CN113011220A - Spike number identification method and device, storage medium and processor - Google Patents
Spike number identification method and device, storage medium and processor Download PDFInfo
- Publication number
- CN113011220A CN113011220A CN201911319103.3A CN201911319103A CN113011220A CN 113011220 A CN113011220 A CN 113011220A CN 201911319103 A CN201911319103 A CN 201911319103A CN 113011220 A CN113011220 A CN 113011220A
- Authority
- CN
- China
- Prior art keywords
- area
- target area
- image
- crop
- target
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 66
- 238000003860 storage Methods 0.000 title claims abstract description 17
- 210000005069 ears Anatomy 0.000 claims abstract description 63
- 235000013399 edible fruits Nutrition 0.000 claims abstract description 35
- 238000012549 training Methods 0.000 claims abstract description 19
- 238000005259 measurement Methods 0.000 claims abstract description 9
- 241000196324 Embryophyta Species 0.000 description 49
- 238000001914 filtration Methods 0.000 description 26
- 241001024327 Oenanthe <Aves> Species 0.000 description 24
- 230000008569 process Effects 0.000 description 21
- 241000209140 Triticum Species 0.000 description 20
- 235000021307 Triticum Nutrition 0.000 description 18
- 238000012545 processing Methods 0.000 description 17
- 238000010586 diagram Methods 0.000 description 9
- 241000495841 Oenanthe oenanthe Species 0.000 description 8
- 238000013135 deep learning Methods 0.000 description 6
- 235000013339 cereals Nutrition 0.000 description 5
- 238000013507 mapping Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 4
- 238000002372 labelling Methods 0.000 description 4
- 238000013528 artificial neural network Methods 0.000 description 3
- 238000013527 convolutional neural network Methods 0.000 description 3
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 238000011946 reduction process Methods 0.000 description 3
- 238000004891 communication Methods 0.000 description 2
- 238000009313 farming Methods 0.000 description 2
- 238000009499 grossing Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 230000011218 segmentation Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/188—Vegetation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Image Analysis (AREA)
Abstract
The application discloses a spike number identification method, a spike number identification device, a storage medium and a processor. Wherein, the method comprises the following steps: acquiring an image of a target area; inputting the image into a preset learning model for analysis to obtain the cluster distribution information of plants in the target area, wherein the preset learning model is obtained by training a plurality of groups of data, and each group of data comprises: image information for each plant, and a label for marking ears in each plant; and determining the number of the fruit clusters of the plants in the target area based on the distribution information of the fruit clusters. The method and the device solve the technical problems that in the related art, the counting speed of the number of the clusters is low and the yield measurement precision is low.
Description
Technical Field
The application relates to the field of crop yield prediction, in particular to a spike number identification method, a spike number identification device, a storage medium and a processor.
Background
The existing grain yield statistical method is extensive, low in measurement accuracy and difficult to apply to large farms due to the fact that the statistics needs a large amount of workload, takes long time and is difficult to finish uniformly at the optimal time point, and a large amount of manual work is needed for basic statistics.
In view of the above problems, no effective solution has been proposed.
Disclosure of Invention
The embodiment of the application provides an ear number identification method, an ear number identification device, a storage medium and a processor, and aims to at least solve the technical problems that the counting speed of the number of ears is low and the measurement precision of yield is low in the related technology.
According to an aspect of the embodiments of the present application, there is provided an ear number identification method, including: acquiring an image of a target area; inputting the image into a preset learning model for analysis to obtain the cluster distribution information of plants in the target area, wherein the preset learning model is obtained by training a plurality of groups of data, and each group of data comprises: image information for each plant, and a label for marking ears in each plant; and determining the number of the fruit clusters of the plants in the target area based on the distribution information of the fruit clusters.
Optionally, the method further comprises: acquiring the number of clusters in a unit area in a target area and the average mass of the reference cluster number; determining the crop yield of the target area based on the average quality of the number of fruit ears per unit area and the reference ear number.
Optionally, acquiring the number of ears per unit area in the target region includes: acquiring image information of a plurality of crop areas; respectively inputting the image information of the plurality of crop areas into a preset learning model for analysis to obtain the cluster distribution information of the plurality of crop areas; determining the quantity of ears in each crop area in the plurality of crop areas based on the ear distribution information; and determining the number of fruit ears in unit area in the target area according to the number of fruit ears in each crop area and the area of each crop area.
Optionally, after determining the crop yield of the target area based on the average quality of the number of fruit ears per unit area and the reference ear number, the method further comprises: adding crop yield of a target area into an electronic map, wherein the electronic map is used for displaying the yield of at least one crop area; and displaying the crop yield of the target area in the electronic map.
Optionally, adding crop yield of the target area to the electronic map comprises: creating a target map layer in an electronic map; and adding the crop yield of the target area into the target image layer, and setting the display attribute of the crop yield of the target area.
Optionally, the method further includes: detecting a frame selection area of a target object in an electronic map; determining the occupation ratio of the frame selection area in the target area, wherein the occupation ratio is used for indicating the ratio of the area of the frame selection area to the area of the target area; and determining the yield of the target crops in the selected area according to the proportion and the crop yield of the target area.
Optionally, the label in the plurality of sets of training data is a label for marking a center point of the ear.
According to another aspect of embodiments of the present application, there is provided a measurement system including: the unmanned aerial vehicle is used for acquiring an image of a target area; the network side equipment is used for inputting the image into a preset learning model for analysis to obtain the cluster distribution information of plants in the target area, wherein the preset learning model is obtained through training of multiple groups of data, and each group of data comprises: image information for each plant, and a label for marking ears in each plant; and determining the number of the fruit clusters of the plants in the target area based on the distribution information of the fruit clusters.
According to another aspect of the embodiments of the present application, there is provided an ear number identification apparatus including: the acquisition module is used for acquiring an image of a target area; the analysis module is used for inputting the image into a preset learning model for analysis to obtain the cluster distribution information of plants in the target area, wherein the preset learning model is obtained by training a plurality of groups of data, and each group of data comprises: image information for each plant, and a label for marking ears in each plant; and the determining module is used for determining the number of the fruit clusters of the plants in the target area based on the distribution information of the fruit clusters.
According to another aspect of the embodiments of the present application, there is provided a non-volatile storage medium including a stored program, wherein the apparatus in which the non-volatile storage medium is controlled to execute the above-mentioned spike number identification method when the program runs
In the embodiment of the application, the spike number identification method comprises the following steps: acquiring an image of a target area; inputting the image into a preset learning model for analysis to obtain the cluster distribution information of plants in the target area, wherein the preset learning model is obtained by training a plurality of groups of data, and each group of data comprises: image information for each plant, and a label for marking ears in each plant; and determining the number of the fruit clusters of the plants in the target area based on the distribution information of the fruit clusters. . Based on the scheme, the preset learning model can be adopted to analyze the image of the target area, and the quantity of the clusters is counted according to the analysis result. Therefore, the automatic identification of the number of the clusters is realized, the scheme of manual statistics is replaced, the purpose of rapidly and accurately counting the number of the clusters in the target area is achieved, and the technical problems of low counting speed of the number of the clusters and low yield measurement precision in the related technology are solved.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the application and together with the description serve to explain the application and not to limit the application. In the drawings:
fig. 1 is a schematic flow chart of a spike number identification method according to an embodiment of the present application;
FIG. 2 is a schematic view of a center point of an ear according to an embodiment of the present application;
FIG. 3 is a schematic diagram of the infrastructure of a network according to an embodiment of the present application;
FIG. 4 is a schematic illustration of a central Gaussian processing according to an embodiment of the application;
FIG. 5 is a schematic diagram of a measurement system according to an embodiment of the present application;
fig. 6 is a schematic structural diagram of an ear number identification device according to an embodiment of the present application.
Detailed Description
In order to make the technical solutions better understood by those skilled in the art, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only partial embodiments of the present application, but not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
It should be noted that the terms "first," "second," and the like in the description and claims of this application and in the drawings described above are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the application described herein are capable of operation in sequences other than those illustrated or described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
First, some terms or terms appearing in the description of the embodiments of the present application are applicable to the following explanations:
gaussian filtering: the method is linear smooth filtering, is suitable for eliminating Gaussian noise, and is widely applied to the noise reduction process of image processing. Generally speaking, gaussian filtering is a process of performing weighted average on the whole image, and the value of each pixel point is obtained by performing weighted average on the value of each pixel point and other pixel values in the neighborhood. The specific operation of gaussian filtering is: each pixel in the image is scanned with a template (or convolution, mask), and the weighted average gray value of the pixels in the neighborhood determined by the template is used to replace the value of the pixel point in the center of the template.
In accordance with an embodiment of the present application, there is provided a method embodiment of spike number identification, it is noted that the steps illustrated in the flowchart of the figures may be performed in a computer system such as a set of computer executable instructions, and that while a logical order is illustrated in the flowchart, in some cases the steps illustrated or described may be performed in an order different than here.
Fig. 1 is a method for identifying the ear number according to an embodiment of the present application, as shown in fig. 1, the method includes the following steps:
step S102, acquiring an image of the target area.
Specifically, the target area is a plot where the ears grow. In an alternative embodiment of the present application, the ear may be an ear. There are various ways to obtain images of the growing areas of the wheat ears, for example: an image of a wheat ear growing plot can be obtained by aerial photography of a surveying and mapping unmanned aerial vehicle, the aerial image is stored in a memory, and the image is obtained from a network side when ear number identification is carried out; or the images of the wheat ear growing plots can be shot manually, and after the shooting is finished, the shot images are input into the processing module manually when the ear number is identified.
Step S104, inputting the image into a preset learning model for analysis to obtain the cluster distribution information of the plants in the target area, wherein the preset learning model is obtained by training a plurality of groups of data, and each group of data comprises: image information for each plant, and a label for marking the ears in each plant.
Specifically, the labels in the multiple sets of training data are labels for marking the center points of the ears of wheat, and the schematic diagram of labeling the center points of the ears of wheat is shown in fig. 2; the distribution information of the wheat ears of the plants can be the distribution position, density and the like of the center points of the wheat ears in the target area image; the image information of the plant may include: the location where each plant is growing, the density of the plants, etc.
Specifically, the deep learning process may be processed by using a preset learning model, and a network infrastructure of the preset learning model is shown in fig. 3, and the network infrastructure may include the following four parts: an Image unit as an input part of the network; the same convolution neural network Encoder and Decode units are used as image processing units; and an Output unit serving as an Output part of the network. The preset learning model may perform the following operations on the photographed image of the target region: firstly, inputting an RGB Image of an originally shot target area into an Image unit, then processing the RGB Image through an Encoder unit and a Decoder unit of a convolutional neural network, and finally outputting contents similar to the labels from an Output unit.
In some embodiments of the application, because the wheat ears are small, and the crossing problem exists, the number of the wheat ears is difficult to accurately count by using a traditional image algorithm, the central point of the wheat ears is directly segmented by adopting a deep learning mode in the scheme, and then the number of the wheat ears is counted. Firstly, marking the center point of each wheat plant, marking the points by taking the center of the wheat ear as a standard, and performing Gaussian filtering processing on the center point in order to complete segmentation for deep learning. The central point labeling schematic diagram of the ear is subjected to gaussian filtering, and the obtained central gaussian processing schematic diagram is shown in fig. 4.
And S106, determining the number of the fruit clusters of the plants in the target area based on the distribution information of the fruit clusters.
In some embodiments of the present application, before determining the ear number of the plant in the target area based on the ear distribution information, the method may further perform a filtering process on the ears in the ear distribution information, and obtain the target distribution information after the filtering process. The filtering process may be a gaussian filtering process in an alternative embodiment of the present application.
Specifically, gaussian filtering is a linear smoothing filter, is suitable for eliminating gaussian noise, and is widely applied to a noise reduction process of image processing. Generally speaking, gaussian filtering is a process of performing weighted average on the whole image, and the value of each pixel point is obtained by performing weighted average on the value of each pixel point and other pixel values in the neighborhood. The specific operation of gaussian filtering is: each pixel in the image is scanned with a template (or convolution, mask), and the weighted average gray value of the pixels in the neighborhood determined by the template is used to replace the value of the pixel point in the center of the template.
In some embodiments of the present application, the method further comprises: and acquiring the number of wheat ears per unit area and the average quality of the reference ear number in the target area, and determining the crop yield of the target area according to the number of the wheat ears per unit area and the average quality of the reference ear number.
Specifically, the average mass of the reference ear number may be the mass of one thousand ears, and the following manner may be adopted when calculating the average mass of the reference ear number: for example: five thousand ears are obtained from a plant, the five thousand ears are averagely divided into five groups, the masses of the five groups of ears are respectively weighed, and then the average value of the masses of the five groups of ears is taken as the average mass of the reference ear number.
In some embodiments of the present application, obtaining the number of ears per unit area in the target region comprises: the method comprises the steps of obtaining image information of a plurality of crop areas, inputting the image information of the crop areas into a preset learning model for analysis, obtaining ear distribution information of the crop areas after the images of the crop areas are analyzed by the preset model, determining the number of ears of each crop area in the crop areas according to the ear distribution information, and determining the number of ears of each crop area in a target area according to the number of ears of each crop area and the area of each crop area.
The determination of the number of ears per unit area in the target region is exemplified by, for example: knowing that the total area of each crop area is 5000 square meters, a user knows that the quantity of wheat ears in each crop area in a plurality of crop areas is 3000000 according to the wheat ear distribution information, and according to a formula: when 3000000/5000 is 600, the number of wheat ears per unit area in the target region is 600.
In some embodiments of the present application, crop yield for a target area is determined based on the average quality of the number of ears per unit area and the reference ear number, for example: the average mass of the reference ear number is 1.25kg, the average ear number per unit area is 600, the total area of the target area is 5000 square meters, and the formula is as follows: (600 × 5000/1000) × 1.25 ═ 3750, and the total ear yield in the target region was 3750 kg.
In some embodiments of the present application, after determining the crop yield of the target area based on the average quality of the number of wheat ears per unit area and the reference ear number, the method may further comprise the steps of: and adding the crop yield of the target area into the electronic map, and displaying the crop yield of the target area in the electronic map. Wherein, the electronic map is used for showing the yield of at least one crop area. After the yield of the object area is displayed on the electronic map, the yield situation of each object area can be visually observed, and the yield situations of different object areas can be compared.
Specifically, adding crop yield of the target area to the electronic map may include: creating a target map layer in the electronic map, then adding the crop yield of the target area into the target map layer, and setting the display attribute of the crop yield of the target area in the target map layer.
The crop yield of the target area can be displayed in various ways in the target map layer, for example: the yield can be directly displayed in the target layer in a digital form; or the shade of the color of the graph layer is used for representing the yield of the crops, such as: the yield of crops is represented by a red layer, the depth of the color of the layer is endowed with a degree value, and the color of the layer of the yield of the crops is darker along with the increase of the yield.
In some embodiments of the present application, the method further comprises: firstly, detecting a frame selection area of a target object in an electronic map, then determining the proportion of the frame selection area in the target area, wherein the proportion is used for indicating the ratio of the area of the frame selection area to the area of the target area, and finally determining the wheat yield in the frame selection area according to the proportion and the wheat yield in the target area.
Specifically, if the user wants to count the wheat yield of any region in the target region, the user needs to select the region to be counted from the target region, then calculate the ratio of the area of the selected region to the area of the target region, and calculate the wheat yield of the selected region according to the ratio and the total wheat yield of the target region. However, because the growth states of wheat in different regions may be different, when the area of the boxed region is smaller, the yield of wheat in the boxed region calculated by the method may have a larger error. In order to reduce the error, a threshold value can be set, namely when the ratio of the area of the selected area to the area of the target area is larger than the threshold value, the method can be selected to calculate the yield of the wheat in the selected area; when the ratio of the area of the frame selection area to the area of the target area is smaller than the threshold value, the method is not recommended to be used for calculating the yield of the wheat in the frame selection area, and the yield of the wheat in the frame selection area can be determined according to the number of fruit ears in unit area, the average quality of the reference ear number and the area of the frame selection area. Here, the size of the threshold is determined according to the area of the target region, and the threshold is different depending on the area of the target region.
For example: assuming that the area of the target region is 5000 square meters, the yield of the target region is 3750kg, and the threshold value is set to 0.01. When the area of the boxed area is 100 square meters, the ratio of the area of the boxed area to the area of the target area is 0.02, and since 0.02 is greater than 0.01, the yield of wheat in the boxed area can be calculated according to the ratio and the total yield of wheat in the target area, namely 3750 x 0.02 is 75 kg. When the area of the frame selection area is 40 square meters, the area ratio is 0.008, and since 0.008 is less than 0.01, the wheat yield of the frame selection area can be determined according to the number of the fruit ears in unit area, the average quality of the reference ear number and the area of the frame selection area, namely the wheat yield of the frame selection area is 50 kg.
In some embodiments of this application, after the model training is accomplished, can be through survey and drawing unmanned aerial vehicle's low-altitude flight random sampling a large amount of farmland pictures, and then obtain the ear of grain number of unit area, and then the average quality of rethread ear of grain number and then obtain the output of whole plot.
Based on above-mentioned process, after surveying and mapping unmanned aerial vehicle took the high definition image, can predict the output of each farming region, generate the output map, can conveniently trace back the crops region like this, and then discover the careless neglect that may exist of planting management link, help peasant household to accumulate the planting management experience better.
Through the steps, the images of the target area can be analyzed by adopting the preset learning model, and the quantity of the ears is counted according to the analysis result. Therefore, the automatic identification of the number of the clusters is realized, the scheme of manual statistics is replaced, the purpose of rapidly and accurately counting the number of the clusters in the target area is achieved, and the technical problems of low counting speed of the number of the clusters and low yield measurement precision in the related technology are solved.
Fig. 5 is a measurement system according to an embodiment of the present application, as shown in fig. 5, the system including:
the unmanned aerial vehicle 1 is used for acquiring the image of the target area 2.
Specifically, the target area 2 is a land where the ears grow. In an alternative embodiment of the present application, the ear may be an ear. The image of the wheat ear growing plot is obtained by utilizing aerial photography of the surveying and mapping unmanned aerial vehicle, the aerial image is stored in a storage, and the image is obtained from a network side when ear number identification is carried out.
Network side equipment 3 for input image to preset learning model and carry out analysis, obtain the ear distribution information of plant in the target area, wherein, preset learning model obtains through multiunit data training, all includes in every group data: image information for each plant, and a label for marking ears in each plant; and determining the number of the fruit clusters of the plants in the target area based on the distribution information of the fruit clusters.
Specifically, as shown in fig. 3, the network infrastructure of the preset learning model may include the following four parts: an Image information (Image) unit as an input part of the network; a co-convolution neural network Encoder (Encoder) and Decoder (Decoder) unit as a processing unit for the image; an Output (Output) unit as an Output part of the network. The preset learning model may perform the following operations on the photographed image of the target region: firstly, inputting an RGB Image of an originally shot target area into an Image unit, then processing the RGB Image through an Encoder unit and a Decoder unit of a convolutional neural network, and finally outputting contents similar to the labels from an Output unit.
In some embodiments of the present application, before determining the ear number of the plant in the target area based on the ear distribution information, the method may further perform a filtering process on the ears in the ear distribution information, and obtain the target distribution information after the filtering process. The filtering process may be a gaussian filtering process in an alternative embodiment of the present application.
In some embodiments of the present application, the method further comprises: and acquiring the number of wheat ears per unit area and the average quality of the reference ear number in the target area, and determining the crop yield of the target area according to the number of the wheat ears per unit area and the average quality of the reference ear number.
In some embodiments of the present application, obtaining the number of ears per unit area in the target region comprises: the method comprises the steps of obtaining image information of a plurality of crop areas, inputting the image information of the crop areas into a preset learning model for analysis, obtaining ear distribution information of the crop areas after the images of the crop areas are analyzed by the preset model, determining the number of ears of each crop area in the crop areas according to the ear distribution information, and determining the number of ears of each crop area in a target area according to the number of ears of each crop area and the area of each crop area.
The preferred embodiment in this embodiment may refer to the description of fig. 1, and is not described herein again.
Fig. 6 is a device for identifying the number of ears according to an embodiment of the present application, as shown in fig. 6, the device including:
an acquiring module 60 is configured to acquire an image of the target area.
Specifically, the target area is a land where the ear grows. In an alternative embodiment of the present application, the ear may be an ear. There are various ways to obtain images of the growing areas of the wheat ears, for example: an image of a wheat ear growing plot can be obtained by aerial photography of a surveying and mapping unmanned aerial vehicle, the aerial image is stored in a memory, and the image is obtained from a network side when ear number identification is carried out; or the images of the wheat ear growing plots can be shot manually, and after the shooting is finished, the shot images are input into the processing module manually when the ear number is identified.
The analysis module 62 is configured to input the image into a preset learning model for analysis, so as to obtain the ear distribution information of the plant in the target region, where the preset learning model is obtained through training of multiple sets of data, and each set of data includes: image information for each plant, and a label for marking the ears in each plant.
Specifically, the labels in the multiple sets of training data are labels for marking the center points of the ears of wheat, and the schematic diagram of labeling the center points of the ears of wheat is shown in fig. 2; the distribution information of the wheat ears of the plants can be the distribution position, density and the like of the center points of the wheat ears in the target area image; the image information of the plant may include: the location where each plant is growing, the density of the plants, etc.
Specifically, the deep learning process may be processed by using a preset learning model, and a network infrastructure of the preset learning model is shown in fig. 3, and the network infrastructure may include the following four parts: an Image unit as an input part of the network; the same convolution neural network Encoder and Decode units are used as image processing units; and an Output unit serving as an Output part of the network. The preset learning model may perform the following operations on the photographed image of the target region: firstly, inputting an RGB Image of an originally shot target area into an Image unit, then processing the RGB Image through an Encoder unit and a Decoder unit of a convolutional neural network, and finally outputting contents similar to the labels from an Output unit.
In some embodiments of the application, because the wheat ears are small, and the crossing problem exists, the number of the wheat ears is difficult to accurately count by using a traditional image algorithm, the central point of the wheat ears is directly segmented by adopting a deep learning mode in the scheme, and then the number of the wheat ears is counted. Firstly, marking the center point of each wheat plant, marking the points by taking the center of the wheat ear as a standard, and performing Gaussian filtering processing on the center point in order to complete segmentation for deep learning. The central point labeling schematic diagram of the ear is subjected to gaussian filtering, and the obtained central gaussian processing schematic diagram is shown in fig. 4.
And the determining module 64 is used for determining the number of the fruit clusters of the plants in the target area based on the distribution information of the fruit clusters.
In some embodiments of the present application, before determining the ear number of the plant in the target area based on the ear distribution information, the method may further perform a filtering process on the ears in the ear distribution information, and obtain the target distribution information after the filtering process. The filtering process may be a gaussian filtering process in an alternative embodiment of the present application.
Specifically, gaussian filtering is a linear smoothing filter, is suitable for eliminating gaussian noise, and is widely applied to a noise reduction process of image processing. Generally speaking, gaussian filtering is a process of performing weighted average on the whole image, and the value of each pixel point is obtained by performing weighted average on the value of each pixel point and other pixel values in the neighborhood. The specific operation of gaussian filtering is: each pixel in the image is scanned with a template (or convolution, mask), and the weighted average gray value of the pixels in the neighborhood determined by the template is used to replace the value of the pixel point in the center of the template.
In some embodiments of the present application, the method further comprises: and acquiring the number of wheat ears per unit area and the average quality of the reference ear number in the target area, and determining the crop yield of the target area according to the number of the wheat ears per unit area and the average quality of the reference ear number.
Specifically, the average mass of the reference ear number may be the mass of one thousand ears, and the following manner may be adopted when calculating the average mass of the reference ear number: for example: five thousand ears are obtained from a plant, the five thousand ears are averagely divided into five groups, the masses of the five groups of ears are respectively weighed, and then the average value of the masses of the five groups of ears is taken as the average mass of the reference ear number.
In some embodiments of the present application, obtaining the number of ears per unit area in the target region comprises: the method comprises the steps of obtaining image information of a plurality of crop areas, inputting the image information of the crop areas into a preset learning model for analysis, obtaining ear distribution information of the crop areas after the images of the crop areas are analyzed by the preset model, determining the number of ears of each crop area in the crop areas according to the ear distribution information, and determining the number of ears of each crop area in a target area according to the number of ears of each crop area and the area of each crop area.
In some embodiments of the present application, after determining the crop yield of the target area based on the average quality of the number of wheat ears per unit area and the reference ear number, the method may further comprise the steps of: and adding the crop yield of the target area into the electronic map, and displaying the crop yield of the target area in the electronic map. Wherein, the electronic map is used for showing the yield of at least one crop area. After the yield of the object area is displayed on the electronic map, the yield situation of each object area can be visually observed, and the yield situations of different object areas can be compared.
Specifically, adding crop yield of the target area to the electronic map may include: creating a target map layer in the electronic map, then adding the crop yield of the target area into the target map layer, and setting the display attribute of the crop yield of the target area in the target map layer.
The crop yield of the target area can be displayed in various ways in the target map layer, for example: the yield can be directly displayed in the target layer in a digital form; or the shade of the color of the graph layer is used for representing the yield of the crops, such as: the yield of crops is represented by a red layer, the depth of the color of the layer is endowed with a degree value, and the color of the layer of the yield of the crops is darker along with the increase of the yield.
In some embodiments of this application, after the model training is accomplished, can be through survey and drawing unmanned aerial vehicle's low-altitude flight random sampling a large amount of farmland pictures, and then obtain the ear of grain number of unit area, and then the average quality of rethread ear of grain number and then obtain the output of whole plot.
Based on above-mentioned process, after surveying and mapping unmanned aerial vehicle took the high definition image, can predict the output of each farming region, generate the output map, can conveniently trace back the crops region like this, and then discover the careless neglect that may exist of planting management link, help peasant household to accumulate the planting management experience better.
For the preferred embodiment of this embodiment, reference may be made to the description related to fig. 1 and fig. 5, which is not described herein again.
According to still another aspect of an embodiment of the present application, there is provided a nonvolatile storage medium including a stored program, the nonvolatile storage medium being configured to run the program that realizes:
acquiring an image of a target area; inputting the image into a preset learning model for analysis to obtain the cluster distribution information of plants in the target area, wherein the preset learning model is obtained by training a plurality of groups of data, and each group of data comprises: image information for each plant, and a label for marking ears in each plant; and determining the number of the fruit clusters of the plants in the target area based on the distribution information of the fruit clusters.
According to still another aspect of embodiments of the present application, there is provided a processor for executing a program stored in a storage medium, the processor being configured to execute the program for:
acquiring an image of a target area; inputting the image into a preset learning model for analysis to obtain the cluster distribution information of plants in the target area, wherein the preset learning model is obtained by training a plurality of groups of data, and each group of data comprises: image information for each plant, and a label for marking ears in each plant; and determining the number of the fruit clusters of the plants in the target area based on the distribution information of the fruit clusters.
The above-mentioned serial numbers of the embodiments of the present application are merely for description and do not represent the merits of the embodiments.
In the above embodiments of the present application, the descriptions of the respective embodiments have respective emphasis, and for parts that are not described in detail in a certain embodiment, reference may be made to related descriptions of other embodiments.
In the embodiments provided in the present application, it should be understood that the disclosed technology can be implemented in other ways. The above-described embodiments of the apparatus are merely illustrative, and for example, the division of the units may be a logical division, and in actual implementation, there may be another division, for example, multiple units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, units or modules, and may be in an electrical or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.
The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application may be substantially implemented or contributed to by the prior art, or all or part of the technical solution may be embodied in a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a removable hard disk, a magnetic or optical disk, and other various media capable of storing program codes.
The foregoing is only a preferred embodiment of the present application and it should be noted that those skilled in the art can make several improvements and modifications without departing from the principle of the present application, and these improvements and modifications should also be considered as the protection scope of the present application.
Claims (10)
1. A spike number identification method is characterized by comprising the following steps:
acquiring an image of a target area;
inputting the image into a preset learning model for analysis to obtain the cluster distribution information of the plants in the target area, wherein the preset learning model is obtained by training a plurality of groups of data, and each group of data comprises: image information of each plant, and a label for marking ears in said each plant;
and determining the number of the fruit clusters of the plants in the target area based on the distribution information of the fruit clusters.
2. The method of claim 1, further comprising:
acquiring the number of clusters in unit area in the target area and the average mass of the reference cluster number;
determining the crop yield of the target area based on the number of fruit ears per unit area and the average quality of the reference ear number.
3. The method of claim 2, wherein obtaining the number of ears per unit area in the target region comprises:
acquiring image information of a plurality of crop areas;
respectively inputting the image information of the plurality of crop areas to the preset learning model for analysis to obtain the cluster distribution information of the plurality of crop areas;
determining a quantity of ears for each crop area of the plurality of crop areas based on the ear distribution information;
and determining the number of the fruit ears in the unit area in the target area according to the number of the fruit ears in each crop area and the area of each crop area.
4. The method of claim 2, wherein after determining the crop yield for the target area based on the average quality of the number of ears per unit area and the reference ear number, the method further comprises:
adding the crop yield of the target area into an electronic map, wherein the electronic map is used for displaying the yield of at least one crop area;
displaying crop yield of the target area in the electronic map.
5. The method of claim 4, wherein adding crop yield of the target area to an electronic map comprises:
creating a target map layer in the electronic map;
and adding the crop yield of the target area into the target image layer, and setting the display attribute of the crop yield of the target area.
6. The method of claim 4, further comprising:
detecting a frame selection area of a target object in the electronic map;
determining the proportion of the selected area in the target area, wherein the proportion is used for indicating the ratio of the area of the selected area to the area of the target area;
and determining the yield of the target crop in the frame selection area according to the proportion and the crop yield of the target area.
7. The method of any one of claims 1 to 6, wherein the labels in the plurality of sets of training data are labels for marking the center point of the ear.
8. A measurement system, comprising:
the unmanned aerial vehicle is used for acquiring an image of a target area;
the network side equipment is used for inputting the image into a preset learning model for analysis to obtain the cluster distribution information of the plants in the target area, wherein the preset learning model is obtained by training a plurality of groups of data, and each group of data comprises: image information of each plant, and a label for marking ears in said each plant; and determining the number of the fruit clusters of the plants in the target area based on the distribution information of the fruit clusters.
9. An ear number identification device, comprising:
the acquisition module is used for acquiring an image of a target area;
the analysis module is used for inputting the image into a preset learning model for analysis to obtain the cluster distribution information of the plants in the target area, wherein the preset learning model is obtained by training a plurality of groups of data, and each group of data comprises: image information of each plant, and a label for marking ears in said each plant;
and the determining module is used for determining the number of the fruit clusters of the plants in the target area based on the distribution information of the fruit clusters.
10. A non-volatile storage medium, characterized in that the storage medium comprises a stored program, wherein when the program runs, the device where the storage medium is located is controlled to execute the spike number identification method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911319103.3A CN113011220A (en) | 2019-12-19 | 2019-12-19 | Spike number identification method and device, storage medium and processor |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911319103.3A CN113011220A (en) | 2019-12-19 | 2019-12-19 | Spike number identification method and device, storage medium and processor |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113011220A true CN113011220A (en) | 2021-06-22 |
Family
ID=76381276
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911319103.3A Pending CN113011220A (en) | 2019-12-19 | 2019-12-19 | Spike number identification method and device, storage medium and processor |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113011220A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113628186A (en) * | 2021-08-09 | 2021-11-09 | 海南青峰生物科技有限公司 | System for measuring rice spike number based on 5G communication and image recognition |
| WO2023072018A1 (en) * | 2021-10-26 | 2023-05-04 | 中国科学院空天信息创新研究院 | Wheat yield observation method based on computer vision and deep learning techniques |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170161560A1 (en) * | 2014-11-24 | 2017-06-08 | Prospera Technologies, Ltd. | System and method for harvest yield prediction |
| CN109145848A (en) * | 2018-08-30 | 2019-01-04 | 西京学院 | A kind of wheat head method of counting |
| US20190066234A1 (en) * | 2017-08-28 | 2019-02-28 | The Climate Corporation | Crop disease recognition and yield estimation |
| CN109409275A (en) * | 2018-10-18 | 2019-03-01 | 广州极飞科技有限公司 | The recognition methods of target object and device, the determination method for being administered information |
| CN109445457A (en) * | 2018-10-18 | 2019-03-08 | 广州极飞科技有限公司 | Determination method, the control method and device of unmanned vehicle of distributed intelligence |
| CN109523509A (en) * | 2018-09-28 | 2019-03-26 | 江苏省无线电科学研究所有限公司 | Detection method, device and the electronic equipment of wheat heading stage |
| CN109740721A (en) * | 2018-12-19 | 2019-05-10 | 中国农业大学 | Wheat head method of counting and device |
| CN109843034A (en) * | 2016-10-19 | 2019-06-04 | 巴斯夫农化商标有限公司 | Production forecast for wheatland |
| CN110569747A (en) * | 2019-08-20 | 2019-12-13 | 南京农业大学 | method for rapidly counting rice ears of paddy field rice by using image pyramid and fast-RCNN |
-
2019
- 2019-12-19 CN CN201911319103.3A patent/CN113011220A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170161560A1 (en) * | 2014-11-24 | 2017-06-08 | Prospera Technologies, Ltd. | System and method for harvest yield prediction |
| CN109843034A (en) * | 2016-10-19 | 2019-06-04 | 巴斯夫农化商标有限公司 | Production forecast for wheatland |
| US20190066234A1 (en) * | 2017-08-28 | 2019-02-28 | The Climate Corporation | Crop disease recognition and yield estimation |
| CN109145848A (en) * | 2018-08-30 | 2019-01-04 | 西京学院 | A kind of wheat head method of counting |
| CN109523509A (en) * | 2018-09-28 | 2019-03-26 | 江苏省无线电科学研究所有限公司 | Detection method, device and the electronic equipment of wheat heading stage |
| CN109409275A (en) * | 2018-10-18 | 2019-03-01 | 广州极飞科技有限公司 | The recognition methods of target object and device, the determination method for being administered information |
| CN109445457A (en) * | 2018-10-18 | 2019-03-08 | 广州极飞科技有限公司 | Determination method, the control method and device of unmanned vehicle of distributed intelligence |
| CN109740721A (en) * | 2018-12-19 | 2019-05-10 | 中国农业大学 | Wheat head method of counting and device |
| CN110569747A (en) * | 2019-08-20 | 2019-12-13 | 南京农业大学 | method for rapidly counting rice ears of paddy field rice by using image pyramid and fast-RCNN |
Non-Patent Citations (1)
| Title |
|---|
| 李毅念 等: "基于小麦群体图像的田间麦穗计数及产量预测方法", 《农业工程学报》, vol. 34, no. 21, 5 November 2018 (2018-11-05), pages 185 - 194 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113628186A (en) * | 2021-08-09 | 2021-11-09 | 海南青峰生物科技有限公司 | System for measuring rice spike number based on 5G communication and image recognition |
| WO2023072018A1 (en) * | 2021-10-26 | 2023-05-04 | 中国科学院空天信息创新研究院 | Wheat yield observation method based on computer vision and deep learning techniques |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110967339B (en) | Method and device for analyzing corn ear characters and corn character analysis equipment | |
| CN104200193A (en) | Fruit tree yield estimation method and device | |
| CN110427933A (en) | A kind of water gauge recognition methods based on deep learning | |
| CN110909640A (en) | Method and device for determining water level line, storage medium and electronic device | |
| CN111985724B (en) | Crop yield estimation method, device, equipment and storage medium | |
| CN113011220A (en) | Spike number identification method and device, storage medium and processor | |
| CN113011221A (en) | Crop distribution information acquisition method and device and measurement system | |
| Rahim et al. | Deep learning-based accurate grapevine inflorescence and flower quantification in unstructured vineyard images acquired using a mobile sensing platform | |
| CN110610438B (en) | Crop canopy petiole included angle calculation method and system | |
| CN113313692B (en) | Automatic banana young plant identification and counting method based on aerial visible light image | |
| CN114581464A (en) | Boundary detection method and device, electronic equipment and computer readable storage medium | |
| CN114419367A (en) | High-precision crop drawing method and system | |
| Guo et al. | Panicle Ratio Network: streamlining rice panicle measurement by deep learning with ultra-high-definition aerial images in the field | |
| CN116579521B (en) | Yield prediction time window determining method, device, equipment and readable storage medium | |
| CN113807143A (en) | Crop connected domain identification method and device and operation system | |
| CN116912265A (en) | Remote sensing image segmentation method and system | |
| CN109523509B (en) | Method and device for detecting heading stage of wheat and electronic equipment | |
| CN114662621B (en) | Agricultural machinery working area calculation method and system based on machine learning | |
| CN116739739A (en) | Loan amount evaluation method and device, electronic equipment and storage medium | |
| CN115294472A (en) | Fruit yield estimation method, model training method, equipment and storage medium | |
| CN115082812A (en) | Agricultural landscape non-agricultural habitat green patch extraction method and related equipment thereof | |
| CN114663652A (en) | Image processing method, image processing apparatus, management system, electronic device, and storage medium | |
| CN114004463A (en) | Visual intelligent agricultural big data analysis management system and method | |
| CN113807135A (en) | Seedling shortage area identification method, equipment and storage medium based on semantic segmentation network | |
| CN116844075B (en) | Tillage environment judging method and system |
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 |