CN114267005A - Harvesting method, device, processor and agricultural machine for crops - Google Patents
Harvesting method, device, processor and agricultural machine for crops Download PDFInfo
- Publication number
- CN114267005A CN114267005A CN202010967102.6A CN202010967102A CN114267005A CN 114267005 A CN114267005 A CN 114267005A CN 202010967102 A CN202010967102 A CN 202010967102A CN 114267005 A CN114267005 A CN 114267005A
- Authority
- CN
- China
- Prior art keywords
- lodging
- image
- crop
- area
- model
- 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
- 238000003306 harvesting Methods 0.000 title claims abstract description 88
- 238000000034 method Methods 0.000 title claims abstract description 70
- 230000008569 process Effects 0.000 claims abstract description 23
- 238000003860 storage Methods 0.000 claims abstract description 17
- 238000012360 testing method Methods 0.000 claims description 18
- 238000012545 processing Methods 0.000 claims description 15
- 238000002372 labelling Methods 0.000 claims description 11
- 238000005070 sampling Methods 0.000 claims description 9
- 230000003321 amplification Effects 0.000 claims description 2
- 238000003199 nucleic acid amplification method Methods 0.000 claims description 2
- 244000038559 crop plants Species 0.000 claims 1
- 239000000284 extract Substances 0.000 claims 1
- 238000010586 diagram Methods 0.000 description 19
- 238000012549 training Methods 0.000 description 17
- 238000004590 computer program Methods 0.000 description 11
- 238000005516 engineering process Methods 0.000 description 7
- 230000006870 function Effects 0.000 description 6
- 238000005520 cutting process Methods 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 3
- 240000005979 Hordeum vulgare Species 0.000 description 2
- 235000007340 Hordeum vulgare Nutrition 0.000 description 2
- 240000007594 Oryza sativa Species 0.000 description 2
- 235000007164 Oryza sativa Nutrition 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 238000005562 fading Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 235000009566 rice Nutrition 0.000 description 2
- 235000007319 Avena orientalis Nutrition 0.000 description 1
- 241000209763 Avena sativa Species 0.000 description 1
- 235000007558 Avena sp Nutrition 0.000 description 1
- 240000008620 Fagopyrum esculentum Species 0.000 description 1
- 235000009419 Fagopyrum esculentum Nutrition 0.000 description 1
- 235000021307 Triticum Nutrition 0.000 description 1
- 241000209140 Triticum Species 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000010191 image analysis Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000000877 morphologic effect Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 239000010902 straw Substances 0.000 description 1
- 238000005303 weighing Methods 0.000 description 1
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The embodiment of the invention provides a harvesting method, a harvesting device, a processor, agricultural machinery and a storage medium for crops. The method comprises the following steps: obtaining crop area images of crops in an area to be harvested in the operation process of the agricultural machinery; inputting the crop area image to a lodging model to determine a lodging state of the crop; according to the lodging state, the operation parameters of the agricultural machine for the area to be harvested are determined, the operation parameters can be timely adjusted when the agricultural machine harvests crops in the lodging area, and more intelligent, accurate and effective crop harvesting operation is achieved.
Description
Technical Field
The invention relates to the technical field of agriculture, in particular to a harvesting method, a harvesting device, a harvesting processor, agricultural machinery and a storage medium for crops.
Background
The intelligent driving and the intelligent operation are two core remarkable characteristics of the intelligent agricultural machine. In recent years, with the development of computer technology, intelligent driving technology with navigation positioning as a core has been applied to agricultural machinery, but intelligent operation related technology has not been applied yet. In the case of a harvesting machine, how to realize automatic harvesting of the lodging crops is one of the key problems that must be solved by an intelligent agricultural machine.
Currently, intelligent agricultural machines generally avoid such problems, for example, some unmanned agricultural machines are handled in the following manner: before harvesting crops, parameters of the cutting table and the reel are manually adjusted in advance, and then the parameters are kept unchanged all the time in the whole harvesting process of the unmanned agricultural machine, so that the lodging crops are obviously harvested inaccurately, and the overall harvesting quality is reduced. Some other methods are that an unmanned aerial vehicle is used for shooting a large number of pictures in a working field, then an image analysis technology, a path planning technology and the like are used for building a working map of the intelligent agricultural machine, areas of lodging crops are marked on the map, then the intelligent agricultural machine carries out field operation according to the planned path according to the map, and the operation is finished by remotely controlling a foreground agricultural machine from a background for the areas of the lodging crops on the map.
This type of agricultural machinery operation is not capable of completing an intelligent unmanned work process and still requires an experienced driver to participate in it, resulting in higher labor costs for crop harvesting.
Disclosure of Invention
The embodiment of the invention aims to provide a harvesting method, a harvesting device, a harvesting processor, an agricultural machine and a storage medium for crops, which can realize unmanned operation and reduce labor cost.
In order to achieve the above object, a first aspect of the present invention provides a harvesting method for crops, comprising:
obtaining crop area images of crops in an area to be harvested in the operation process of the agricultural machinery;
inputting the crop area image into a lodging model to determine the lodging state of the crop;
and determining the operation parameters of the agricultural machine aiming at the area to be harvested according to the lodging state.
In an embodiment of the present invention, inputting the crop area image to the lodging model to determine the lodging status of the crop comprises: inputting the crop area image into the lodging model; acquiring a prediction label output by the lodging model and corresponding to each image pixel in the crop area image; and determining the lodging state of the crop according to the prediction label.
In an embodiment of the present invention, determining the lodging status of the crop according to the predictive label comprises: selecting a sub-image from the crop area image, wherein the sub-area corresponding to the sub-image in the area to be harvested is closest to the agricultural machine; and determining the lodging state of the crop according to the prediction label corresponding to each image pixel in the selected sub-images.
In an embodiment of the present invention, determining the lodging state of the crop according to the prediction label corresponding to each image pixel in the selected sub-image comprises: acquiring the number of pixels of the image which are indicated to belong to the lodging by the prediction label in the sub-image; and determining that the crop belongs to the lodging state under the condition that the ratio of the pixel number to the total image pixel number in the sub-image is greater than a preset ratio. In an embodiment of the invention, determining the operation parameters of the agricultural machine for the area to be harvested according to the lodging state comprises the following steps: and determining operation parameters according to the ratio when the crops are determined to be in the lodging state.
In an embodiment of the present invention, the crop area image includes a plurality of crop area images for an area to be cut, which are acquired at a preset image acquisition frequency.
In an embodiment of the invention, determining the operation parameters of the agricultural machine for the area to be harvested according to the lodging state comprises the following steps: determining the actual position of a region belonging to the lodging state in the crop region image; and determining the adjustment time interval of the operation parameters according to the distance between the agricultural machine and the actual position and the moving speed of the agricultural machine.
In an embodiment of the invention, the lodging model is trained by the following steps: obtaining a plurality of crop image samples; extracting a preset area in a crop image sample; and inputting the preset area into the lodging model so as to train the lodging model.
In an embodiment of the present invention, the operational parameters include at least one of: header height, reel rotating speed, reel height, fan rotating speed and agricultural machine moving speed.
In an embodiment of the present invention, the lodging model includes at least one of a residual convolution network and a residual upsampling network; inputting the preset area into the lodging model to train the lodging model, wherein the training comprises the following steps: inputting the preset area into a residual convolution network to extract image sample characteristics in the preset area; and extracting partial image sample characteristics by the residual up-sampling network, and amplifying to train the lodging model through the image sample characteristics and the amplified image sample characteristics.
In the embodiment of the invention, the crop image sample carries the labeling information; inputting the preset area into the lodging model to train the lodging model, wherein the training comprises the following steps: inputting the preset area into the lodging model, and acquiring a test label output by the lodging model; and adjusting the model parameters of the lodging model according to the test label and the labeling information so as to train the lodging model.
A second aspect of the invention provides a processor configured to perform a harvesting method for a crop as described above.
In a third aspect the present invention provides a harvesting apparatus for a crop, for use with an agricultural machine, the harvesting apparatus comprising:
an image capture device configured to obtain crop area images of crops in an area to be harvested in front of the agricultural machine during operation of the agricultural machine; and
such as the processor described above.
A fourth aspect of the invention provides an agricultural machine comprising a harvesting apparatus for crop as described above.
A fifth aspect of the invention provides a machine-readable storage medium having instructions stored thereon which, when executed by a processor, cause the processor to be configured to perform the harvesting method for a crop described above.
Above-mentioned technical scheme is through constantly gathering the crops territory image of the crops in the area of waiting to cut at agricultural machine in-process to with crops territory image input to lodging model, after determining the lodging state of crops through the lodging model, can determine agricultural machine to the operation parameter of waiting to cut the area, thereby make agricultural machine when reaping the crops in lodging area, adjustment operation parameter that can be timely, realize that more intelligent and accurate effectual crops reap the operation.
Additional features and advantages of embodiments of the invention will be set forth in the detailed description which follows.
Drawings
The accompanying drawings, which are included to provide a further understanding of the embodiments of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the embodiments of the invention without limiting the embodiments of the invention. In the drawings:
fig. 1 schematically shows an application environment diagram of a harvesting method for crops according to an embodiment of the present invention;
fig. 2 schematically shows a flow diagram of a harvesting method for a crop according to an embodiment of the invention;
fig. 3 schematically shows a flow diagram of a harvesting method for a crop according to another embodiment of the invention;
FIG. 4 schematically illustrates a work flow diagram of an agricultural machine according to an embodiment of the present disclosure;
FIG. 5 schematically illustrates a schematic representation of an identification process of a lodging model according to an embodiment of the invention;
fig. 6 schematically shows a block diagram of a harvesting apparatus for crop according to an embodiment of the present invention;
fig. 7 schematically shows a block diagram of a harvesting apparatus for crop according to another embodiment of the present invention;
FIG. 8 is a block diagram schematically illustrating the construction of an agricultural machine according to an embodiment of the present invention;
fig. 9 schematically shows an internal configuration diagram of a computer apparatus according to an embodiment of the present invention.
Detailed Description
The following detailed description of embodiments of the invention refers to the accompanying drawings. It should be understood that the detailed description and specific examples, while indicating embodiments of the invention, are given by way of illustration and explanation only, not limitation.
The harvesting method for crops can be applied to the application environment shown in figure 1. The image acquisition equipment 101 is communicated with the lodging model 102, the image acquisition equipment 101 acquires crop area images of crops in an area to be harvested in real time in the operation process of the agricultural machine, the crop area images are transmitted to the lodging model 102, the lodging state of the crops is determined through the lodging model 102, and the operation parameter controller of the agricultural machine can determine operation parameters aiming at the area to be harvested according to the determined lodging state. The image capturing device 101 may include, but is not limited to, a personal computer, a notebook computer, a smart phone, a tablet computer, a portable wearable device, a camera, a video camera, a recorder, an unmanned aerial vehicle, and other electronic products having an image capturing function, and may capture an image in real time, and the lodging model 102 may include at least one of a residual convolution network and a residual upsampling network.
Fig. 2 schematically shows a flow diagram of a harvesting method for a crop according to an embodiment of the invention. In one embodiment of the present invention, as shown in fig. 2, there is provided a harvesting method for crops, comprising the steps of:
In step 202, the crop area image is input to the lodging model to determine the lodging state of the crop.
And step 203, determining the operation parameters of the agricultural machine aiming at the area to be harvested according to the lodging state.
The harvesting method for the crops is applied to agricultural machinery, and the agricultural machinery can be provided with image acquisition equipment for acquiring the crop area images corresponding to the crops in the area to be harvested in the operation process of the agricultural machinery. The region to be harvested is a farmland region which has a certain distance from the agricultural machine and has not been subjected to harvesting operation. And the object region image is the image corresponding to the region to be cut. After the corresponding crop area images are collected, the crop area images can be input into the lodging model, and the crop area images are analyzed and recognized through the lodging model, so that whether crops in the area to be harvested belong to the lodging state or not can be determined. After the lodging state of the crops is determined, the operation parameters of the agricultural machine aiming at the area to be harvested can be determined.
In one embodiment, the operational parameters may include at least one of: header height, reel rotating speed, reel height, fan rotating speed and agricultural machine moving speed.
In the operation process of the agricultural machine, the operation parameters of all parts contained in the machine can be continuously adjusted, and the operation parameters comprise at least one of the height of a cutting table, the rotating speed of a reel, the height of the reel, the rotating speed of a fan, the moving speed of the agricultural machine and the like. If the crops are determined to be in the lodging state, when the agricultural machine harvests the crops in the area to be harvested, the operation parameters of the agricultural machine need to be correspondingly adjusted, so that the agricultural machine can comprehensively harvest the lodging crops. For example, when crops in an area to be harvested have a large area of lodging, the header cannot harvest the lodged crops without reducing the height of the header when the crops in the area to be harvested are harvested. Therefore, when the agricultural machine harvests the crop in the partial lodging area, the height of the header needs to be lowered, so that the cutting knife of the header can effectively harvest the crop lodging on the ground. The agricultural machine can be a harvesting machine, such as a harvester, and can be used for harvesting crops such as wheat, barley, buckwheat, oat, highland barley, rice and the like.
The harvesting method applied to the agricultural machine and used for the crops comprises the steps of continuously collecting crop region images of the crops in the region to be harvested in the agricultural machine process, inputting the crop region images into the lodging model, determining the lodging state of the crops through the lodging model, determining operation parameters of the agricultural machine in the region to be harvested, enabling the agricultural machine to harvest the crops in the lodging region, adjusting the operation parameters in time, and achieving more intelligentization and accurate and effective crop harvesting operation.
In one embodiment, inputting the crop area image to the lodging model to determine the lodging status of the crop comprises: inputting the crop area image into the lodging model; acquiring a prediction label output by the lodging model and corresponding to each image pixel in the crop area image; and determining the lodging state of the crop according to the prediction label.
After the crop region image is input into the lodging model, the lodging model can perform image recognition on the crop region image, and output a prediction label corresponding to each image pixel in the crop region image, so that the lodging state of the crop can be determined according to the prediction label output by the lodging model, namely, whether the crop is lodging or not can be determined according to the prediction label output by the lodging model. That is, the input data of the lodging model is a composition area image, and the output data is a prediction label corresponding to each pixel. The prediction label represents an analysis recognition result of the lodging model, and the recognition result of the lodging model can be obtained according to the prediction label output by the lodging model. For example, 0 represents no lodging and 1 represents lodging. When the prediction label corresponding to a certain image pixel output by the lodging model is 0, the representative lodging model judges that the crop corresponding to the image pixel does not belong to the lodging state; when the prediction label corresponding to a certain image pixel output by the lodging model is 1, the representative lodging model judges that the crop corresponding to the image pixel belongs to the lodging state.
In one embodiment, determining the lodging status of the crop according to the predictive label comprises: selecting a sub-image from the crop area image, wherein the sub-area corresponding to the sub-image in the area to be harvested is closest to the agricultural machine; and determining the lodging state of the crop according to the prediction label corresponding to each image pixel in the selected sub-images.
The agricultural machine can only harvest a fixed range around the agricultural machine due to the limitation of the machine. For example, if the agricultural machine has only one header with a width of 2 meters and is installed right in front of the machine body, the header represents that the agricultural machine can only harvest crops in the area right in front of the header when harvesting, and the harvesting width is less than or equal to 2 meters. Firstly, the collected crop area image corresponds to the whole area to be harvested, and in the actual operation process of the agricultural machine, only the front area of the agricultural machine during operation can be concerned, and the area can be harvested in the next process of the agricultural machine. Secondly, the area is located within the working range of the agricultural machine, and other areas in the area to be harvested are also located outside the working range of the agricultural machine, so that other areas in the area to be harvested can be ignored first.
Therefore, the sub-image corresponding to the sub-area closest to the agricultural machine can be selected from the crop area image, and the lodging state of the crop can be determined according to the prediction label corresponding to each image pixel in the selected sub-image.
In one embodiment, determining the lodging status of the crop according to the predictive label corresponding to each image pixel in the selected sub-image comprises: acquiring the number of pixels of the image which are indicated to belong to the lodging by the prediction label in the sub-image; and determining that the crop belongs to the lodging state under the condition that the ratio of the pixel number to the total image pixel number in the sub-image is greater than a preset ratio. When the lodging state of the crop is determined according to the prediction label corresponding to each image pixel in the selected sub-images, the lodging state can be determined according to the image pixels belonging to the lodging state. Specifically, the number of pixels in the sub-image for which the prediction tag indicates that the pixels belong to the fallen image can be obtained, and the ratio of the number of pixels to the total number of pixels in the sub-image can be calculated. When the proportion is larger than a preset ratio, determining that the crop belongs to a lodging state; when the ratio is less than or equal to the preset ratio, the crop can be determined not to belong to the lodging state. Assuming that the size of the sub-image is 400 × 400, the total number of image pixels of the sub-image is 400 × 400 — 160000. From the data output by the lodging model, the number of pixels with a prediction label of 1 is 150000, i.e. the number of pixels in the sub-image with a prediction label indicating that the pixels belong to the image that is lodging is 150000, the occupation ratio can be calculated as 150000/160000-0.93. If the preset ratio is 0.75 and the ratio 0.93 is greater than the preset ratio 0.75, the crop in the sub-image can be determined to be in the lodging state. If the preset ratio is 0.95 and the above-mentioned ratio 0.93 is less than 0.95, it can be determined that the crop in the sub-image does not belong to the lodging state. The preset ratio can be self-defined according to actual needs.
In one embodiment, determining the working parameters of the agricultural machine for the area to be harvested from the lodging status comprises: and determining operation parameters according to the ratio when the crops are determined to be in the lodging state.
Indicating the ratio of the number of pixels belonging to a lodging image pixel to the total number of pixels in the sub-image, in effect represents the size of the lodging area of the crop in the sub-image, i.e. the lodging ratio in the area of the sub-image. Therefore, after the crops in the sub-images are determined to belong to the lodging state, specific operation parameters of the agricultural machine when the agricultural machine operates in the area can be determined according to the specific occupation ratio. For example, whether the height of the header needs to be reduced or not, and the reduction range can be adjusted according to parameters such as the rotating speed and the height of a reel of the agricultural machine, the rotating speed of a fan, the moving speed of the agricultural machine and the like. Specifically, the lodging percentage is inversely proportional to the moving speed of the agricultural machine and the rotation speed of the reel, and is directly proportional to the rotation speed of the fan.
In one embodiment, the crop area image includes a plurality of crop area images for an area to be cut acquired at a preset image acquisition frequency.
In order to improve the accuracy of the judgment of the lodging state of the crops in the area to be harvested, the crop area images corresponding to the crops in the area to be harvested can be acquired by adopting a preset image acquisition frequency. When the lodging state of crops in the area to be harvested is determined, the lodging state of the crops is actually determined together according to a plurality of crop area images collected every second. For example, the preset image acquisition frequency may be 5fps, which means 5 images per second are acquired. When crop area images of crops in an area to be harvested are collected, 5 crop area images are collected every second, the 5 crop area images are all input into the lodging model, the lodging model outputs a prediction label corresponding to each image pixel in each crop area image, the lodging state corresponding to each crop area image is determined, and the final result, namely the lodging state of the crops, is determined by combining the judgment results of the 5 crop area images.
As shown in fig. 3, in an embodiment of the present invention, there is also provided a harvesting method for crops, comprising the steps of:
And 308, determining the operation parameters of the area to be cut according to the occupation ratio.
In the operation process of the agricultural machine, the image acquisition equipment arranged on the agricultural machine can continuously acquire images of the area to be harvested in front of the agricultural machine, so that crop area images of crops in the area to be harvested can be acquired. The image capturing device may capture images at a preset image capturing frequency, such as 5 images per second. Therefore, a plurality of crop area images of crops in the area to be harvested can be acquired, the crop area images are all input into the lodging model, and the lodging model can perform image recognition on each crop area image to determine the lodging state of the crops in each crop area image. Specifically, for each input crop region image, the lodging model may output a prediction label corresponding to each image pixel in the crop region image, where the prediction label may be 0 or 1. Where a0 may be indicated as not belonging to the lodging and a 1 is indicated as belonging to the lodging.
Further, a sub-image corresponding to a sub-area closest to the agricultural machine can be selected from the crop area image, and the lodging state of the crop can be determined according to the prediction label corresponding to each image pixel in the selected sub-image. In addition, because the camera has a principle of large and small imaging, the sub-area corresponding to the sub-image is trapezoidal in the image, and the sub-image needs to be corrected to be rectangular. Therefore, the lodging state of the crop can be determined according to the prediction label corresponding to each image pixel in the sub-image corrected to be rectangular. The number of pixels in the sub-image for which the prediction tag indicates that the image pixel belongs to a fall may be obtained and a ratio of the number of pixels to the total number of image pixels in the sub-image calculated. And comparing the occupation ratio with a preset ratio, and if the occupation ratio is greater than the preset ratio, judging that the crops in the subimage belong to a lodging state, so that the lodging occupation ratio of the lodging area of the crops in the subimage can be determined according to the occupation ratio, and the operation parameters of the area to be harvested are determined according to the lodging occupation ratio. Specifically, an optimal adjustment instruction for the agricultural machine can be determined according to the lodging proportion, and the optimal adjustment instruction comprises the following steps: the height of the cutting table, the height of the reel, the rotating speed of the reel, the speed of the agricultural machine and other optimal value ranges of all components.
For example, when the lodging percentage reaches 90%, it indicates that the crops in the area corresponding to the sub-image are almost in the lodging state, and when the agricultural machine harvests the area, the operation parameters of the agricultural machine need to be adjusted to harvest the crops in the area in a better state. For example, the height of the header needs to be reduced, the reduction range needs to be reduced, and parameters such as the rotating speed and the height of a reel of the agricultural machine, the rotating speed of a fan, the moving speed of the agricultural machine and the like can be adjusted. Wherein, the lodging proportion is inversely proportional to the moving speed of the agricultural machine and the rotating speed of the reel, and is directly proportional to the rotating speed of the fan. If the proportion is less than or equal to the preset ratio, the crops in the sub-images can be judged not to belong to the lodging state, and when the area is harvested, the operation parameters of the agricultural machine are set to be the corresponding operation parameters in the non-lodging state. Therefore, the operation parameter of the region to be cut can be determined to be the operation parameter corresponding to the non-lodging state, and more specifically, the operation parameter of the sub-region corresponding to the sub-image in the region to be cut is determined to be the non-lodging state.
It will be appreciated that if the current operating parameters under which the agricultural machine is operating correspond to harvesting of crop in an unbent condition, then no adjustment to the operating parameters is required when the agricultural machine is harvesting crop in the area to be harvested if the crop in the area to be harvested is also in an unbent condition. If the crops in the area to be harvested are in the lodging state, the agricultural machine needs to adjust the operation parameters to adapt to harvesting of the crops in the lodging state. Similarly, if the operation parameters of the current agricultural machine in operation correspond to the harvesting of the crops in the lodging state, if the crops in the area to be harvested are in the non-lodging state, the agricultural machine needs to adjust the operation parameters, and the operation parameters are adjusted from the harvesting suitable for the lodging state to the harvesting suitable for the crops in the non-lodging state. If the crops in the area to be harvested are also in the lodging state, the agricultural machine does not need to adjust the operation parameters, or the agricultural machine can correspondingly adjust the current operation parameters according to the size of the lodging area of the crops in the area to be harvested.
In one embodiment, determining the working parameters of the agricultural machine for the area to be harvested from the lodging status comprises: determining the actual position of a region belonging to the lodging state in the crop region image; and determining the adjustment time interval of the operation parameters according to the distance between the agricultural machine and the actual position and the moving speed of the agricultural machine.
After the crops in the crop area image are determined to be in the lodging state, the actual positions of the lodging crops in the farmland can be further determined, the distance between the agricultural machine and the actual position of the lodging area can be calculated, the adjusting time interval of the operation parameters of the agricultural machine can be determined by combining the speed of the agricultural machine, and the adjusting starting time and the adjusting ending time of the operation parameters of the agricultural machine can be determined.
In one embodiment, the lodging model is trained by: obtaining a plurality of crop image samples; extracting a preset area in a crop image sample; and inputting the preset area into the lodging model so as to train the lodging model.
Before putting the lodging model into practical use, can train the lodging model earlier, the lodging model that trains can be more accurate to the discernment of crops lodging state. Therefore, the crop images obtained from the database can be used as training samples of the lodging model, and are called as crop image samples. Specifically, the plurality of crop images in the database may be pictures of various shapes of lodging areas acquired by a camera when the agricultural machine harvests a land parcel with lodging by mounting the camera in front of a windshield of the agricultural machine. Furthermore, the pictures acquired by the method are too large and have a wide range, and the agricultural machine only needs to pay attention to the front area of the current harvesting area in the actual operation process, so that the preset area in the crop image sample can be extracted for training aiming at each crop image sample in order to reduce the calculation amount. The preset region may refer to a middle region of a picture in the crop image sample, for example, a middle 720 × 720 region of the crop image sample may be selected. Further, the selected preset region may be compressed, for example, the resolution of the image corresponding to the preset region is compressed to 512 × 512. The processed preset region can then be input into a lodging model, and the lodging model is trained through the preset region. By the method, the calculated amount of the lodging model can be greatly reduced, meanwhile, redundant information on two sides of the image, such as people, ridges, discharged straws and the like, can be removed, the image is not distorted, the image is zoomed on the basis of the original image, and the morphological characteristic extraction of crops in the image by the lodging model is not influenced. Therefore, the processing speed of the lodging model can be improved, the data structure is not changed due to image distortion, and the performance of the lodging model is not influenced by redundant information.
In one embodiment, the crop image sample carries annotation information. Inputting the preset area into the lodging model to train the lodging model, wherein the training comprises the following steps: inputting the preset area into the lodging model, and acquiring a test label output by the lodging model; and adjusting the model parameters of the lodging model according to the test label and the labeling information so as to train the lodging model.
The marking information refers to a marking label of each pixel in the crop image sample; the label indicates whether the pixel belongs to lodging or not; inputting the preset area into the lodging model to train the lodging model, wherein the training comprises the following steps: inputting the preset area into a lodging model, and outputting a test label of each pixel contained in the preset area by the lodging model; adjusting model parameters of the lodging model according to the test label and the labeling information so as to train the lodging model; the model parameters are adjusted according to the difference between the test label and the label, so that the test label is consistent with the label. When the lodging model is trained, in order to determine whether the lodging model accurately identifies the lodging state of the input sample data, the sample data can be labeled first, that is, the crop image sample can be labeled, so that the crop image sample can carry labeling information. In order to make the labeling information more accurate, the experience of the driver of the agricultural machine can be combined, and the crop image sample can be labeled manually. Specifically, the crop image sample can be labeled by adopting a pixel point labeling mode, that is, a closed-loop area is drawn in the crop image sample, a corresponding numerical value is labeled for each area according to the state of the crop in each area, the area labeled as 1 is specified to represent that the area belongs to a lodging state, and the area labeled as 0 represents that the area does not belong to the lodging state.
After the preset area is input into the lodging model, the test label output by the lodging model can be obtained, and the test label is compared with the labeled information, so that whether the prediction result of the lodging model is correct or not can be known, and the prediction accuracy of the lodging model can be calculated. If the prediction accuracy does not reach the preset accuracy, the lodging model can be continuously trained, namely, model parameters of the lodging model are adjusted, and the continuous training of the lodging model is realized. For example, the preset accuracy is 95%, which indicates that the prediction accuracy of the lodging model needs to be greater than or equal to 95%, and the lodging model is considered to be trained completely. Otherwise, the training of the lodging model is required to be continued.
In one embodiment, the lodging model includes at least one of a residual convolutional network and a residual upsampling network; inputting the preset area into the lodging model to train the lodging model, wherein the training comprises the following steps: inputting the preset area into a residual convolution network to extract image sample characteristics in the preset area; and extracting partial image sample characteristics by the residual up-sampling network, and amplifying to train the lodging model through the image sample characteristics and the amplified image sample characteristics.
In order to extract deeper features, the lodging model can include a full-residual network architecture, i.e., a residual convolutional network. The value before convolution calculation is introduced at the tail of each convolution module for addition, so that information fading is reduced, and the problem that the model cannot well learn the characteristics of data due to the fact that the model has too much fading in network transmission because the depth is not enough or the model is too deep is prevented. The lodging model can also comprise a residual error upsampling network. The features are extracted through a layer-by-layer convolution network, the size of the feature map is reduced layer by layer, so that the feature map can be amplified layer by layer through a residual up-sampling network, and more information can be reserved by the feature map with the same size in the process of cascade convolution. Specifically, the lodging model comprises three different networks, namely a residual convolution network, a residual upsampling network and a convolution network, and each network can comprise a plurality of networks. The cascading mode among the networks can be various, and the three networks can be mixed according to various modes, namely, the connection sequence among the networks can be various.
Therefore, when the preset region is input into the lodging model and the lodging model is trained, the preset region is actually input into the residual convolution network, and the features in the preset region are extracted through the residual convolution network to obtain a plurality of image sample features. Meanwhile, partial image sample features can be extracted from all image sample features through a residual up-sampling network for amplification processing, so that the lodging model can be trained through the image sample features and the amplified image sample features.
Specifically, the extracted features in the lodging model can be calculated in a Softmax and Dice loss manner. Softmax can "compress" a K-dimensional vector containing arbitrary real numbers into another K-dimensional real vector, so that each element has a value ranging from 0 to 1, and the sum of all elements is 1. Therefore, the pixels corresponding to the characteristics of each image sample can be classified twice through Softmax, and whether each pixel is in a lodging state or a non-lodging state is judged. Because the lodging scene accounts for a small amount in the harvesting process of crops, the whole lodging scene accounts for 10% or less, and the lodging pixels account for 20% or less in the image containing the lodging. There is a problem of extreme imbalance of positive and negative samples (where positive samples are lodging pixels and negative samples are non-lodging pixels), so the problem of too small foreground proportion can be solved by a dice loss, which is derived from two classifications, essentially weighing the overlapping part of two samples. The index ranges from 0 to 1, where "1" indicates complete overlap. Therefore, the rice loss can be well suitable for the lodging of crops.
In one embodiment, the work flow of an agricultural machine is as shown in FIG. 4. Firstly, can acquire the place ahead image of the current operation of agricultural machine in real time through the image acquisition equipment of installation on the agricultural machine, then with the image input who gathers to the lodging model in, determine whether crops belong to the state of lodging to and the regional proportion of lodging, control end among the agricultural machine, control module promptly can be according to the lodging proportion of occupying and adjusting its operation parameter to this region, thereby can realize the accurate harvesting to crops. Meanwhile, intelligent unmanned operation can be realized, personnel participation is not needed, operation parameters can be automatically adjusted according to crops in a lodging area, and labor cost is effectively reduced.
In one embodiment, as shown in fig. 5, the specific identification process of the lodging model for the input image may be: inputting training sample images, preprocessing the input images, such as extracting the middle area of each image, then performing operations of feature extraction, calculation processing and the like on the input images through a residual convolution network, a residual up-sampling network and a convolution network in a lodging model, and performing secondary classification on each extracted pixel point through a Softmax + dice loss mode to determine whether crops belong to a lodging state or not.
In one embodiment, as shown in fig. 6, there is provided a harvesting apparatus for a crop, which may be used to perform a harvesting method for a crop as described above, in particular, the harvesting apparatus comprising:
an image acquisition module 601 configured to acquire crop area images of crops in an area to be harvested during operation of the agricultural machine.
An image processing module 602 configured to input the crop area image to the lodging model to determine a lodging status of the crop.
And a control module 603 configured to determine operation parameters of the agricultural machine for the area to be harvested according to the lodging state.
In one embodiment, the image processing module 602 is further configured to input the crop area image to a lodging model; acquiring a prediction label output by the lodging model and corresponding to each image pixel in the crop area image; and determining the lodging state of the crop according to the prediction label.
In one embodiment, the image processing module 602 is further configured to select a sub-image from the crop area image, wherein a sub-area of the area to be harvested corresponding to the sub-image is closest to the agricultural machine; and determining the lodging state of the crop according to the prediction label corresponding to each image pixel in the selected sub-images.
In one embodiment, the image processing module 602 is further configured to obtain a number of pixels in the sub-image for which the prediction tag indicates that the image pixel belongs to a lodging; and determining that the crop belongs to the lodging state under the condition that the ratio of the pixel number to the total image pixel number in the sub-image is greater than a preset ratio.
In one embodiment, the image processing module 602 is further configured to determine the operation parameters based on the ratio if the crop is determined to be in a lodging state.
In one embodiment, the crop area image includes a plurality of crop area images for an area to be cut acquired at a preset image acquisition frequency.
In one embodiment, the operational parameters include at least one of: header height, reel rotating speed, reel height, fan rotating speed and agricultural machine moving speed.
In one embodiment, the control module 603 is further configured to determine an actual position of a region in the composition area image that belongs to the lodging state; and determining the adjustment time interval of the operation parameters according to the distance between the agricultural machine and the actual position and the moving speed of the agricultural machine.
In one embodiment, the harvesting apparatus for crop further comprises a training module 604 configured to train the lodging model, comprising: obtaining a plurality of crop image samples; extracting a preset area in a crop image sample; and inputting the preset area into the lodging model so as to train the lodging model.
In one embodiment, the lodging model includes at least one of a residual convolutional network and a residual upsampling network. The training module 604 is further configured to input the preset region to a residual convolutional network to extract image sample features in the preset region; and extracting partial image sample characteristics by the residual up-sampling network, and amplifying to train the lodging model through the image sample characteristics and the amplified image sample characteristics.
In one embodiment, the crop image sample carries annotation information. The training module 604 is further configured to input the preset region into the lodging model, and obtain a test label output by the lodging model; and adjusting the model parameters of the lodging model according to the test label and the labeling information so as to train the lodging model.
The harvesting device for the crops comprises a processor and a memory, wherein the image acquisition module, the image processing module, the control module and the like are stored in the memory as program units, and the processor executes the program modules stored in the memory to realize corresponding functions.
The processor comprises a kernel, and the kernel calls the corresponding program unit from the memory. The kernel can be set to be one or more than one, and the crops can be harvested by adjusting the kernel parameters.
The memory may include volatile memory in a computer readable medium, Random Access Memory (RAM) and/or nonvolatile memory such as Read Only Memory (ROM) or flash memory (flash RAM), and the memory includes at least one memory chip.
An embodiment of the present invention provides a storage medium having stored thereon a program that, when executed by a processor, implements the harvesting method for crops described above.
An embodiment of the invention provides a processor for running a program, wherein the program is run to execute the harvesting method for crops.
In one embodiment, as shown in fig. 7, there is also provided a harvesting apparatus 700 for crop,
an image capture device 701 configured to obtain crop area images of crops in an area to be harvested in front of the agricultural machine during operation of the agricultural machine;
a processor 702 configured to perform the harvesting method for the crop described above.
In one embodiment, as shown in fig. 8, an agricultural machine 800 is provided that includes a harvesting apparatus 700 for crops as described above.
In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as shown in fig. 9. The computer device includes a processor a01, a network interface a02, a memory (not shown), and a database (not shown) connected by a system bus. Wherein processor a01 of the computer device is used to provide computing and control capabilities. The memory of the computer device comprises an internal memory a03 and a non-volatile storage medium a 04. The non-volatile storage medium a04 stores an operating system B01, a computer program B02, and a database (not shown in the figure). The internal memory a03 provides an environment for the operation of the operating system B01 and the computer program B02 in the nonvolatile storage medium a 04. The database of the computer device is used for storing data such as crop area images and the like. The network interface a02 of the computer device is used for communication with an external terminal through a network connection. The computer program B02 is executed by the processor a01 to implement a harvesting method for a crop.
Those skilled in the art will appreciate that the architecture shown in fig. 9 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.
The embodiment of the invention provides equipment, which comprises a processor, a memory and a program which is stored on the memory and can run on the processor, wherein the processor executes the program and realizes the following steps: obtaining crop area images of crops in an area to be harvested in the operation process of the agricultural machinery; inputting the crop area image into a lodging model to determine the lodging state of the crop; and determining the operation parameters of the agricultural machine aiming at the area to be harvested according to the lodging state.
In one embodiment, inputting the crop area image to the lodging model to determine the lodging status of the crop comprises: inputting the crop area image into the lodging model; acquiring a prediction label output by the lodging model and corresponding to each image pixel in the crop area image; and determining the lodging state of the crop according to the prediction label.
In one embodiment, determining the lodging status of the crop according to the predictive label comprises: selecting a sub-image from the crop area image, wherein the sub-area corresponding to the sub-image in the area to be harvested is closest to the agricultural machine; and determining the lodging state of the crop according to the prediction label corresponding to each image pixel in the selected sub-images.
In one embodiment, determining the lodging status of the crop according to the predictive label corresponding to each image pixel in the selected sub-image comprises: acquiring the number of pixels of the image which are indicated to belong to the lodging by the prediction label in the sub-image; and determining that the crop belongs to the lodging state under the condition that the ratio of the pixel number to the total image pixel number in the sub-image is greater than a preset ratio.
In one embodiment, determining the working parameters of the agricultural machine for the area to be harvested from the lodging status comprises: and determining operation parameters according to the ratio when the crops are determined to be in the lodging state.
In one embodiment, the crop area image includes a plurality of crop area images for an area to be cut acquired at a preset image acquisition frequency.
In one embodiment, determining the working parameters of the agricultural machine for the area to be harvested from the lodging status comprises: determining the actual position of a region belonging to the lodging state in the crop region image; and determining the adjustment time interval of the operation parameters according to the distance between the agricultural machine and the actual position and the moving speed of the agricultural machine.
In one embodiment, the lodging model is trained by: obtaining a plurality of crop image samples; extracting a preset area in a crop image sample; and inputting the preset area into the lodging model so as to train the lodging model.
In one embodiment, the operational parameters include at least one of: header height, reel rotating speed, reel height, fan rotating speed and agricultural machine moving speed.
In one embodiment, the lodging model includes at least one of a residual convolutional network and a residual upsampling network; inputting the preset area into the lodging model to train the lodging model, wherein the training comprises the following steps: inputting the preset area into a residual convolution network to extract image sample characteristics in the preset area; and extracting partial image sample characteristics by the residual up-sampling network, and amplifying to train the lodging model through the image sample characteristics and the amplified image sample characteristics.
In one embodiment, the crop image sample carries annotation information; inputting the preset area into the lodging model to train the lodging model, wherein the training comprises the following steps: inputting the preset area into the lodging model, and acquiring a test label output by the lodging model; and adjusting the model parameters of the lodging model according to the test label and the labeling information so as to train the lodging model.
The present application further provides a computer program product adapted to perform a program for initializing the following method steps when executed on a data processing device: obtaining crop area images of crops in an area to be harvested in the operation process of the agricultural machinery; inputting the crop area image into a lodging model to determine the lodging state of the crop; and determining the operation parameters of the agricultural machine aiming at the area to be harvested according to the lodging state.
In one embodiment, inputting the crop area image to the lodging model to determine the lodging status of the crop comprises: inputting the crop area image into the lodging model; acquiring a prediction label output by the lodging model and corresponding to each image pixel in the crop area image; and determining the lodging state of the crop according to the prediction label.
In one embodiment, determining the lodging status of the crop according to the predictive label comprises: selecting a sub-image from the crop area image, wherein the sub-area corresponding to the sub-image in the area to be harvested is closest to the agricultural machine; and determining the lodging state of the crop according to the prediction label corresponding to each image pixel in the selected sub-images.
In one embodiment, determining the lodging status of the crop according to the predictive label corresponding to each image pixel in the selected sub-image comprises: acquiring the number of pixels of the image which are indicated to belong to the lodging by the prediction label in the sub-image; and determining that the crop belongs to the lodging state under the condition that the ratio of the pixel number to the total image pixel number in the sub-image is greater than a preset ratio.
In one embodiment, determining the working parameters of the agricultural machine for the area to be harvested from the lodging status comprises: and determining operation parameters according to the ratio when the crops are determined to be in the lodging state.
In one embodiment, the crop area image includes a plurality of crop area images for an area to be cut acquired at a preset image acquisition frequency.
In one embodiment, determining the working parameters of the agricultural machine for the area to be harvested from the lodging status comprises: determining the actual position of a region belonging to the lodging state in the crop region image; and determining the adjustment time interval of the operation parameters according to the distance between the agricultural machine and the actual position and the moving speed of the agricultural machine.
In one embodiment, the lodging model is trained by: obtaining a plurality of crop image samples; extracting a preset area in a crop image sample; and inputting the preset area into the lodging model so as to train the lodging model.
In one embodiment, the operational parameters include at least one of: header height, reel rotating speed, reel height, fan rotating speed and agricultural machine moving speed.
In one embodiment, the lodging model includes at least one of a residual convolutional network and a residual upsampling network; inputting the preset area into the lodging model to train the lodging model, wherein the training comprises the following steps: inputting the preset area into a residual convolution network to extract image sample characteristics in the preset area; and extracting partial image sample characteristics by the residual up-sampling network, and amplifying to train the lodging model through the image sample characteristics and the amplified image sample characteristics.
In one embodiment, the crop image sample carries annotation information; inputting the preset area into the lodging model to train the lodging model, wherein the training comprises the following steps: inputting the preset area into the lodging model, and acquiring a test label output by the lodging model; and adjusting the model parameters of the lodging model according to the test label and the labeling information so as to train the lodging model.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
The memory may include forms of volatile memory in a computer readable medium, Random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). The memory is an example of a computer-readable medium.
Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, 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 the process, method, article, or apparatus that comprises the element.
The above are merely examples of the present application and are not intended to limit the present application. Various modifications and changes may occur to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.
Claims (15)
1. A harvesting method for crops, applied to agricultural machinery, characterized in that it comprises:
obtaining crop area images of crops in an area to be harvested in the operation process of the agricultural machinery;
inputting the crop area image to a lodging model to determine a lodging state of the crop;
and determining the operation parameters of the agricultural machine aiming at the area to be harvested according to the lodging state.
2. The harvesting method of claim 1, wherein inputting the crop area image to a lodging model to determine a lodging status of the crop comprises:
inputting the crop area image to the lodging model;
acquiring a prediction label output by the lodging model and corresponding to each image pixel in the crop area image;
and determining the lodging state of the crop according to the prediction label.
3. The harvesting method of claim 2, wherein the determining the lodging of the crop according to the predictive label comprises:
selecting a sub-image from the crop area image, wherein a sub-area corresponding to the sub-image in the area to be harvested is closest to the agricultural machine;
and determining the lodging state of the crop according to the prediction label corresponding to each image pixel in the selected sub-images.
4. The harvesting method of claim 3, wherein determining the lodging of the crop according to the predictive label for each image pixel in the selected sub-image comprises:
acquiring the number of pixels of the image, of which the prediction labels indicate that the pixels belong to the lodging, in the sub-image;
and under the condition that the ratio of the pixel number to the total image pixel number in the sub-image is larger than a preset ratio, determining that the crop belongs to a lodging state.
5. The harvesting method according to claim 4, wherein the determining of the operating parameters of the agricultural machine for the area to be harvested from the lodging includes:
and under the condition that the crops are determined to be in the lodging state, determining the operation parameters according to the proportion.
6. The harvesting method according to claim 1, wherein the crop area images comprise a plurality of crop area images for the area to be harvested acquired at a preset image acquisition frequency.
7. The harvesting method according to claim 1, wherein the determining of the operating parameters of the agricultural machine for the area to be harvested from the lodging includes:
determining the actual position of a region belonging to the lodging state in the crop region image;
and determining the adjustment time interval of the operation parameters according to the distance between the agricultural machine and the actual position and the moving speed of the agricultural machine.
8. The harvesting method of claim 1, wherein the lodging model is trained by:
obtaining a plurality of crop image samples;
extracting a preset area in the crop image sample;
and inputting the preset area into the lodging model so as to train the lodging model.
9. The harvesting method according to any one of claims 1 to 8, wherein the operating parameters include at least one of:
header height, reel rotating speed, reel height, fan rotating speed and the moving speed of the agricultural machine.
10. The harvesting method of claim 8, wherein the lodging model includes at least one of a residual convolutional network and a residual upsampling network;
the inputting the preset area into the lodging model to train the lodging model comprises:
inputting the preset area into the residual convolution network to extract the image sample characteristics in the preset area;
and the residual up-sampling network extracts part of the image sample characteristics to carry out amplification processing so as to train the lodging model through the image sample characteristics and the amplified image sample characteristics.
11. The harvesting method of claim 8, wherein the crop image sample carries annotation information;
inputting the preset region into the lodging model to train the lodging model comprises:
inputting the preset area into the lodging model, and acquiring a test label output by the lodging model;
and adjusting the model parameters of the lodging model according to the test label and the labeling information so as to train the lodging model.
12. A processor configured to perform a harvesting method for a crop according to any one of claims 1 to 11.
13. A harvesting apparatus for crops for use in agricultural machinery, the harvesting apparatus comprising:
an image capture device configured to obtain crop area images of crops in an area to be harvested in front of the agricultural machine during operation of the agricultural machine; and
the processor of claim 12.
14. An agricultural machine, comprising a harvesting device for a crop according to claim 13.
15. A machine readable storage medium having instructions stored thereon, which when executed by a processor causes the processor to be configured to perform a harvesting method for crop plants according to any one of claims 1 to 11.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010967102.6A CN114267005A (en) | 2020-09-15 | 2020-09-15 | Harvesting method, device, processor and agricultural machine for crops |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010967102.6A CN114267005A (en) | 2020-09-15 | 2020-09-15 | Harvesting method, device, processor and agricultural machine for crops |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114267005A true CN114267005A (en) | 2022-04-01 |
Family
ID=80824118
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010967102.6A Pending CN114267005A (en) | 2020-09-15 | 2020-09-15 | Harvesting method, device, processor and agricultural machine for crops |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114267005A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115399139A (en) * | 2022-08-12 | 2022-11-29 | 中联农业机械股份有限公司 | Method, apparatus, storage medium, and processor for determining crop yield |
-
2020
- 2020-09-15 CN CN202010967102.6A patent/CN114267005A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115399139A (en) * | 2022-08-12 | 2022-11-29 | 中联农业机械股份有限公司 | Method, apparatus, storage medium, and processor for determining crop yield |
| CN115399139B (en) * | 2022-08-12 | 2024-04-26 | 中联农业机械股份有限公司 | Method, apparatus, storage medium and processor for determining crop yield |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| DE102020204363A1 (en) | IMAGE SELECTION FOR MACHINE CONTROL | |
| CN109977924A (en) | For real time image processing and system on the unmanned plane machine of crops | |
| CN111985724B (en) | Crop yield estimation method, device, equipment and storage medium | |
| CN112020982A (en) | Automatic lodging crop harvesting method and harvester | |
| CN112541395A (en) | Target detection and tracking method and device, storage medium and electronic device | |
| CN111508260A (en) | Vehicle parking space detection method, device and system | |
| JP7299213B2 (en) | Information processing equipment | |
| KR20210047230A (en) | Fruit tree disease Classification System AND METHOD Using Generative Adversarial Networks | |
| CN115661650A (en) | Farm management system based on data monitoring of Internet of things | |
| CN114267005A (en) | Harvesting method, device, processor and agricultural machine for crops | |
| CN114255394A (en) | Method, device and processor for adjusting operation parameters of agricultural machine | |
| CN113989649B (en) | Remote sensing land parcel recognition method based on deep learning | |
| CN114140396A (en) | Road surface damage detection method, system, device and medium based on unmanned aerial vehicle image | |
| CN118196823A (en) | Intelligent recognition method and device for power grid engineering drawing and drawing recognition system | |
| CN113867406A (en) | Unmanned aerial vehicle-based line inspection method and system, intelligent equipment and storage medium | |
| CN117636314A (en) | Seedling missing identification method, device, equipment and medium | |
| CN117197595A (en) | Fruit tree growth period identification method, device and management platform based on edge calculation | |
| CN107368847A (en) | A kind of crop leaf diseases recognition methods and system | |
| CN116739739A (en) | Loan amount evaluation method and device, electronic equipment and storage medium | |
| CN113837107A (en) | Model training method, video processing method, electronic device and readable storage medium | |
| CN118658069B (en) | Unmanned aerial vehicle-based intelligent plant protection method, device, equipment and medium | |
| CN117689481B (en) | Natural disaster insurance processing method and system based on unmanned aerial vehicle video data | |
| CN117014584B (en) | Vegetation remote sensing product acquisition system and method | |
| CN114485612B (en) | Route generation method and device, unmanned operation vehicle, electronic equipment and storage medium | |
| US20240257529A1 (en) | Method and system for automatically capturing and extracting data from images of agricultural field crops and weeds using machine learning processes |
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 |