CN108537164B - Method and device for monitoring germination rate of dibbling and sowing based on unmanned aerial vehicle remote sensing - Google Patents
Method and device for monitoring germination rate of dibbling and sowing based on unmanned aerial vehicle remote sensing Download PDFInfo
- Publication number
- CN108537164B CN108537164B CN201810305437.4A CN201810305437A CN108537164B CN 108537164 B CN108537164 B CN 108537164B CN 201810305437 A CN201810305437 A CN 201810305437A CN 108537164 B CN108537164 B CN 108537164B
- Authority
- CN
- China
- Prior art keywords
- dibbling
- plant
- remote sensing
- sowing
- seeding
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/20—Scenes; Scene-specific elements in augmented reality scenes
Abstract
The invention provides a method and a device for monitoring the germination rate of dibbling and sowing based on unmanned aerial vehicle remote sensing, wherein the method comprises the following steps: taking an independent image patch of a crop vegetation area in a complete point sowing farmland image as a monitoring unit, and dividing an overlapped plant unit and a single plant unit by monitoring the outline shape and the size characteristic of each monitoring unit; determining the number of pixels covered by a single plant by using a data statistical method, and calculating the overlapping times of plants in the overlapping plant unit based on the number of pixels covered by the single plant; and calculating the total number of budding plants by counting the total number of the single plant units and the overlapping times of each overlapping plant unit, and calculating the budding rate of the target point sowing based on the total number of the budding plants and the total number of the sowing holes. The method can effectively improve the monitoring efficiency, accuracy and applicability of the germination rate of the crops, thereby greatly saving the consumption of human resources and improving the informatization management level.
Description
Technical Field
The invention relates to the technical field of agricultural information, in particular to a method and a device for monitoring the germination rate of dibbling and sowing based on unmanned aerial vehicle remote sensing.
Background
The method has the advantages that the budding condition of the sowed crops can be rapidly and accurately monitored, and the method has very important significance for timely reseeding management of plots with low budding rate, so that the yield of the lands is effectively improved, the cost is saved, and the generation benefit is improved.
Currently, the monitoring of the germination rate of crops is generally realized by manually performing visual search and statistics on farmlands. Or, the green crop vegetation and the surface soil are obtained by collecting the image of the crop planting area, dividing the image into pixel groups with fixed pixel numbers, and then classifying the pixel groups. And then carrying out topology conversion on the obtained pixels classified as the crop vegetation to obtain vectors of crop plants, counting the number of the vectors to obtain the total amount of the crop plants, and finally calculating the ratio of the total amount of the crop plants to the seeding amount to obtain the germination rate of the crops. Here, topology conversion refers to a process of extracting and converting a boundary of an isolated pixel set formed by gathering a plurality of crop pixels into a vector.
The manual land parcel investigation method has high precision and good effect, but consumes manpower resources and has low efficiency, and particularly in large-area crop planting areas, the subsequent crop yield is influenced because the seeds of part of farmlands cannot be timely reseeded after budding. And, as the labor cost rises, the method will be increasingly difficult to be widely used.
By collecting the image of the crop plot and carrying out pixel group division, classification, topology conversion and counting, the method adopts a fixed pixel quantity standard when the pixel group division is carried out, image data processing is carried out by taking an image tuple as a unit, and the problems of plant overlapping and plant size are ignored, so that the estimated budding rate has deviation. For example, the potential for overlapping of adjacent seeded spot-sprouted crop plants due to differences in the field environment may result in an estimated germination rate that is lower than is practical, or a plant that is larger may be divided into multiple image tuples, resulting in repeated calculations and thus higher estimates.
Disclosure of Invention
In order to overcome the problems or at least partially solve the problems, the invention provides a method and a device for monitoring the germination rate of dibbling and sowing based on unmanned aerial vehicle remote sensing, which are used for effectively improving the monitoring efficiency, accuracy and applicability of the germination rate of crops.
On one hand, the invention provides a method for monitoring the germination rate of dibbling and sowing based on unmanned aerial vehicle remote sensing, which comprises the following steps: s1, taking the independent image patches of the crop vegetation areas in the complete dibbling and seeding farmland images as monitoring units, and dividing overlapping plant units and single plant units by monitoring the outline shape and size characteristics of each monitoring unit; s2, determining the number of pixels covered by a single plant by using a data statistics method, and calculating the overlapping times of plants in the overlapping plant units based on the number of pixels covered by the single plant; s3, calculating the total number of total plants after germination by counting the total number of the single plant units and the overlapping times of each overlapping plant unit, and calculating the germination rate of the target dibbling seeding based on the total number of the total plants after germination and the total number of the seeding dibbling holes.
Wherein the step of S1 further comprises: and according to the contour shape and size characteristics of the single plant of the target dibbling crop, a plant overlapping judgment model is made, and according to the contour shape and size characteristics of each monitoring unit, each monitoring unit is judged to belong to the overlapping plant unit or the single plant unit by using the plant overlapping judgment model.
Wherein the step of S2 further comprises: utilizing a data statistical method to make a frequency distribution percentage diagram of the number of pixels covered by a single plant, and calculating the number of pixels covered by the single plant based on the frequency distribution percentage diagram; and calculating the average contour diameter of the single plant based on the number of pixels covered by the single plant, and calculating the overlapping times of the plants based on the average contour diameter of the single plant and the maximum axial length of the overlapping plant units.
Further, before the step of S1, the method further includes: s01, controlling the unmanned aerial vehicle to fly above a target dibbling and seeding farmland, and collecting remote sensing images of the target dibbling and seeding farmland in a plurality of discrete areas; and S02, splicing the remote sensing images of the plurality of discrete region point sowing farmlands to obtain the complete point sowing farmland image, classifying the complete point sowing farmland image by using a supervision and classification method, and extracting the crop vegetation area.
Further, between the steps of S01 and S02, the method further comprises: carrying out quality monitoring on the remote sensing images of the plurality of discrete region point sowing farmlands, and outputting the remote sensing images of the discrete regions with quality indexes meeting set standards; correspondingly, the step of splicing the remote sensing images of the multiple discrete region point sowing farmlands to obtain the complete point sowing farmland image in the step S02 further includes: and splicing the remote sensing images of the discrete areas with the quality indexes meeting the set standard to obtain the complete dibbling and seeding farmland image.
Further, in step S02, between the step of classifying the complete dibbling and seeding farmland image and the step of extracting the crop vegetation area by using the supervised classification method, the method further comprises: and removing isolated spots in the remote sensing image classified into the vegetation area, wherein the isolated spots represent isolated vegetation areas with the number of covered pixels smaller than a given value.
In step S02, the step of stitching the remote sensing images of the discrete area hill-drop sowing farmland to obtain the complete hill-drop sowing farmland image further includes: inputting the multiple discrete region point sowing farmland remote sensing images into Agisoft Scan software, and outputting orthographic remote sensing images containing geographic coordinates as the complete point sowing farmland images after splicing processing.
Further, the step of S02 further includes: extracting a naked area; after the step of S02, the method further includes: and the crop vegetation area and the bare area are distinguished and displayed according to the geographic coordinates, and a reseeding area is indicated according to a display result.
Further, before the step of calculating the germination rate of the target dibbling seeding based on the total number of the germinated plants and the total number of the seeding dibbles in step S3, the method further comprises: and calculating the total number of the dibbling and dibbling points according to the total number of the dibbling and sowing lines, the length of each dibbling and sowing line, the dibbling and sowing speed and the time interval of the dibbling and sowing between adjacent dibbling points.
On the other hand, the invention provides a device for monitoring the germination rate of dibbling and seeding based on unmanned aerial vehicle remote sensing, which comprises: at least one memory, at least one processor, a communication interface, and a bus; the memory, the processor and the communication interface complete mutual communication through the bus, and the communication interface is used for information transmission between the monitoring device and the unmanned aerial vehicle data communication interface and the human-computer interaction interface; the memory has stored therein a computer program operable on the processor, which when executed implements the method as described above.
According to the method and the device for monitoring the germination rate of the dibbling seeding based on the unmanned aerial vehicle remote sensing, provided by the invention, the unmanned aerial vehicle remote sensing image data of a farmland after the dibbling seeding is processed, the number of pixels covered by a plant in a remote sensing image, the overlapping condition of the plant and the like are analyzed, the automatic and accurate estimation of the target dibbling seeding germination rate is realized, the monitoring efficiency, the accuracy and the applicability of the crop germination rate can be effectively improved, the consumption of human resources is greatly saved, and the informatization management level is improved.
Drawings
FIG. 1 is a flow chart of a method for monitoring the germination rate of dibbling and seeding based on unmanned aerial vehicle remote sensing in the embodiment of the invention;
FIG. 2 is a schematic diagram showing the percentage of frequency distribution of the number of pixels occupied by a single plant of a cotton crop;
FIG. 3 is a flow chart of the method for monitoring the germination rate of dibbling seedlings based on unmanned aerial vehicle remote sensing according to the embodiment of the invention for calculating the overlapping times of plants in overlapping plant units;
FIG. 4 is a flow chart of the method for monitoring the germination rate of dibbling seedlings based on unmanned aerial vehicle remote sensing according to the embodiment of the invention for extracting the vegetation area of the crops;
fig. 5 is a structural block diagram of a device for monitoring the germination rate of dibbling and seeding based on remote sensing of an unmanned aerial vehicle in the embodiment of the invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be obtained by a person skilled in the art without any inventive step based on the embodiments of the present invention, are within the scope of the present invention.
As an aspect of the embodiment of the present invention, the present embodiment provides a method for monitoring a germination rate of a dibbling seeder based on remote sensing of an unmanned aerial vehicle, and with reference to fig. 1, is a flowchart of the method for monitoring a germination rate of a dibbling seeder based on remote sensing of an unmanned aerial vehicle according to the embodiment of the present invention, and includes:
s1, taking the independent image patches of the crop vegetation areas in the complete dibbling and seeding farmland images as monitoring units, and dividing overlapping plant units and single plant units by monitoring the outline shape and size characteristics of each monitoring unit;
s2, determining the number of pixels covered by a single plant by using a data statistics method, and calculating the overlapping times of plants in the overlapping plant units based on the number of pixels covered by the single plant;
s3, calculating the total number of total plants after germination by counting the total number of the single plant units and the overlapping times of each overlapping plant unit, and calculating the germination rate of the target dibbling seeding based on the total number of the total plants after germination and the total number of the seeding dibbling holes.
It can be understood that, in the embodiment, the germination rate estimation of the planting area of the dibbling crop is rapidly performed by utilizing the efficient data acquisition characteristic of the unmanned aerial vehicle and combining the technologies of computer vision, image processing and the like. It should be understood that, for the target farmland after dibbling and seeding, before monitoring the germination rate according to the embodiment, the unmanned aerial vehicle needs to be used for acquiring the whole remote sensing image of the farmland in advance, and the crop vegetation area in the farmland can be identified.
The dibbling method is a planting method in which holes are dug at fixed intervals, seeds are placed in the holes, and soil is covered.
The complete dibbling and seeding farmland image is a full-area remote sensing image of a farmland for dibbling and seeding. For a target dibbling and seeding farmland to be monitored, a continuous area range is covered, and a full-area remote sensing image formed in the continuous area range is a complete dibbling and seeding farmland image.
Then, in step S1, a statistical monitoring unit for plants that have germinated from seeds after dibbling is first determined, that is, an independent image patch in a plant vegetation area is determined as a monitoring unit. Wherein the crop vegetation area is a plant coverage area identified from the complete point-seeding farmland image. The independent image patch in the plant vegetation area refers to an independent image shape which is formed by a plurality of picture elements and is not adjacent to other plant picture elements.
Then, according to the shape of the independent image formed by each monitoring unit, the outline shape and the size characteristic of the monitoring unit are counted, and the plant overlapping state of each monitoring unit is judged by combining the plant outline characteristic of the dibbling crops, namely whether plant overlapping exists or not, and the monitoring unit with the plant overlapping exists, namely an overlapping plant unit, and the monitoring unit without the plant overlapping exists, namely an individual plant unit, are divided.
Optionally, the step of S1 further includes: and according to the contour shape and size characteristics of the single plant of the target dibbling crop, a plant overlapping judgment model is made, and according to the contour shape and size characteristics of each monitoring unit, each monitoring unit is judged to belong to the overlapping plant unit or the single plant unit by using the plant overlapping judgment model.
It can be understood that different crop types have different plant contour shapes and size differences, so that the plant overlapping state of each monitoring unit can be judged according to the contour and size characteristics of a single plant. And for the target dibbling and seeding crops, a plant overlapping judgment model is made according to the outline shape characteristics of the individual plants or the size characteristics of the individual plants.
Then, the outline shape, namely the size characteristic, of each monitoring unit to be monitored is input into the plant overlapping judgment model, and the plant overlapping state of each monitoring unit can be output through comparison and judgment inside the model, namely, the plant overlapping state belongs to either an overlapping plant unit or a single plant unit.
Taking a target dibbling crop as cotton as an example, after splicing a plurality of cotton field remote sensing images acquired by an unmanned aerial vehicle, a cotton field true color image example, namely a complete dibbling farmland image, is obtained. The crop vegetation area can be identified from the complete dibbling and seeding farmland image, but the cotton plants between adjacent dibbling in the identified crop vegetation area have overlapping condition. The extent to which these overlapping regions occur varies for different stages of budding growth, for which identification and extraction is required to improve the accuracy of the monitoring.
It should be understood that the number of seeds sown by one sowing and dibbling is uncertain in actual production, and the reseeding of farmers is carried out by dibbling, namely, the reseeding is not carried out when cotton seedlings exist in one dibbling, and the number of cotton plants is not considered. Therefore, in the embodiment of the invention, one point is taken as a unit, namely one point is considered as one monitoring point, whether cotton seedlings exist in each point is considered, and the number of the cotton seedlings in each point is not considered. That is, a plurality of or one cotton in one acupoint is treated as one strain. Thus, a cotton plant in the description of the embodiments of the present invention refers to a collection of cotton plants formed in one sowing point.
However, it should be appreciated that by increasing the resolution of the captured images reasonably, a detailed monitoring of the actual plant count of cotton plants contained in a single point can be achieved.
Because cotton plant leaves appear circular, a circle can be used to approximately fit a cotton plant. When two cotton plants with the radius of r are adjacent but not completely overlapped, the maximum axial length of the shape forming the image is less than 4r and more than 2r, and the minimum axial length is more than 0 and less than r. Therefore, in practice, the overlapping rate of two plants is not fixed, so that it is difficult to describe the overlapping rate by using a specific formula. In the embodiment of the invention, the maximum axial length L is used for classifying the shape of cotton plants in the imagemaxGreater than 3r and a minimum axial length LminIs less than
The judgment condition for the presence of mutual coverage between plants is specifically represented by the following formula:
in the formula, LmaxIndicating the maximum axial length of the formed shape, i.e. the maximum axial length of the monitoring unit, LminIndicating the minimum axial length of the formed shape, i.e. the minimum axial length of the monitoring unit, rmeanDenotes the average cotton plant radius, SmeanThe area of the average cotton plant is indicated.
It should be noted here that the shape formed by the cotton plant, i.e. the division rule of each monitoring unit, is: an independent image patch which is composed of a plurality of picture elements and is not adjacent to other plant picture elements is a monitoring unit.
Step S2 can be understood as the difficulty in determining the number of pixels that an individual plant occupies, due to the different sizes of the plants in different growing seasons. However, statistical analysis shows that the sizes of the corresponding crop plants in a larger area accord with certain probability distribution, for example, the number of pixels covered by the remote sensing image of a single cotton plant accords with the off-normal distribution. As shown in fig. 2, which is a schematic diagram of the frequency distribution percentage of the number of the pixels occupied by a single plant of the cotton crop, it can be clearly seen that the number of the pixels occupied by the cotton plant has a relatively significant distribution, that is, the area of the pixels occupied by most of the cotton plants is less than 200 pixels, and only 10% of the pixels are greater than 200 pixels.
Therefore, the number of the pixels covered by a single plant is determined by counting the number of the pixels covered by the single plant of the existing target dibbling crop or by sampling statistics. And calculating the overlapping times of the plants in each overlapping plant unit according to the number of the pixels covered by the determined single plant and the size characteristics of the overlapping plant unit divided by the steps, namely calculating the number of the plants contained in each overlapping plant unit.
Optionally, referring to fig. 3, the step of S2 is a flowchart illustrating a method for monitoring a germination rate of a plant in an overlapped plant unit by dibbling and seeding based on remote sensing by an unmanned aerial vehicle according to an embodiment of the present invention, and the method includes:
and S21, making a frequency distribution percentage diagram of the number of the pixels covered by the single plant by using a data statistical method, and calculating the number of the pixels covered by the single plant based on the frequency distribution percentage diagram.
It will be appreciated that according to the above embodiments, the number of pixels covered by the plant outline is statistically regular for different crop plants, or for the same crop during different growth periods. For the existing crop plant types, a frequency distribution percentage diagram of the number of pixels covered by a single plant is formulated through statistics of the number of pixels covered by the outline of the single plant. The frequency distribution percentage diagram of the number of the pixels covered by a single plant is formulated, wherein the horizontal axis of the diagram represents the number of the pixels covered by the single plant, and the vertical axis represents the statistical occurrence frequency of the number of the corresponding pixels.
For example, for the cotton plants of the above examples, the size of the cotton plants is in accordance with the segregation distribution over a larger area. FIG. 2 is a graph of the percentage of frequency distribution of the number of pixels covered by a single cotton plant, showing that the area of the majority of the cotton plants occupied pixels is less than 200 pixels, and only 10% of the pixels are greater than 200.
Then, according to the made frequency distribution percentage diagram of the number of the pixels covered by the single plant, the number of the pixels covered by the single plant is estimated through statistical calculation. For example, considering that most cotton plants do not overlap during the germination stage, the method considers that the number of pixels with the largest occurrence frequency of area pixels is the number of pixels occupied by the single cotton plant to be determined. In addition, considering that certain fluctuation exists in practice, the average value of the number of the image elements within 10% of the maximum image element occurrence frequency is selected as the number of the image elements occupied by the single cotton plant which needs to be determined, and the specific formula is shown in the following formula.
In which p denotes the frequency of the picture element, pmaxIndicating maximumFrequency of occurrence of the number of picture elements, NumpNumber of pixels corresponding to frequency of occurrence, NtotalRepresenting the sum of the numbers of the frequencies of occurrence, SGSDRepresenting the resolution of the image acquisition, SmeanThe area of the average cotton plant is expressed as the mean of the pixel size covered by the individual plants.
The meaning of the above formula is that the area of the pixel with the largest number of current times is taken as Smean. In practice, it may be required to take the area of a nearby pixel with the highest occurrence frequency of pixels as a value range, and for this reason, the embodiment of the present invention is designed as Pmax-10％. It will be appreciated that the area of the cotton over a large area exhibits a specific distribution, and for most plants the leaf area must be the value which occurs the most frequently, and this value is taken as the desired mean value in the embodiment of the invention.
S22, calculating the average contour diameter of the single plant based on the number of pixels covered by the single plant, and calculating the overlapping times of the plants based on the average contour diameter of the single plant and the maximum axial length of the overlapping plant units.
It can be understood that after the number of pixels covered by a single plant is calculated according to the above steps, the evaluation contour diameter of the single plant can be calculated according to the value. It should be understood that the plant contour radius is also possible here. For example, in the case of a single plant with a circular outline, such as a cotton plant, the average outline radius of the single plant is calculated as follows:
in the formula, rmeanDenotes the average cotton plant radius, SmeanThe area of the average cotton plant is expressed as the mean of the pixel size covered by the individual plants.
And meanwhile, observing the axial length of the overlapped plant units in the vegetation area of the crops, and calculating the overlapping times of the plants of the overlapped plant units according to the maximum axial length of the overlapped plant units and the average contour diameter of the single plant obtained by calculation.
For example, for the cotton plants of the above steps, the average cotton plant radius r is obtained in the calculationmeanThen, the number of plant overlaps for each overlapping plant unit of the cotton plant is calculated according to the following formula:
in the formula, NumoverlayRepresenting the number of overlapping plants of overlapping plant units, i.e.several cotton plants overlap, rmeanDenotes the average cotton plant radius, LmaxThe maximum axial length of the formed shape, i.e. the maximum axial length of the monitoring unit, is indicated.
As can be appreciated, the calculation of the germination percentage generally calculates the ratio of the total number of sprouts to the total number of plantings at step S3. In consideration of the actual point sowing, the number of seeds sown in each point does not necessarily have to be one, and the number of plants in one point does not necessarily have to be one. The point hitting is also used as a unit for the seed supplementing, and the seed supplementing is carried out in the point hitting without budding, so that the total planting number is replaced by the total seeding point hitting number.
Dividing single plant units and overlapped plant units according to the steps, calculating the overlapping times of the overlapped plant units, counting the total number of the single plant units and the total number of plants in the overlapped plant units, and summing the two numbers to obtain the total number of the germinated plants. Meanwhile, the total number of the seeding points can be counted according to the actual planting condition. And finally, calculating the ratio of the total number of the germinated plants to the total number of the holes sowed by the sowing to obtain the required germination rate, namely the germination rate of the target hole-sowing.
Wherein, considering that the spacing and the speed of sowing are constant due to the adoption of machine sowing, the total number of sowing points can be calculated conveniently, and then the germination rate is calculated as follows:
in the formula, Rate represents monitoringGermination Rate, NumisolatedDenotes the independent non-overlapping cotton plant, NumsowingIndicates the total number of sowing dibbles, (Num)overlay)iIndicates the number of overlapping plants of the i-th overlapping plant unit, i.e., several cotton plants overlap, and n indicates the total number of overlapping plant units.
Wherein, before the step of calculating the germination rate of the target dibbling seeding based on the total number of the germinated plants and the total number of the seeding dibbles in step S3, the method further comprises: and calculating the total number of the dibbling and dibbling points according to the total number of the dibbling and sowing lines, the length of each dibbling and sowing line, the dibbling and sowing speed and the time interval of the dibbling and sowing between adjacent dibbling points.
It is to be understood that, for the total number of seeding point holes of the above-described embodiment, considering that the spacing and speed of seeding at the time of mechanical planting are constant, the total number of rows, the length of each row, the seeding speed, and the like of seeding in the field of hole seeding at the target point after planting is completed are determined, and therefore, the total number of seeding point holes can be calculated based on the above-described known data and the number relationship between the data. The method comprises the following specific steps:
in the formula, NumsowingIndicates the total number of sowing holes, NrowIndicates the total row number of dibbling seeds, LrowThe length of each row of dibbling and seeding is shown, Speed is shown as the dibbling and seeding Speed, and T is the time interval between adjacent dibbling and seeding.
According to the method for monitoring the germination rate of the dibbling seeding based on the unmanned aerial vehicle remote sensing, provided by the embodiment of the invention, the unmanned aerial vehicle remote sensing image data of the farmland after the dibbling seeding is processed, the number of pixels covered by a plant in a remote sensing image, the overlapping condition of the plant and the like are analyzed, the automatic and accurate estimation of the target dibbling seeding germination rate is realized, the monitoring efficiency, the accuracy and the applicability of the crop germination rate can be effectively improved, the consumption of human resources is greatly saved, and the informatization management level is improved.
In one embodiment, before the step S1, the method further includes a processing flow shown in fig. 4, which is a flow chart of extracting a crop vegetation area in a method for monitoring the germination rate of dibbling and seeding based on unmanned aerial vehicle remote sensing according to an embodiment of the present invention, and the method includes:
and S01, controlling the unmanned aerial vehicle to fly above the target dibbling and seeding farmland, and collecting remote sensing images of the target dibbling and seeding farmland in a plurality of discrete areas.
It can be understood that because the scope of farmland is all relatively broad usually, unmanned aerial vehicle can't once only gather the global remote sensing image in farmland. Therefore, the unmanned aerial vehicle is firstly controlled to fly above the target dibbling and seeding farmland to traverse and collect single remote sensing images of different subareas of the target dibbling and seeding farmland, and a plurality of discrete area dibbling and seeding farmland remote sensing images are formed.
For example, image acquisition in a field cotton field is performed using a Xinjiang fairy 4 drone, the flying height is 10m to 15m, and the resolution of the obtained image is not less than 12 mm.
And S02, splicing the remote sensing images of the plurality of discrete region point sowing farmlands to obtain the complete point sowing farmland image, classifying the complete point sowing farmland image by using a supervision and classification method, and extracting the crop vegetation area.
It can be understood that to many discrete region dibbling seeding farmland remote sensing images that above-mentioned step unmanned aerial vehicle gathered, the emergence rate condition in whole farmland is not convenient for observe from global visual angle. Therefore, before monitoring according to the farmland image, firstly, splicing a plurality of discrete region point-sowing farmland remote sensing images collected by the unmanned aerial vehicle one by one according to adjacent regions, and obtaining a complete classified cotton plant image.
Optionally, in step S02, the step of stitching the remote sensing images of the multiple discrete area hill-drop sowing farmlands to obtain the complete hill-drop sowing farmlands further includes: inputting the multiple discrete region point sowing farmland remote sensing images into Agisoft Scan software, and outputting orthographic remote sensing images containing geographic coordinates as the complete point sowing farmland images after splicing processing.
It can be understood that, in this embodiment, the image stitching is performed by using agisoft scan software, a plurality of discrete region hole-pressing sowing farmland remote sensing images are input, and an image obtained by automatic stitching by the agisoft scan software is an ortho remote sensing image containing geographic coordinates, that is, the image is used as a complete hole-pressing sowing farmland image, and the resolution of the complete hole-pressing sowing farmland image is consistent with that of the original image.
And then, classifying the complete hole-pressing sowing farmland images obtained after splicing by using a supervision classification method to obtain the areas covered by the crop plants in the images, wherein the areas are used as vegetation areas of the crops, and the areas not covered by the plants are used as bare areas.
The supervision and classification method is realized by selecting a representative training field from a research area as a sample. According to samples provided by a known training area, a discriminant function is established by selecting characteristic parameters (such as pixel brightness mean, variance and the like), and the attribution type of non-sample pixels is identified according to the characteristics of the sample type, so that the sample pixels are classified. Specifically, in the embodiment of the invention, ENVI software is used for classifying the complete point-pressing sowing farmland images.
That is, the required image, namely the complete dibbling and seeding farmland image, is loaded in the ENVI software, and two classification categories are created: and manually selecting a region of interest which is the ground surface in the plurality of images and a region which is a plant in the images as training sample regions. Then, images are classified by utilizing the maximum interpretation classification algorithm carried by the ENVI software.
Wherein, on the basis of the above embodiment, between the steps of S01 and S02, the method further comprises: carrying out quality monitoring on the remote sensing images of the plurality of discrete region point sowing farmlands, and outputting the remote sensing images of the discrete regions with quality indexes meeting set standards;
correspondingly, the step of splicing the remote sensing images of the multiple discrete region point sowing farmlands to obtain the complete point sowing farmland image in the step S02 further includes: and splicing the remote sensing images of the discrete areas with the quality indexes meeting the set standard to obtain the complete dibbling and seeding farmland image.
It can be understood that, in the practical application process, because of the farmland topography factor, the images of unmanned aerial vehicle flight environment etc. can make the partial image data quality who gathers not reach categorised identification standard, for example resolution ratio is low excessively, overexposure etc. to lead to image data classification's failure. Therefore, in this embodiment, after the unmanned aerial vehicle collects the remote sensing images of the target point sowing farmland in a plurality of discrete areas, the quality of the collected remote sensing images is monitored. Specifically, if the detected image quality index meets the set standard, the corresponding remote sensing image in the discrete area is reserved, and if the detected image quality index does not meet the set standard, the corresponding remote sensing image in the discrete area is discarded. In addition, the re-flying operation is carried out on the area which is not covered by the primary flight so as to obtain the remote sensing image of the global area covering the target point sowing farmland.
Correspondingly, as the remote sensing image data which cannot meet the set standard are filtered, the remote sensing image data which can meet the set standard are remained. In addition, when the discrete remote sensing image data of the entire region of the farmland is acquired through the flying compensation operation, when the discrete region point sowing farmland remote sensing images are spliced in step S02, only the discrete region remote sensing images whose quality index satisfies the set standard are spliced.
Further, on the basis of the above embodiment, between the step of classifying the complete hole-sowing agricultural land image and the step of extracting the crop vegetation area by using the supervised classification method in step S02, the method further includes:
and removing isolated spots in the remote sensing image classified into the vegetation area, wherein the isolated spots represent isolated vegetation areas with the number of covered pixels smaller than a given value.
It can be understood that due to the influence of factors such as data noise, after the complete dibbling and seeding farmland image is classified, the number of pixels covered by some isolated spots in the formed classified image is obviously less than that of the pixels covered by normal plants, for example, the number of covered pixels is less than 3, so that the classified image cannot be the plant image of the target crop. Therefore, the isolated spots with the number of covered pixels obviously less than that of the pixels covered by normal plants are removed, so that the data processing scale is reduced, and the calculation efficiency is improved. Wherein an isolated blob represents an isolated vegetation zone covering less than a given number of pixels.
Further, on the basis of the foregoing embodiment, the step of S02 further includes: extracting a naked area;
after the step of S02, the method further includes: and the crop vegetation area and the bare area are distinguished and displayed according to the geographic coordinates, and a reseeding area is indicated according to a display result.
It can be understood that, according to the above-described embodiment, when the classification of the complete point-sowing farmland image is performed, two surface types of the crop vegetation area and the bare area can be simultaneously classified. Therefore, for convenience of reseeding, after the crop vegetation area and the bare land area are classified, the two land surface types are displayed on the display unit in different states, and related personnel can perform targeted reseeding on seedling lacking positions according to display results. Therefore, the device can help related personnel to quickly position the seedling lacking point, thereby effectively improving the working efficiency and reducing the labor intensity.
As another aspect of the embodiment of the present invention, the present embodiment provides a device for monitoring a germination rate of a dibbling seed based on remote sensing of an unmanned aerial vehicle, and referring to fig. 5, the structural block diagram of the device for monitoring a germination rate of a dibbling seed based on remote sensing of an unmanned aerial vehicle according to the embodiment of the present invention includes: at least one memory 1, at least one processor 2, a communication interface 3 and a bus 4. Wherein the content of the first and second substances,
the memory 1, the processor 2 and the communication interface 3 complete mutual communication through a bus 4, and the communication interface 3 is used for information transmission between the monitoring device and the unmanned aerial vehicle data communication interface and the human-computer interaction interface; the memory 1 stores a computer program which can run on the processor 2, and the processor 2 executes the program to realize the method for monitoring the germination rate of dibbling and sowing based on unmanned aerial vehicle remote sensing.
It can be understood that the device for monitoring the germination rate of dibbling and seeding based on remote sensing of the unmanned aerial vehicle at least comprises a memory 1, a processor 2, a communication interface 3 and a bus 4, wherein the memory 1, the processor 2 and the communication interface 3 are in communication connection with each other through the bus 4 and can complete communication with each other.
When the germination rate monitoring device is operated, the processor 2 calls the program instructions in the memory 1 to execute the methods provided by the embodiments of the methods, for example, the methods include: and calculating the total number of budding plants by counting the total number of the single plant units and the overlapping times of each overlapping plant unit, and calculating the budding rate of the target point sowing based on the total number of the budding plants and the total number of the sowing holes.
In another embodiment of the present invention, a non-transitory computer-readable storage medium is provided, which stores computer instructions for causing a computer to execute the method for monitoring the germination rate of dibbling seeds based on drone remote sensing according to the above embodiment.
It will be appreciated that the logic instructions in the memory 1 described above may be implemented in the form of software functional units and stored in a computer readable storage medium when sold or used as a stand-alone product. Alternatively, all or part of the steps of implementing the method embodiments may be implemented by hardware related to program instructions, where the program may be stored in a computer-readable storage medium, and when executed, the program performs the steps including the method embodiments; and the aforementioned storage medium includes: various media capable of storing program codes, such as a U disk, a removable hard disk, a ROM, a RAM, a magnetic disk, or an optical disk.
The above described embodiments of the apparatus for monitoring the germination rate of dibbling seedlings based on remote sensing by unmanned aerial vehicles are merely illustrative, wherein the units illustrated as separate parts may or may not be physically separated, either located in one place or distributed on different network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the scheme of the embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.
Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware. Based on such understanding, the technical solutions mentioned above may be embodied in the form of a software product, which may be stored in a computer-readable storage medium, such as a usb disk, a removable hard disk, a ROM, a RAM, a magnetic or optical disk, etc., and includes several instructions for causing a computer device (such as a personal computer, a server, or a network device, etc.) to execute the methods described in the method embodiments or some parts of the method embodiments.
According to the monitoring device for the bud ratio of the dibbling seeding based on the unmanned aerial vehicle remote sensing, provided by the embodiment of the invention, the program instruction in the memory 1 is called through the processor 2, the unmanned aerial vehicle remote sensing image data of the farmland after the dibbling seeding is processed, the number of pixels covered by a plant in a remote sensing image, the overlapping condition of the plant and the like are analyzed, the automatic and accurate estimation of the target dibbling seeding bud ratio is realized, the monitoring efficiency, the accuracy and the applicability of the crop bud ratio can be effectively improved, the consumption of manpower resources is greatly saved, and the informatization management level is improved.
In addition, it should be understood by those skilled in the art that the terms "comprises," "comprising," or any other variation thereof, in the specification of the present invention, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
In the description of the present invention, numerous specific details are set forth. However, it is understood that embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures and techniques have not been shown in detail in order not to obscure an understanding of this description. Similarly, it should be appreciated that in the foregoing description of exemplary embodiments of the invention, various features of the invention are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the disclosure and aiding in the understanding of one or more of the various inventive aspects.
However, the disclosed method should not be interpreted as reflecting an intention that: that the invention as claimed requires more features than are expressly recited in each claim. Rather, as the following claims reflect, inventive aspects lie in less than all features of a single foregoing disclosed embodiment. Thus, the claims following the detailed description are hereby expressly incorporated into this detailed description, with each claim standing on its own as a separate embodiment of this invention.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (7)
1. A method for monitoring the germination rate of dibbling and sowing based on unmanned aerial vehicle remote sensing is characterized by comprising the following steps:
s1, taking the independent image patches of the crop vegetation areas in the complete dibbling and seeding farmland images as monitoring units, and dividing overlapping plant units and single plant units by monitoring the outline shape and size characteristics of each monitoring unit; the method comprises the following steps:
according to the outline shape characteristic of a single plant of a target dibbling and seeding crop, a plant overlapping judgment model is made, and according to the outline shape and the size characteristic of each monitoring unit, each monitoring unit is judged to belong to the overlapping plant unit or the single plant unit by using the plant overlapping judgment model;
s2, determining the number of pixels covered by a single plant by using a data statistics method, and calculating the overlapping times of plants in the overlapping plant units based on the number of pixels covered by the single plant; the method comprises the following steps:
utilizing a data statistical method to make a frequency distribution percentage diagram of the number of pixels covered by a single plant, and calculating the number of pixels covered by the single plant based on the frequency distribution percentage diagram;
calculating the average contour diameter of the single plant based on the number of pixels covered by the single plant, and calculating the overlapping times of the plants based on the average contour diameter of the single plant and the maximum axial length of the overlapping plant units;
s3, calculating the total number of total plants after germination by counting the total number of the single plant units and the overlapping times of each overlapping plant unit, and calculating the germination rate of the target dibbling seeding based on the total number of the total plants after germination and the total number of the seeding dibbling; the method comprises the following steps:
and calculating the total number of the dibbling and dibbling points according to the total number of the dibbling and sowing lines, the length of each dibbling and sowing line, the dibbling and sowing speed and the time interval of the dibbling and sowing between adjacent dibbling points.
2. The method of claim 1, further comprising, before the step of S1:
s01, controlling the unmanned aerial vehicle to fly above a target dibbling and seeding farmland, and collecting remote sensing images of the target dibbling and seeding farmland in a plurality of discrete areas;
and S02, splicing the remote sensing images of the plurality of discrete region point sowing farmlands to obtain the complete point sowing farmland image, classifying the complete point sowing farmland image by using a supervision and classification method, and extracting the crop vegetation area.
3. The method of claim 2, further comprising, between the steps of S01 and S02:
carrying out quality monitoring on the remote sensing images of the plurality of discrete region point sowing farmlands, and outputting the remote sensing images of the discrete regions with quality indexes meeting set standards;
correspondingly, the step of splicing the remote sensing images of the multiple discrete region point sowing farmlands to obtain the complete point sowing farmland image in the step S02 further includes:
and splicing the remote sensing images of the discrete areas with the quality indexes meeting the set standard to obtain the complete dibbling and seeding farmland image.
4. The method of claim 3, wherein between the step of classifying the full hill-drop seeding farmland image and the step of extracting the crop vegetation zone using a supervised classification method in step S02, further comprising:
and removing isolated spots in the remote sensing image classified into the vegetation area, wherein the isolated spots represent isolated vegetation areas with the number of covered pixels smaller than a given value.
5. The method according to claim 2, wherein the step of stitching the plurality of discrete region dibbling and seeding farmland remote sensing images in step S02 to obtain the complete dibbling and seeding farmland image further comprises:
inputting the multiple discrete region point sowing farmland remote sensing images into Agisoft Scan software, and outputting orthographic remote sensing images containing geographic coordinates as the complete point sowing farmland images after splicing processing.
6. The method of claim 5, wherein the step of S02 further comprises: extracting a naked area;
after the step of S02, the method further includes:
and the crop vegetation area and the bare area are distinguished and displayed according to the geographic coordinates, and a reseeding area is indicated according to a display result.
7. The utility model provides a dibbling seeding germination percentage monitoring devices based on unmanned aerial vehicle remote sensing which characterized in that includes: at least one memory, at least one processor, a communication interface, and a bus;
the memory, the processor and the communication interface complete mutual communication through the bus, and the communication interface is used for information transmission between the monitoring device and the unmanned aerial vehicle data communication interface and the human-computer interaction interface;
the memory has stored therein a computer program operable on the processor, which when executed implements the method of any of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810305437.4A CN108537164B (en) | 2018-04-08 | 2018-04-08 | Method and device for monitoring germination rate of dibbling and sowing based on unmanned aerial vehicle remote sensing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810305437.4A CN108537164B (en) | 2018-04-08 | 2018-04-08 | Method and device for monitoring germination rate of dibbling and sowing based on unmanned aerial vehicle remote sensing |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108537164A CN108537164A (en) | 2018-09-14 |
| CN108537164B true CN108537164B (en) | 2020-08-28 |
Family
ID=63482853
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810305437.4A Active CN108537164B (en) | 2018-04-08 | 2018-04-08 | Method and device for monitoring germination rate of dibbling and sowing based on unmanned aerial vehicle remote sensing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108537164B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11580729B2 (en) * | 2019-11-22 | 2023-02-14 | Intelinair, Inc. | Agricultural pattern analysis system |
| CN112560833B (en) * | 2021-03-01 | 2021-05-11 | 广州汇图计算机信息技术有限公司 | Information display system based on remote sensing image |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9058633B2 (en) * | 2010-10-25 | 2015-06-16 | Trimble Navigation Limited | Wide-area agricultural monitoring and prediction |
| CN105046909A (en) * | 2015-06-17 | 2015-11-11 | 中国计量学院 | Agricultural loss assessment assisting method based on small-sized unmanned aerial vehicle |
| CN106971167A (en) * | 2017-03-30 | 2017-07-21 | 北京兴农丰华科技有限公司 | Crop growth analysis method and its analysis system based on unmanned aerial vehicle platform |
| CN107148633A (en) * | 2014-08-22 | 2017-09-08 | 克莱米特公司 | Method for carrying out agronomy and agricultural monitoring using UAS |
| CN107493920A (en) * | 2017-09-26 | 2017-12-22 | 界首市俊良种植专业合作社 | A kind of cultural method for peanut high yield |
-
2018
- 2018-04-08 CN CN201810305437.4A patent/CN108537164B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9058633B2 (en) * | 2010-10-25 | 2015-06-16 | Trimble Navigation Limited | Wide-area agricultural monitoring and prediction |
| CN107148633A (en) * | 2014-08-22 | 2017-09-08 | 克莱米特公司 | Method for carrying out agronomy and agricultural monitoring using UAS |
| CN105046909A (en) * | 2015-06-17 | 2015-11-11 | 中国计量学院 | Agricultural loss assessment assisting method based on small-sized unmanned aerial vehicle |
| CN106971167A (en) * | 2017-03-30 | 2017-07-21 | 北京兴农丰华科技有限公司 | Crop growth analysis method and its analysis system based on unmanned aerial vehicle platform |
| CN107493920A (en) * | 2017-09-26 | 2017-12-22 | 界首市俊良种植专业合作社 | A kind of cultural method for peanut high yield |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108537164A (en) | 2018-09-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108647652B (en) | Cotton development period automatic identification method based on image classification and target detection | |
| CN106971167B (en) | Crop growth analysis method and system based on unmanned aerial vehicle platform | |
| CN110163138B (en) | Method for measuring and calculating wheat tillering density based on multispectral remote sensing image of unmanned aerial vehicle | |
| CN107274384B (en) | Wheat seedling emergence uniformity determination method | |
| CN110909679B (en) | Remote sensing identification method and system for fallow crop rotation information of winter wheat historical planting area | |
| CN113469112B (en) | Crop growth condition image identification method and system | |
| CN113011221A (en) | Crop distribution information acquisition method and device and measurement system | |
| CN108537164B (en) | Method and device for monitoring germination rate of dibbling and sowing based on unmanned aerial vehicle remote sensing | |
| CN102592118B (en) | Automatic detection method for time emergence of seedling of corns | |
| CN114359746A (en) | Machine learning multispectral remote sensing image crop straw field-leaving extraction method and system | |
| CN106683069A (en) | Method for recognizing inline crops and weeds in seedling stage of farmland | |
| CN115619286B (en) | Method and system for evaluating quality of sample plot of breeding field district | |
| CN117036926A (en) | Weed identification method integrating deep learning and image processing | |
| CN115620151A (en) | Method and device for identifying phenological period, electronic equipment and storage medium | |
| CN115424151A (en) | Agricultural intelligent platform based on image processing | |
| Cui et al. | Phenotypic characteristics of ramie (Boehmeria nivea L) germplasm resources based on UAV remote sensing | |
| CN116823918B (en) | Crop seedling number measuring method, device, electronic equipment and storage medium | |
| CN115660236B (en) | Crop weather period prediction method, device, electronic equipment and storage medium | |
| Manfrini et al. | Development of precision apple fruit growing techniques: monitoring strategies for yield and high-quality fruit production | |
| CN116453003B (en) | Method and system for intelligently identifying rice growth vigor based on unmanned aerial vehicle monitoring | |
| US20230401703A1 (en) | Apparatus, systems and methods for image plant counting | |
| US11393193B1 (en) | Zone management generation from point samples | |
| US11915421B2 (en) | Auditing task performance | |
| CN116168292A (en) | Soybean seedling emergence uniformity judging method and system | |
| CN116543319A (en) | Evaluation method for planting characteristics of sugarcane transverse planter |
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 |