, can reduce the pollution damage of chemical fertilizer to the field environment, and has good market prospect.
Drawings
FIG. 1 is a schematic diagram of a prior art on-line real-time variable fertilization method based on a spectrum sensor;
FIG. 2 is a schematic view of a targeted variable fertilizing device for corn according to the present invention;
FIG. 3 is a schematic view of the flat fan nozzle spray pattern of the present invention;
FIG. 4 is a schematic view of the nozzle spray in relation to the camera field of view according to the present invention;
FIG. 5 is a schematic view of the targeting detection of the present invention;
FIG. 6 is a schematic diagram of image pixel segmentation according to the present invention;
FIG. 7 is a schematic illustration of the control of the target application of the present invention;
FIG. 8 is a schematic diagram of pulse width modulation according to the present invention;
FIG. 9 is a schematic diagram of a PWM based variable spray of the present invention;
FIG. 10 is a schematic flow chart of the targeted variable fertilization method of the present invention.
Detailed Description
Preferred embodiments of the present invention will be described in detail with reference to the following examples. It is to be understood that the following examples are given for illustrative purposes only and are not intended to limit the scope of the present invention. Various modifications and alterations of this invention will become apparent to those skilled in the art without departing from the spirit and scope of this invention.
Example 1
The embodiment provides a corn targeted variable fertilizing device, and a schematic diagram of the device is shown in fig. 2. The device includes: the fertilizer spraying device comprises a plurality of image acquisition devices, a control device and a plurality of fertilizer spraying devices; the image acquisition device acquires image information of an area to be fertilized in real time when the fertilization device moves forward, and transmits the image information to the control device, and the control device respectively controls the fertilization device according to the image information.
The image acquisition device is an RGB-D camera and can acquire color image information, spectral image information and depth image information of the area to be fertilized.
The fertilizer spraying device comprises a flat fan-shaped nozzle, the spraying shape of the flat fan-shaped nozzle is fan-shaped, as shown in figure 3, the front view of the spraying shape of the nozzle is fan-shaped, the spraying angle of the flat fan-shaped nozzle is theta, and the right view of the flat fan-shaped nozzle is a straight line, so that the accurate positioning of the spraying position can be ensured. When the fertilizer applicator is installed, the fan-shaped surface (the plane where the fan-shaped surface is located) is vertical to the horizontal plane, and the vertical projection of the fan-shaped surface on the plane of the area to be fertilized is vertical to the advancing direction of the fertilizer applicator.
The angle of view of the RGB-D cameras perpendicular to the advancing direction of the fertilizing device is 2 times of the spraying angle of the flat fan-shaped nozzles (namely each RGB-D camera is responsible for image acquisition of 2 flat fan-shaped nozzle fertilizing areas), and the number of the flat fan-shaped nozzles is twice of that of the RGB-D cameras. Each flat fan nozzle is individually controlled by an on-off solenoid valve.
The RGB-D cameras (hereinafter referred to as cameras) and the flat fan-shaped nozzles (hereinafter referred to as nozzles) are arranged on the same horizontal plane (namely a spray rod) in parallel, and each camera is arranged in the middle of two nozzles and is responsible for simultaneously carrying out target detection on the two nozzles. As shown in the schematic diagram of the position relationship between the nozzle spraying and the camera view field in fig. 4, the nozzle 1 and the nozzle 2 are respectively responsible for fertilizing the target in the left and right half areas in the camera view field. The vertical projection of the spray sector of the nozzle in the camera field of view is a straight line perpendicular to the advancing direction of the fertilizing mechanism (fertilizing device). And the straight line is positioned in the middle of the visual field of the camera in the advancing direction of the fertilizing mechanism.
The installation height h of the nozzle and the camera responsible for image acquisition of the fertilization area of the nozzle, the spraying angle theta of the nozzle and the field angle theta of the camera1、θ2The following constraint relationships are satisfied:
d1≈2×x1 (3)
d2≈n×x2 (4)
wherein d is1The distance of the field of view of the camera parallel to the advancing direction of the fertilizing mechanism, d2Distance of field of view, x, of camera perpendicular to direction of travel of fertilizing mechanism1For planting corn at a plant spacing of x2For planting row spacing of corn, theta1The angle of view, theta, of the camera parallel to the direction of travel of the fertilizing mechanism2The angle of view of the camera perpendicular to the advancing direction of the fertilizing mechanism, h is the height of the camera and the nozzle from the ground, n is the number of the nozzles corresponding to the image acquisition area of the camera, and theta are1、θ2In degrees, h in centimeters, d1、d2、x1、x2In centimeters. In this example, d1Is 100 cm, d2Is 60cm, x1Is 50 cm, x2Is 30 cm, theta is 33.4 degrees, theta1Is 90 degrees, theta2At 62 degrees, h is 50 cm and n is 2.
The control device comprises a corn detection and positioning module, the corn detection and positioning module detects whether corns exist in the area to be fertilized according to the color image information of the area to be fertilized, which is acquired by the camera, and when the corns are detected, the corns are further positioned. The corn detection positioning module comprises a YOLOv3 target detection network.
The control device further comprises a nutrition demand diagnosis module and a fertilizer spraying device control module, the nutrition demand diagnosis module predicts the nutrition demand of the corn according to spectral image information and depth image information of the corn acquired by the camera, and the fertilizer spraying device control module controls the fertilizer spraying amount of the corresponding fertilizer spraying device according to the nutrition demand.
The control device also comprises a response distance determining module and a response position judging module, the response distance determining module determines the response distance of the fertilizer spraying device according to the following formulas (6) to (7),
D＝v0×(t+m) (6)
D≤x1/100 (7)
wherein D is the response distance, v0The advancing speed of the fertilizing device is represented by t, the response delay time of the fertilizer spraying device is represented by m, the processing time of the control device for predicting the nutritional requirement of the corn is represented by m, v is represented by m, and0the unit of (d) is meter/second, the unit of t is second, and the unit of m is second; in this example, D is 0.36 m, v00.3 m/s, t 1 s, m 0.2 s.
And the response position judging module judges whether to send a fertilizer spraying instruction to the fertilizer spraying device control module according to the position of the fertilizer spraying device, the position information of the corn obtained from the corn detecting and positioning module and the response distance.
The embodiment also provides a fertilizing method adopting the fertilizing device, which specifically comprises the following steps:
(1) when the fertilizing devices advance, the image acquisition devices acquire image information of areas to be fertilized corresponding to the visual fields of the image acquisition devices in real time;
(2) and the control device gives control instructions after analyzing the image information in real time so as to respectively control the corresponding fertilizer spraying devices.
The step (2) specifically comprises the following steps:
s1: the method comprises the steps of obtaining color image information in each image information by a corn detection and positioning module, detecting whether corn exists in a region to be fertilized or not by adopting a target detection network YOLOv3 trained by a deep learning image processing technology according to the color image information, positioning the corn and judging the region to be fertilized as a fertilization region when the corn is detected, and judging the region to be fertilized as a non-fertilization region when the corn is not detected.
As shown in fig. 5, during the movement of the fertilizing device, the target is detected in real time, and a target frame is formed around the target. And positioning the target by positioning the pixel distance between the target frame and the edge of the image.
S2: after the nutrition demand of the corns in the fertilizing area is predicted by the nutrition demand diagnosis module, the fertilizer spraying device control module controls the fertilizer spraying amount of the fertilizer spraying device according to the nutrition demand, and the method comprises the following specific steps:
s21: segmenting all the corns at a pixel level from the color image information corresponding to the fertilization area to obtain segmented areas;
s22: extracting spectral image information of the segmented areas, wherein the spectral image information comprises infrared spectrum information and near infrared spectrum information, and calculating vegetation indexes (spectral parameters) of the corns respectively according to the spectral image information;
s23: extracting depth image information of the segmented regions to respectively obtain spatial morphological parameters of the corns;
s24: predicting nutritional requirements of each of the corn in real time according to the vegetation index and the spatial morphological parameters of each of the corn;
the real-time diagnosis method of the present embodiment is to perform pixel-level segmentation on the maize plant in the target frame when the maize plant is detected, as shown in fig. 6. Spectral information of the red light channel and the near infrared channel of the segmented region is then extracted by the nutritional requirement diagnosis module, and a vegetation index (such as Normalized Difference Vegetation Index (NDVI)) is calculated in combination. In addition, the segmented regions are mapped to a depth image, the depth information of each pixel in the segmented regions is extracted, and the spatial form parameter, namely the three-dimensional leaf area index (3D-LAI), of the whole corn plant is obtained through accumulation of the depth information of each pixel in the segmented regions. And establishing a corn plant nutrition diagnosis model by using the obtained spectral parameters and spatial morphological parameters for predicting the nutrition requirement of the corn plant in real time.
S25: and the fertilizer spraying device control module respectively controls the fertilizer spraying amount of the fertilizer spraying device corresponding to each corn according to the nutrition requirement of each corn.
S3: and the response position judging module judges whether to send a fertilizer spraying instruction to the fertilizer spraying device control module according to the position information of the corns, the position of the fertilizer spraying device and the response distance determined by the response distance determining module.
Specifically, to avoid the problem of inaccurate real-time target fertilizer spraying caused by response delay, a control schematic diagram of the specific target fertilizer spraying of the embodiment is shown in fig. 7, where "On" represents opening a nozzle, "Off" represents closing the nozzle, and a square frame is a target frame, where the target frame is located. The traveling direction of the spraying mechanism (fertilizing device) is taken as the positive direction, the left side frame of each target frame in the figure 7 is taken as the front side, and the right side frame is taken as the back side. When the control device judges that the distance between the spraying position of the nozzle and the front edge of the target frame is equal to the response distance, a control instruction for opening the electromagnetic valve is sent, and the nozzle is opened for fertilizing; and when the control device judges that the distance between the spraying position of the nozzle and the rear side of the target frame is equal to the response distance, a control instruction for closing the electromagnetic valve is sent, and the nozzle is closed to stop fertilizing. Therefore, the on-off of the electromagnetic valve is controlled according to the detection result of the position of the target object so as to realize accurate target fertilization.
The present embodiment controls the opening and closing time of the electromagnetic valve by a Pulse Width Modulation (PWM) technique, and further controls the opening and closing time of the nozzle to change the fertilizing output. Through Arduino and singlechip etc. as the next computer controller, can output the PWM control signal of different duty cycles, the duty cycle is bigger, and the time that the solenoid valve was opened is longer, and the fertilizer sprayed by the nozzle is more, otherwise, the fertilizer that the nozzle sprayed is less. As shown in FIG. 8, "On" indicates opening the solenoid valve; "Off" indicates that the solenoid valve is closed.
The response position judging module determines the time for sending a control instruction to the lower computer according to the position relation between the spraying position of the nozzle and the corn plant target, and the reference is made to fig. 9. And determining the duty ratio of the PWM control signal according to the prediction result of the nutrition diagnosis model. And normalizing the output result of the diagnosis model, then carrying out grade division, and establishing a mapping relation with the duty ratio of PWM. If the model prediction result is in the level 1[0,0.3], the duty ratio of PWM is 75%; when the model prediction result is in the level 2[0.3,0.6], the duty ratio of PWM is 50%; when the model prediction result is at level 3[0.6,1], the duty cycle of the PWM is 25%. The lower computer receives the instruction of the control device, outputs PWM control signals with different duty ratios, controls the opening and closing time of the electromagnetic valve, and further controls the on-off time of the nozzle so as to realize variable fertilization.
In the advancing process of the fertilizing device, the electromagnetic valve is controlled to be switched on and off according to the real-time processing result of the target detection network on each frame of image, the on-off state of the nozzle is further controlled, and accurate target variable fertilizing can be achieved.
The flow of the target variable fertilization method of the embodiment is shown in fig. 10.
The method comprises the steps that when the fertilizing device moves forward, firstly, image data of a coverage area below a nozzle are collected in real time through an RGB-D camera, secondly, the RGB image data are input into a corn plant detection network (YOLOv3), the RGB image data are processed in real time through a deep learning and image processing technology, and corn plants are detected in real time. When a corn plant is detected, the nozzle is determined to pass through the fertilization area (i.e., the corn plant location). When no corn plant is detected, judging that the nozzle passes through a non-fertilization area (namely weeds or a soil background), not performing subsequent variable fertilization operation, and continuously processing the next frame of RGB image. And thirdly, when and only when the nozzle passes through the fertilizing area, carrying out segmentation processing on the corn plant, extracting and processing spectrum information of a red light channel and a near infrared channel of the RGB-D camera, fusing depth information to calculate three-dimensional shape information of the corn plant, and predicting the corn nutrition demand according to the spectrum information and the three-dimensional shape information. And then, judging whether the corn plants reach a preset response position or not according to the positions of the corn plants, if so, determining the duty ratio of a PWM control signal according to the nutrition demand grade of the corn plants, and sending a control instruction to control the opening and closing time of an electromagnetic valve so as to change the fertilization output quantity. When the nozzle passes through the non-fertilization area, the electromagnetic valve is always in a closed state, and the fertilization device does not perform fertilization operation.
Although the invention has been described in detail hereinabove with respect to a general description and specific embodiments thereof, it will be apparent to those skilled in the art that modifications or improvements may be made thereto based on the invention. Accordingly, such modifications and improvements are intended to be within the scope of the invention as claimed.
Claims (6)
1. The utility model provides a maize is to target variable rate fertilizer injection unit which characterized in that includes: the device comprises at least one image acquisition device, a control device and at least one fertilizer spraying device;
the image acquisition device acquires image information of an area to be fertilized in real time when the fertilization device moves, and the control device respectively controls the fertilization device according to the image information;
the fertilizer spraying device comprises a flat fan-shaped nozzle, the spraying shape of the flat fan-shaped nozzle is fan-shaped, the spraying angle of the flat fan-shaped nozzle is theta, and the vertical projection of the plane where the fan-shaped nozzle is located on the plane of the area to be fertilized is perpendicular to the advancing direction of the fertilizer spraying device;
the image acquisition device and the fertilizer spraying device are arranged on the same horizontal plane, and the vertical projection of the fan shape on the plane of the area to be fertilized is positioned in the middle position of the field distance of the image acquisition device parallel to the advancing direction;
the arrangement of the image acquisition device and the fertilizer spraying device responsible for fertilizing in the field of view area of the image acquisition device satisfies the following formulas (1) to (5):
d1≈2×x1 (3)
d2≈n×x2 (4)
wherein d is1Distance of field of view parallel to the direction of travel for the image acquisition device, d2A field distance, x, perpendicular to the direction of travel for the image acquisition device1For planting corn at a plant spacing of x2For planting row spacing of corn, theta1Angle of view, θ, of said image capturing device parallel to said direction of travel2The field angle of the image acquisition device perpendicular to the advancing direction, h is the height of the image acquisition device and the fertilizer spraying device from the ground, n is the number of the fertilizer spraying devices, and theta are1、θ2In degrees, h in centimeters, d1、d2、x1、x2In units of centimeters;
the image information comprises color image information, spectral image information and depth image information; the image acquisition device is an RGB-D camera;
the control device comprises a corn detection and positioning module, the corn detection and positioning module detects whether corn exists in the area to be fertilized according to the color image information, and when the corn is detected, the corn is positioned;
the control device further comprises a nutritional requirement diagnosis module and a fertilizer spraying device control module, the nutritional requirement diagnosis module predicts the nutritional requirement of the corn according to the spectral image information and the depth image information of the corn, and the fertilizer spraying device control module controls the fertilizer spraying amount of the fertilizer spraying device according to the nutritional requirement;
the control device also comprises a response distance determining module and a response position judging module, the response distance determining module determines the response distance of the fertilizer spraying device according to the following formulas (6) to (7),
D＝v0×(t+m) (6)
D≤x1/100 (7)
wherein D is the response distance, v0Is the advancing speed of the fertilizing device, t is the response delay time of the fertilizing device, m is the processing time of the control device for predicting the nutrition requirement of the corn, D is the unit of meter, v0The unit of (d) is meter/second, the unit of t is second, and the unit of m is second;
and the response position judging module judges whether to send a fertilizer spraying instruction to the fertilizer spraying device control module according to the position of the fertilizer spraying device, the position information of the corn obtained from the corn detecting and positioning module and the response distance.
2. A fertilizing method using the fertilizing apparatus of claim 1, characterized by comprising the steps of:
(1) when the fertilizing device advances, the image acquisition device acquires the image information of the area to be fertilized in real time;
(2) and the control device gives control instructions after analyzing the image information in real time so as to respectively control the fertilizer spraying devices.
3. The method according to claim 2, wherein step (2) comprises in particular:
s1: and a corn detection positioning module is used for acquiring color image information in the image information, detecting whether corn exists in the area to be fertilized or not according to the color image information, positioning the corn and judging the area to be fertilized as a fertilization area when the corn is detected, and judging the area to be fertilized as a non-fertilization area when the corn is not detected.
4. The method according to claim 3, further comprising, after the step S1:
s2: after the nutrition demand of the corns in the fertilizing area is predicted by the nutrition demand diagnosis module, the fertilizer spraying device control module controls the fertilizer spraying amount of the fertilizer spraying device according to the nutrition demand, and the specific mode is as follows:
s21: segmenting all the corns at a pixel level from the color image information corresponding to the fertilization area to obtain segmented areas;
s22: extracting spectral image information of the segmented areas, wherein the spectral image information comprises infrared spectrum information and near infrared spectrum information, and calculating vegetation indexes of the corns respectively according to the spectral image information;
s23: extracting depth image information of the segmented regions to respectively obtain spatial morphological parameters of the corns;
s24: predicting nutritional requirements of each of the corn in real time according to the vegetation index and the spatial morphological parameters of each of the corn;
s25: and the fertilizer spraying device control module respectively controls the fertilizer spraying amount of the fertilizer spraying device corresponding to each corn according to the nutrition requirement of each corn.
5. The method according to claim 4, further comprising, after the step S2:
s3: and the response position judging module judges whether to send a fertilizer spraying instruction to the fertilizer spraying device control module according to the position information of the corns, the position of the fertilizer spraying device and the response distance determined by the response distance determining module.
6. Use of the fertilizing apparatus according to claim 1 or the method according to any of claims 2-5 in corn planting.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911229580.0A CN111066442B (en) | 2019-12-04 | 2019-12-04 | Targeted variable fertilization method and device for corn and application |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911229580.0A CN111066442B (en) | 2019-12-04 | 2019-12-04 | Targeted variable fertilization method and device for corn and application |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111066442A CN111066442A (en) | 2020-04-28 |
| CN111066442B true CN111066442B (en) | 2021-08-06 |
Family
ID=70312844
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911229580.0A Active CN111066442B (en) | 2019-12-04 | 2019-12-04 | Targeted variable fertilization method and device for corn and application |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111066442B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113692793A (en) * | 2021-08-27 | 2021-11-26 | 淄博市农业科学研究院 | Millet field operation machine for intertillage, weeding, ridging and variable fertilization |
| CN113760016A (en) * | 2021-09-03 | 2021-12-07 | 霍山县安绿园农业发展有限公司 | Environment management and control system for vegetable and fruit greenhouse |
| CN114514914A (en) * | 2021-12-28 | 2022-05-20 | 中国农业大学 | Intelligent sensing fertilization and pesticide spraying method and device |
| CN115226458B (en) * | 2022-07-29 | 2024-05-03 | 驻马店志恒信息科技有限公司 | Unmanned agricultural machinery fertilization control method |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN2640609Y (en) * | 2003-07-29 | 2004-09-15 | 中国农业大学 | Automatic targetting atomizing machine |
| CN201500606U (en) * | 2009-07-03 | 2010-06-09 | 中国科学院武汉植物园 | Environmental-friendly portable spray machine |
| CN202588134U (en) * | 2012-05-16 | 2012-12-12 | 绿友机械集团股份有限公司 | Novel trailer-type pesticide spray machine |
| CN103782705A (en) * | 2014-01-26 | 2014-05-14 | 北京农业信息技术研究中心 | Variable target liquid fertilizer apparatus of orchard and fertilizing method |
| CN103814664A (en) * | 2012-11-16 | 2014-05-28 | 中国农业机械化科学研究院 | Variable NPK (nitrogen, phosphorus and potassium) fertilizer applicator and method for determining fertilizing formulation |
| CN103999841A (en) * | 2014-04-18 | 2014-08-27 | 浙江大学 | Automatic targeting and spraying system |
| CN104115809A (en) * | 2014-07-24 | 2014-10-29 | 中国农业大学 | Toward-target spraying machine and method for crops in field |
| RU2572301C1 (en) * | 2014-07-11 | 2016-01-10 | Федеральное государственное бюджетное образовательное учреждение Высшего профессионального образования "Оренбургский государственный агарарный университет" | Electrodischarge fertiliser distributor |
| CN107976921A (en) * | 2017-11-01 | 2018-05-01 | 中国农业大学 | A kind of fertilizer apparatus and method |
-
2019
- 2019-12-04 CN CN201911229580.0A patent/CN111066442B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN2640609Y (en) * | 2003-07-29 | 2004-09-15 | 中国农业大学 | Automatic targetting atomizing machine |
| CN201500606U (en) * | 2009-07-03 | 2010-06-09 | 中国科学院武汉植物园 | Environmental-friendly portable spray machine |
| CN202588134U (en) * | 2012-05-16 | 2012-12-12 | 绿友机械集团股份有限公司 | Novel trailer-type pesticide spray machine |
| CN103814664A (en) * | 2012-11-16 | 2014-05-28 | 中国农业机械化科学研究院 | Variable NPK (nitrogen, phosphorus and potassium) fertilizer applicator and method for determining fertilizing formulation |
| CN103782705A (en) * | 2014-01-26 | 2014-05-14 | 北京农业信息技术研究中心 | Variable target liquid fertilizer apparatus of orchard and fertilizing method |
| CN103999841A (en) * | 2014-04-18 | 2014-08-27 | 浙江大学 | Automatic targeting and spraying system |
| RU2572301C1 (en) * | 2014-07-11 | 2016-01-10 | Федеральное государственное бюджетное образовательное учреждение Высшего профессионального образования "Оренбургский государственный агарарный университет" | Electrodischarge fertiliser distributor |
| CN104115809A (en) * | 2014-07-24 | 2014-10-29 | 中国农业大学 | Toward-target spraying machine and method for crops in field |
| CN107976921A (en) * | 2017-11-01 | 2018-05-01 | 中国农业大学 | A kind of fertilizer apparatus and method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111066442A (en) | 2020-04-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111066442B (en) | Targeted variable fertilization method and device for corn and application | |
| US11237563B2 (en) | Method for analyzing individual plants in an agricultural field | |
| US11601587B2 (en) | System and method for monitoring plants in plant growing areas | |
| Tian | Development of a sensor-based precision herbicide application system | |
| CN110163138B (en) | Method for measuring and calculating wheat tillering density based on multispectral remote sensing image of unmanned aerial vehicle | |
| CN103999841B (en) | A kind of automatic target detection spraying system | |
| CN109191074A (en) | Wisdom orchard planting management system | |
| CN105173085A (en) | Automatic control system and method for variable pesticide spraying of unmanned aerial vehicle | |
| CN103782705B (en) | A kind of orchard is to target variable liquid fertilizing device and fertilizing method | |
| CN103891697A (en) | Drug spraying robot capable of moving indoors autonomously and variable drug spraying method thereof | |
| CN107295818B (en) | Automatic seedling avoiding system and method for paddy field weeding machine | |
| CN209749385U (en) | Accurate irrigation monitoring system | |
| CN110140704A (en) | A kind of intelligent pesticide spraying method and system for plant protection drone | |
| CN102172233A (en) | Method for carrying out real-time identification and targeted spraying on cotton field weeds | |
| CN112136423A (en) | Multifunctional agricultural robot | |
| CN103530643A (en) | Pesticide positioned spraying method and system on basis of crop interline automatic identification technology | |
| CN107748886A (en) | A kind of rail mounted contemporary standard orchard information sensory perceptual system based on depth camera | |
| CN113439520A (en) | Intelligent decision-making method and system for crop irrigation and fertilization | |
| CN113608551A (en) | Unmanned agricultural machinery group cooperation system and application method thereof | |
| Ma et al. | Automatic detection of crop root rows in paddy fields based on straight-line clustering algorithm and supervised learning method | |
| CN204157280U (en) | A kind of small rail car drip irrigation type Variable Fertilization System | |
| CA3170566A1 (en) | Systems and methods for monitoring plants in plant growing areas | |
| CN104221571B (en) | Small rail car drip irrigation type variable fertilization system | |
| KR101109337B1 (en) | Automatic pest recognition and control system and method | |
| CN108323389A (en) | The detection method and device of the rice transplanting rice shoot spacing in the rows and cave rice shoot number of rice transplanter |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |