WO2020207225A1 - 用于农割作业机器的图像采集装置及其处理方法 - Google Patents

用于农割作业机器的图像采集装置及其处理方法 Download PDF

Info

Publication number
WO2020207225A1
WO2020207225A1 PCT/CN2020/080322 CN2020080322W WO2020207225A1 WO 2020207225 A1 WO2020207225 A1 WO 2020207225A1 CN 2020080322 W CN2020080322 W CN 2020080322W WO 2020207225 A1 WO2020207225 A1 WO 2020207225A1
Authority
WO
WIPO (PCT)
Prior art keywords
collector
collectors
image
images
harvesting
Prior art date
Application number
PCT/CN2020/080322
Other languages
English (en)
French (fr)
Inventor
沈永泉
王波
范顺
吴迪
姚远
张虓
陈睿
Original Assignee
丰疆智能科技股份有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Priority claimed from CN201920473543.3U external-priority patent/CN210466135U/zh
Priority claimed from CN201910280375.0A external-priority patent/CN110209155B/zh
Application filed by 丰疆智能科技股份有限公司 filed Critical 丰疆智能科技股份有限公司
Priority to JP2021559955A priority Critical patent/JP7397880B2/ja
Priority to US17/602,726 priority patent/US20220167555A1/en
Priority to CA3135646A priority patent/CA3135646A1/en
Priority to EP20788625.0A priority patent/EP3954190A4/en
Priority to AU2020256953A priority patent/AU2020256953B2/en
Publication of WO2020207225A1 publication Critical patent/WO2020207225A1/zh

Links

Images

Classifications

    • A—HUMAN NECESSITIES
    • A01—AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
    • A01D—HARVESTING; MOWING
    • A01D41/00—Combines, i.e. harvesters or mowers combined with threshing devices
    • A01D41/12—Details of combines
    • A01D41/127—Control or measuring arrangements specially adapted for combines
    • A01D41/1278—Control or measuring arrangements specially adapted for combines for automatic steering
    • A—HUMAN NECESSITIES
    • A01—AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
    • A01D—HARVESTING; MOWING
    • A01D41/00—Combines, i.e. harvesters or mowers combined with threshing devices
    • A01D41/12—Details of combines
    • A01D41/127—Control or measuring arrangements specially adapted for combines
    • A—HUMAN NECESSITIES
    • A01—AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
    • A01B—SOIL WORKING IN AGRICULTURE OR FORESTRY; PARTS, DETAILS, OR ACCESSORIES OF AGRICULTURAL MACHINES OR IMPLEMENTS, IN GENERAL
    • A01B69/00—Steering of agricultural machines or implements; Guiding agricultural machines or implements on a desired track
    • A01B69/001—Steering by means of optical assistance, e.g. television cameras
    • A—HUMAN NECESSITIES
    • A01—AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
    • A01B—SOIL WORKING IN AGRICULTURE OR FORESTRY; PARTS, DETAILS, OR ACCESSORIES OF AGRICULTURAL MACHINES OR IMPLEMENTS, IN GENERAL
    • A01B79/00—Methods for working soil
    • A01B79/005—Precision agriculture
    • A—HUMAN NECESSITIES
    • A01—AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
    • A01D—HARVESTING; MOWING
    • A01D41/00—Combines, i.e. harvesters or mowers combined with threshing devices
    • A01D41/02—Self-propelled combines
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
    • G05D1/02—Control of position or course in two dimensions
    • G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
    • G05D1/0231—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means
    • G05D1/0246—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means using a video camera in combination with image processing means

Definitions

  • the invention relates to the field of agricultural equipment, in particular to an image acquisition device used for agricultural harvesting machines and an image processing method thereof.
  • the final job of large-scale planting operations is to harvest crops.
  • the final harvest includes cutting, harvesting, and even threshing of the crops.
  • large-scale harvesting machinery is essential for agricultural production. Higher harvesting efficiency can maximize harvest, and good finishing work is more conducive to subsequent re-seeding or straw recovery.
  • the current harvesting machinery is basically a simple cutter machine, in conjunction with a conveying and threshing machine, to simply enlarge and expand the cutter, there are still many shortcomings.
  • An existing harvesting machine is mainly composed of a header, a conveyor, a thresher and a storage bin. Moreover, manual operation of the harvesting machinery is required, because the harvesting of those areas depends on manual experience at what rate. Some harvesting machinery is equipped with a camera, but the image is difficult to fully reflect the specific status of the harvesting operation.
  • Figure 1 in agricultural production, the state of traditional harvesting methods is displayed. The position of the crop to be harvested, the position of the crop that has been harvested, and the state of the crop being harvested are all status information required for traditional harvesting. For example, in a farmer's harvesting operation with a sickle, the next harvesting operation is judged from the position of the crop to be harvested.
  • the position of the crop that has been harvested is behind and the position of the knife of the crop being harvested. Therefore, it can be said to be quite intelligent. A set of operating procedures. However, the existing harvesting machinery basically ignores these status information, and performs harvesting operations mechanically, repeatedly, and without differentiation.
  • One of the main advantages of the present invention is to provide an image acquisition device and processing method for agricultural harvesting machines, in which the information required for harvesting operations is fully considered, and the image of the location of agricultural harvesting machines is collected with high efficiency. Process multiple captured images.
  • Another advantage of the present invention is to provide an image acquisition device for agricultural harvesting machines and a processing method thereof, wherein the acquired image has the position of the crop to be harvested, the position of the crop that has been harvested, and the state of the crop being harvested. Information to assist agricultural harvesting operations.
  • Another advantage of the present invention is to provide an image acquisition device and its processing method for agricultural harvesting machines, wherein the image acquisition device and its processing method are suitable for unmanned agricultural harvesting operations, reducing production costs and promoting Intelligent information coordination.
  • Another advantage of the present invention is to provide an image acquisition device for agricultural cutting machines and a processing method thereof, wherein the collected images and processed images can collect the status of the cutting operation on the surface in real time, using simple and efficient control, and Increase productivity.
  • Another advantage of the present invention is to provide an image acquisition device for agricultural harvesting machines and a processing method thereof, in which the acquired images comprehensively cover the operation area of the harvesting operation, and then the information required for the harvesting operation is grasped.
  • Another advantage of the present invention is to provide an image acquisition device and processing method for agricultural harvesting machines, wherein based on the effectiveness of the acquired images, the processing method quickly obtains the status of the harvesting operation in real time, which is beneficial for subsequent Control execution.
  • Another advantage of the present invention is to provide an image acquisition device for agricultural cutting work machines and a processing method thereof, wherein the processing method assists in positioning the position of the work machine, which facilitates the determination of the traveling position and speed in the subsequent control. .
  • Another advantage of the present invention is to provide an image acquisition device for agricultural harvesting machines and a processing method thereof, wherein the acquired images assist in obtaining the status of the crops being harvested, which facilitates real-time adjustment of the working mode of harvesting operations. Maximize harvesting benefits.
  • Another advantage of the present invention is to provide an image capture device for agricultural cutting machines and a processing method thereof, in which effective captured images and relatively simple processing methods are used, so that subsequent control can be performed directly according to the operating status Response, suitable for unmanned operation.
  • Another advantage of the present invention is to provide an image acquisition device and processing method for agricultural cutting operation machines, in which the acquired image is suitable for identification, and it is convenient to obtain the required operation information from the image feedback. Problems can also be alerted in real time.
  • Another advantage of the present invention is to provide an image acquisition device for agricultural harvesting machines and a processing method thereof, wherein the acquired images do not have redundant information, and are collected for the operation area of agricultural harvesting operations, which fully Utilize resources and simplify processing difficulties.
  • Another advantage of the present invention is to provide an image acquisition device for agricultural cutting machines and a processing method thereof, wherein the acquired images are suitable for remote transmission, making it possible to remotely control the harvesting operation.
  • Another advantage of the present invention is to provide an image acquisition device and processing method for agricultural cutting machines, in which the collected images are not only suitable for simplified recognition, but also suitable for extracting information to quickly understand the working machine status.
  • Another advantage of the present invention is to provide an image acquisition device for agricultural cutting work machines and a processing method thereof, wherein the image acquisition device and the image processing method have low requirements on the work machine and can be configured as additional equipment For machines, promote the evolution of different types of harvesting machines.
  • Another advantage of the present invention is to provide an image capture device for agricultural cutting work machines and a processing method thereof, wherein the image capture device is adaptively configured according to different work machines, the installation configuration is optimized, and the agricultural Stability of production.
  • Another advantage of the present invention is to provide an image capture device for agricultural cutting machines and a processing method thereof, wherein the image capture device is suitable for overall management, so that the captured images are processed cooperatively, Simplify the processing process and increase the processing speed.
  • Another advantage of the present invention is to provide an image acquisition device for agricultural cutting machines and a processing method thereof, wherein according to a preset control program, combined with real-time feedback of the image acquisition device, intelligent unmanned operation Can be achieved and can be supervised.
  • Another advantage of the present invention is to provide an image acquisition device for agricultural cutting operation machines and a processing method thereof, wherein the image acquisition device is not only highly adaptable, but also has high effect stability for different types of operation machines and is suitable for a wide range of applications. application.
  • Another advantage of the present invention is to provide an image acquisition device for agricultural cutting machines and a processing method thereof, in which the transformation cost of the working machines is lower, and more working conditions are provided for reference and control, which is more effective for manual operations. It also provides comprehensive information assistance.
  • an image acquisition device for agricultural cutting work machines of the present invention which can achieve the foregoing objectives and other objectives and advantages, includes a body and a photographing component, wherein the photographing component is installed on On the surface of the body, where the body is activated for harvesting operations, an operation area is formed in the front of the body, and the collection range of the photographing component is covered and installed on the body and faces Image acquisition is performed in the operating area, wherein the body includes a body frame, at least one wheel frame, and a harvesting frame, wherein the harvesting frame is placed in the front of the body frame, and the wheel frame is supported on the At the bottom of the frame, the wheel frame is operated to drive the harvesting frame and the frame forward to cut and transport the crops in front.
  • the photographing component includes at least one collector and a processor, wherein a plurality of collectors of the photographing component are pre-installed on the front and sides of the frame , wherein the image obtained by the collector is processed, summarized and analyzed by the processor.
  • the collector is implemented as a panoramic camera.
  • the collector is implemented as a plurality of wide-angle cameras.
  • the number of the collectors is at least three, wherein the plurality of collectors of the photographing assembly are pre-installed on the front and the sides of the frame, wherein the collectors
  • the obtained images are processed, summarized and analyzed by the processor.
  • the processor performs image processing according to the position where the collector is installed.
  • the collector further includes a distant scene collector, at least one close scene collector, and at least two side scene collectors, wherein the long scene collector and the close scene collector are suitable for installation At the front of the body frame, and respectively facing above and below the harvesting frame, wherein the side scene collectors are respectively installed on both sides of the body frame, and respectively face the two sides of the body side.
  • the distant view collector maintains an orientation along the upper side of the wheel frame
  • the close-range collector maintains its orientation along a position between the wheel frame and the wheel frame
  • the distant scene collector, the close scene collector, and the side scene collector are respectively numbered according to the installed position, wherein the processor obtains corresponding numbers according to each different number.
  • the position of the collector is matched with the position where the image is collected for image processing.
  • a plurality of the distant view collectors are installed at a certain interval, and a plurality of the close-range collectors are installed at a certain interval.
  • the processor performs image processing according to the position where the collector is installed, wherein the images collected by the close-range collector are spliced first, and then combined with the images collected by other collectors Splicing.
  • the processor performs image processing according to the position where the collector is installed, wherein the images collected by the distant view collector are spliced first, and then combined with the images collected by other collectors Splicing.
  • the images collected by a plurality of the collectors in adjacent positions are spliced preferentially, and then are spliced with the images collected by other collectors.
  • the images collected by the distant view collector and the close view collector located above and correspondingly below the harvesting rack are preferentially stitched.
  • the images collected by the close-range collector located below the end of the harvesting frame and the side-view collector at the edge of the end are preferentially spliced.
  • the present invention further provides an agricultural cutting machine including a body and an image acquisition device, wherein the image acquisition device includes a photographing component and a communication machine, wherein the image acquisition device is Installed on the body, wherein the communication machine is communicably connected to the photographing component to obtain collected image information from the photographing component, thereby assisting in adjusting the working position and operating speed of the body.
  • the communication device uploads the information provided by the photographing component to be acquired.
  • the communication machine transmits the processed information of the image collected by the photographing component to the remote control terminal.
  • the photographing component includes: at least one collector and a processor, wherein a plurality of collectors of the photographing component are pre-installed on the front and sides of the body , Wherein the image obtained by the collector is processed, summarized and analyzed by the processor.
  • the number of the collectors is at least three, and the plurality of collectors of the photographing assembly are pre-installed on the front and sides of the body, and the collectors are The obtained images are processed and aggregated by the processor and transmitted to the communication device.
  • the processor performs image processing according to the position where the collector is installed.
  • the present invention further provides an image processing method for agricultural cutting machines, wherein the image processing method includes the following steps:
  • step III further comprising the steps:
  • the state information of the body is transmitted remotely.
  • step II image processing is performed according to the positions where the multiple collectors installed in the body are installed.
  • step II images collected by a plurality of the collectors with similar orientations are preferentially stitched.
  • step II the images collected by a plurality of adjacent collectors are preferentially stitched.
  • Figure 1 is a schematic diagram of an agricultural harvesting method and scene.
  • Fig. 2 is an overall schematic diagram of an image acquisition device for a farming machine according to a preferred embodiment of the present invention.
  • Fig. 3 is a schematic side view of an image capture device for a farming machine according to the above preferred embodiment of the present invention.
  • Fig. 4 is a partial schematic diagram of an image processing method for a farming machine according to the above preferred embodiment of the present invention.
  • Fig. 5 is a partial schematic diagram of an image processing method for a farming machine according to the above preferred embodiment of the present invention.
  • Fig. 6 is a partial schematic diagram of an image processing method for a farming machine according to the above preferred embodiment of the present invention.
  • Fig. 7 is a schematic diagram of an image processing method for a farming machine according to the above preferred embodiment of the present invention.
  • Fig. 8 is a partial schematic diagram of an image processing method for a farming machine according to another feasible manner of the above-mentioned preferred embodiment of the present invention.
  • Fig. 9 is a partial schematic diagram of an image processing method for a farming machine in the above feasible manner according to the above preferred embodiment of the present invention.
  • Fig. 10 is an application schematic diagram of an image acquisition device and a processing method thereof for a farming machine according to the above-mentioned preferred embodiment of the present invention.
  • the invention provides an image acquisition device for agricultural harvesting machines and a processing method thereof, which collect and process status information in agricultural harvesting operations in real time, provide a direct data source for subsequent control and supervision, and optimize operation efficiency Provide the basis.
  • the agricultural cutting operation machine includes a body 10 and a photographing assembly 20, wherein the photographing assembly 20 is installed on the surface of the body 10.
  • an operation area 100 is formed in the front of the machine body 10, and the machine body 10 only performs harvesting operations in the operation area 100.
  • the harvesting operation of the body 10 sequentially performs operations such as cutting, conveying, threshing, and storing crops during the forward movement.
  • the operation area 100 of the body 10 includes not only crops being harvested, but also crops that have been cut and crops to be harvested, and the operation area 100 is mobile.
  • the image capturing component 20 is installed on the body 10 in a covering manner, and collects images toward the operating area 100. In other words, the collection range of the image capturing component 20 covers the front and the sides of the body 10. It is worth mentioning that the cutting operation has been saved by the body 10.
  • the imaging component 20 collects images of the crops being harvested and the operation area 100 of the crops to be harvested.
  • the photographing assembly 20 is installed from the front and both sides of the body 10. That is to say, the imaging component 20 collects the operation area 100 in the front half of the periphery.
  • the image collected by the imaging component 20 basically covers the operating area 100.
  • the images collected from the photographing component 20 can reflect the images around the body 10, and after further processing, the collected images can express the state information of the operating area 100.
  • the image captured by the image capturing component 20 may be multiple copies, and the image provided by the output is one copy. More preferably, the output of the photographing component 20 is the status information of the operation area 100. The operating state of the body 10 is obtained according to the images collected by the photographing component 20.
  • the image acquisition device includes the photographing component 20 and a communication device 30, wherein the communication device 30 is communicably connected to the photographing component 20 to obtain information from the photographing component 20 Collected image information.
  • the image capture device is installed in the body 10.
  • the photographing component 20 acquires an image of the operating area 100
  • the communication device 30 uploads the information provided by the photographing component 20 to be acquired.
  • the communication machine 30 transmits an image provided by the photographing component 20 to the remote monitoring terminal, thereby ensuring real-time transmission. More preferably, the communication machine 30 transmits the processed information of the image collected by the photographing component 20 to the remote control terminal, so as to ensure the transmission speed.
  • the photographing component 20 is distributed on the front and side of the body 10.
  • the multiple images collected by the photographing component 20 are spliced and then transmitted through the communication machine 30.
  • the body 10 includes a body frame 11, at least one wheel frame 12, and a harvesting frame 13, wherein the harvesting frame 13 is placed at the front of the body frame 11.
  • the wheel frame 12 is supported on the bottom of the body frame 11.
  • the wheel frame 12 is operated to drive the harvesting frame 13 and the body frame 11 forward to cut and transport the crops in front.
  • the image acquisition device is placed on the frame 11 of the agricultural cutting machine. More specifically, the photographing component 20 is placed on the body frame 11, and the operation area 100 that the photographing component 20 faces avoids collecting the wheel frame 12 and the harvesting frame 13.
  • the image capturing component 20 includes multiple collectors and a processor 24.
  • the multiple collectors of the photographing assembly 20 are pre-installed on the front and sides of the body frame 11 in an orientation.
  • the images obtained by the collector are processed, summarized and analyzed by the processor 24.
  • the processor 24 performs image processing according to the position where the collector is installed.
  • the collector may be implemented as a plurality of wide-angle cameras, or may be implemented as a panoramic camera.
  • the collectors are called distributed installations as shown in FIG. 3 and classified as a long-range collector 21, at least one close-range collector 22, and at least one side-view collector 23.
  • the long-range collector 21 and the close-range collector 22 are adapted to the front of the frame 11 and face upward and downward of the harvesting frame 13 respectively. More specifically, the distant view collector 21 collects along the upper side of the wheel frame 12, and the close view collector 22 collects along the position between the wheel frame 12 and the wheel frame 12.
  • the side scene collectors 23 are respectively installed on both sides of the body frame 11 to collect images of crop conditions on both sides.
  • the images collected by the distant view collector 21 are related to the state of the crop to be harvested
  • the images collected by the close-range collector 22 are related to the crop being harvested.
  • the image collected by the side view collector 23 is related to the position of the operation area 100.
  • the distant view collector 21, the close view collector 22, and the side view collector 23 are respectively numbered according to the installed positions.
  • the processor 24 obtains the position of the corresponding collector according to each different number, and performs image processing in accordance with the position where the image is collected. Since the processing is based on the location of the image, the processing speed is increased.
  • image splicing processing based on the installation position of the distant view collector 21, the close view collector 22, and the side view collector 23, the images around the body 10 can be obtained quickly, thereby obtaining The state of the operating area 100.
  • the body frame 11, the wheel frame 12, and the harvester in the collected images Shelf 13 can be omitted by default.
  • the lower edge of the image collected by the distant view collector 21 may be regarded as the harvesting frame 13, and is pre-cut or ignored in the processing degree.
  • the upper edge of the image collected by the close-range collector 22 may be regarded as the harvesting frame 13, and the lower edge may be regarded as the wheel frame 12, so as to obtain status information of the crop being cut.
  • the lower edge of the image collected by the side view collector 23 can be regarded as the wheel frame 12, so that part of the body 10 can be excluded. Therefore, the images collected by the distant view collector 21, the close view collector 22, and the side view collector 23 are quickly and preliminarily processed by the processor 24 and spliced to obtain the status information of the operation area 100 .
  • FIGS. 4 to 6 An image processing method of this preferred embodiment is shown in FIGS. 4 to 6.
  • the image processing method for agricultural cutting machines includes the following steps:
  • a plurality of the distant view collectors 21 collect toward the upper side of the harvesting rack 13.
  • a plurality of the distant view collectors 21 are installed at a certain interval so that the operation area 100 at the front of the body 10 is covered.
  • a plurality of the distant view collectors 21 performs image collection for the crops to be harvested, that is, step 1.
  • the collected images are sequentially spliced, so that the image of the crop to be harvested in front of the harvesting rack 13 is collected. It is worth mentioning that, unlike the existing image stitching, there is no need to identify each image to determine the content of the image, but directly based on the installation position or the harvesting frame 13 in each image to stitch multiple images. Thus, a comprehensive image in front of the body 10 is obtained, that is, step II.
  • the operating state 100 in front of the body 10 is acquired in real time and represents state information.
  • the collected images can be further analyzed in detail later.
  • a plurality of the close-range collectors 22 collects toward the bottom of the harvesting frame 13 and between the wheel frames 12.
  • a plurality of the close-range collectors 22 are installed at a certain interval so that the operation area 100 in the middle of the body 10 is covered. That is to say, a plurality of the close-range collectors 22 perform image collection for the crops being harvested, that is, step 1.
  • the collected images are sequentially spliced, so that the images of the crops being harvested under the harvesting rack 13 are collected. It is worth mentioning that, unlike the existing image stitching, there is no need to identify each image to determine the content of the image, but directly based on the installation position or the harvesting frame 13 in each image to stitch multiple images. In this way, a comprehensive image of the working body 10 is obtained, that is, step II.
  • the cutting state in the operation area 100 of the body 10 is acquired and characterized in real time.
  • the collected images can be further analyzed in detail later.
  • a plurality of side scene collectors 23 collect toward both sides of the body 10.
  • a plurality of side scene collectors 23 are respectively installed on the left and right sides of the body 10 so that the crop status on both sides of the body 10 can be obtained. That is to say, generally, the side scene collector 23 performs image collection on the field where the harvest has been completed and the field to be harvested separately, that is, step I.
  • the collected images are sequentially spliced, so that the images of the field that has been harvested and the field to be harvested are collected.
  • the installation position such as the side view collector 23 installed on the left side of the body 10
  • the side view collector 23 installed on the right side of the body 10 obtains the crop status of the field to be harvested, so as to obtain the status of the left and right crops that the body 10 is working on, and then obtain all the crops.
  • the location of the body 10 is step II.
  • the forward direction and speed of the body 10 can be obtained, ensuring the accuracy of the forward direction of harvesting, and assisting in improving the harvesting efficiency.
  • the images collected by the distant view collector 21, the close view collector 22, and the side view collector 23 are finally stitched together to obtain the image of the operating area 100 of the body 10.
  • the three-sided state of the body 10 is comprehensively obtained.
  • the photographing component 20 finally obtains an observation angle similar to that of the body 10 from top to bottom, including the crops being harvested, the crops to be harvested, and the location of the harvesting operations in the field.
  • the image processing method after step III, further includes the steps:
  • the status information of the body 10 is transmitted remotely.
  • the working state of the body 10 can be obtained remotely. Further, the machine body 10 is remotely controlled for parameters in the harvesting operation. For example, there is a certain gap between the crops obtained from the images collected by the close-range collector 22 and the side-view collector 23 and the harvesting frame 13, that is, the work of the body 10 is relatively deviated from the required crop area. , The harvesting rack 13 needs to be fully filled to achieve high-efficiency crops. More specifically, for example, in the feedback image, it is obtained that 80% of the left side of the harvesting rack 13 is harvesting crops, then this status information can be obtained remotely, and the forward direction of the body 10 can be adjusted to the left. The harvesting rack 13 can fully cover the crops to be harvested.
  • the adjustment of the operation of the body 10 can be manually operated remotely, or can be adjusted by a predetermined program, and the image obtained by the image acquisition device is assisted to control to realize unmanned agricultural production.
  • FIGS. 8-9 Another possible way of the above-mentioned preferred embodiment of the present invention is an image acquisition device for agricultural cutting machines, as shown in FIGS. 8-9, wherein the agricultural cutting machine includes a body 10A and a photographing assembly 20A.
  • the structure of the body 10A and the photographing component 20A is similar to the body 10 and the photographing component 20 of the agricultural harvesting machine in the above-mentioned preferred embodiment, and will not be repeated in the present invention.
  • FIGS. 4 to 6 are mosaic images according to the same type of installation position, and the last three different perspective collectors 21A, close-range collectors 22A, and side-view collectors 23A Then composite the final image.
  • the images collected by the collectors with similar positions are preferably spliced, and then processed subsequently As an image.
  • the images collected by the long-range collector 21A and the close-range collector 22A located above and correspondingly below the harvesting rack 13A are spliced first.
  • the multiple distant view collectors 21A and the multiple close-range collectors 22A are considered to be adjacent to each other in installation positions.
  • the distant view collector 21A and the close view collector 22A perform image collection for the crops to be harvested and the crops being harvested at the same position relative to the harvesting rack 13A, that is, step I.
  • the collected images are sequentially spliced so that they are located above and correspondingly below the harvesting rack 13A
  • the image is captured. It is worth mentioning that, unlike the existing image stitching, there is no need to identify each image to determine the content of the image, but directly based on the installation position, so as to obtain the crop at a certain position of the body 10A before and after being cut. A comprehensive picture, step II.
  • the operating state 100A in front of the body 10A is acquired in real time and represents state information.
  • the collected images can be further analyzed in detail later, similar to the viewing angle of traditional manual cutting.
  • the images collected by the close-range collector 22A located below the end of the harvesting frame 13A and the side-view collector 23A at the edge of the end are spliced first.
  • the side scene collector 23A on one side and the multiple close scene collectors 22A are considered to be installed adjacent to each other.
  • the side view collector 23A and the close view collector 22A perform image collection for the crops being harvested and the crops that have been harvested opposite to the end of the harvesting rack 13A, that is, step I.
  • the installation positions of the close-range collector 22A under the end of the harvesting rack 13A and the side-view collector 23A at the edge of the end are sequentially spliced with the collected images so that they are located at the end of the harvesting rack 13A
  • the images are collected. It is worth mentioning that, unlike the existing image stitching, there is no need to identify each image to determine the content of the image, but directly based on the installation position, so as to obtain the crop at a certain position of the body 10A before and after being cut. A comprehensive picture, step II.
  • the operating state 100A in front of the body 10A is acquired in real time and represents state information.
  • the collected images can be further analyzed in detail later, similar to the viewing angle of traditional manual cutting.
  • the edge working state of the machine body 10A is further obtained, which is convenient to assist in adjusting the working position and working speed of the machine body 10A.
  • the images collected by all the collectors are finally stitched to obtain the body An image of the operation area 100A of 10A.
  • the state before and after operation of the body 10A is comprehensively obtained.
  • the image capturing component 20A finally obtains an observation angle similar to that of the body 10A from top to bottom, including the crops being harvested and the crops to be harvested, and the location of the harvesting operations in the field.
  • the working state of the body 10A in this embodiment can be obtained remotely. Further, the machine body 10A is remotely controlled for parameters in the harvesting operation. For example, there is a certain gap between the crops obtained from the images collected by the close-range collector 22A and the side-view collector 23A and the harvesting rack 13A, that is, the work of the body 10A is relatively deviated from the required crop area , The harvesting rack 13A needs to be close to the crops to be harvested to achieve high-efficiency crops.
  • the feedback image it is obtained that there are unharvested crops on the left side of the harvesting rack 13A and displayed in the side view collector 23A, then this status information can be obtained remotely and adjusted The forward direction of the body 10A is to the left, so that the harvesting frame 13A fully covers the crops to be harvested.

Landscapes

  • Life Sciences & Earth Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Environmental Sciences (AREA)
  • Physics & Mathematics (AREA)
  • Mechanical Engineering (AREA)
  • Soil Sciences (AREA)
  • Electromagnetism (AREA)
  • Multimedia (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Aviation & Aerospace Engineering (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • General Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Combines (AREA)
  • Guiding Agricultural Machines (AREA)

Abstract

一种用于农业收割作业机器的图像采集装置及其图像处理方法，其中农割作业机器包括一机体(10)和一图像采集装置，其中所述图像采集装置包括一摄图组件(20)和一通信机(30)，其中图像采集装置被安装于机体(10)，其中通信机(30)被可通信地连接于摄图组件(20)，以从摄图组件(20)获得所采集图像信息，进而辅助调整机体(10)的工作位置和作业速度。另提供一种图像处理方法，充分考虑收割作业所需信息，对农割作业机器所处位置的图像进行采集，并高效率地处理多个所采集的图像。

Description

用于农割作业机器的图像采集装置及其处理方法 技术领域
本发明涉及农业器械领域，尤其涉及一种用于农业收割作业机器的图像采集装置及其图像处理方法。
背景技术
农业生产中，大范围的种植作业最后的工作是对作物进行收获。对于茎秆的作物而言，最后的收获作业包括了对作物的切割、收取、甚至于脱粒等等。目前，大型的收割机械对于农业生产至关重要。较高的收割效率可以使得收获最大化，良好的收尾工作更利于后续的再次播种或者秸秆回收。但是，目前现有的收割机械基本是单纯地割刀机器，配合输送脱粒机器，单纯地将割刀放大化，扩宽化，还是有很多不足之处。
一种现有的收割机械主要由割台、输送器、脱粒器及储存仓所组成。而且，需要人工的操作收割机械进行作业，因为要对那些区域进行收割，以何种速率进行都是依靠人工经验的。有些收割机械设置了摄像头，但是图像很难全面地反应收割作业的具体状态。如图1所示，农业生产中，传统的收割作业方式状态被展示。待收割作物的位置、已经收割的作物的位置、正在收割的作物的状态，都是传统收割所需要的状态信息。例如，农民的使用镰刀收割作业中，从待收割作物的位置中判断下次收割作业的对象，已经收割的作物位置处于身后，正在收割的作物的下刀位置，因此可以说是相当智能化的一套操作流程。但是，现有的收割机械基本地忽视掉了这些状态信息，机械地，重复地，无差异化地进行收割作业。
虽然大型的现有收割机械可以快速完成大范围的作业，但是高效率的作业完成是离不开人工经验的。需要人工判断哪里进行何种速率的收割，突发问题也需要人工的介入解决。无人化操作是很难完成的。而且，现有的收割机械不能实时地获得所需要的状态信息，不能全面地得到收割作业所需的信息，也就不会做出正确的判断。
另外，因为农业生产的特殊性，所需要的并不仅仅是机械周围的场景。也可以说，即使获得了机械周围的场景图像，也需要后续很多识别处理。这些算法对于农业生产来说成本过高，耗时较长，对于实时性较强的收割作业来说，机械周 围的场景并不重要。而且，与城市交通所需要的远度不同，农业生产中较远距离的场景图像并没有较大用处，仅仅是辽阔土地的冗余图像而已。
考虑到农业领域的大范围生产的需要，用于农业收割作业机器在作业中的图像采集装置及其图像处理方法被市场所追求。
发明内容
本发明的一个主要优势在于提供一种用于农割作业机器的图像采集装置及其处理方法，其中充分考虑收割作业所需信息，对农割作业机器所处位置的图像进行采集，并高效率地处理多个所采集的图像。
本发明的另一个优势在于提供一种用于农割作业机器的图像采集装置及其处理方法，其中所采集的图像具有待收割作物的位置、已经收割的作物的位置、正在收割的作物的状态的信息，以辅助农业收割作业。
本发明的另一个优势在于提供一种用于农割作业机器的图像采集装置及其处理方法，其中所述图像采集装置及其处理方法适用于无人化的农业收割作业，降低生产成本，促进智能化信息统筹。
本发明的另一个优势在于提供一种用于农割作业机器的图像采集装置及其处理方法，其中所采集图像以及处理后的图像可以实时地表征收割作业的状态，利用简单高效的控制，并提高生产效率。
本发明的另一个优势在于提供一种用于农割作业机器的图像采集装置及其处理方法，其中所采集的图像全面性地覆盖收割作业的操作区域，进而掌握收割作业所需要的信息。
本发明的另一个优势在于提供一种用于农割作业机器的图像采集装置及其处理方法，其中基于所采集的图像的有效性，所述处理方法实时迅速地获得收割作业的状态，利于后续的控制执行。
本发明的另一个优势在于提供一种用于农割作业机器的图像采集装置及其处理方法，其中所述处理方法辅助地定位所述作业机器的位置，利于后续的控制中判断行进位置与速度。
本发明的另一个优势在于提供一种用于农割作业机器的图像采集装置及其处理方法，其中所采集的图像辅助地得到正在收割的作物的状态，利于实时地调整收割作业的工作方式，最大地获得收割效益。
本发明的另一个优势在于提供一种用于农割作业机器的图像采集装置及其处理方法，其中利用有效的所采集的图像和相对简单的处理方法，使得后续控制可以直接根据作业状态而进行反应，适于无人化的作业。
本发明的另一个优势在于提供一种用于农割作业机器的图像采集装置及其处理方法，其中所采集的图像适于被识别，方便从图像反馈中得到所需要的作业信息，对于出现的问题也可以得到实时地报警。
本发明的另一个优势在于提供一种用于农割作业机器的图像采集装置及其处理方法，其中所采集的图像并无冗余的信息，针对农业收割作业的操作区域而进行采集，充分地利用资源并简化处理难度。
本发明的另一个优势在于提供一种用于农割作业机器的图像采集装置及其处理方法，其中所采集的图像适于被远程传输，使得远程控制收割作业变得可能。
本发明的另一个优势在于提供一种用于农割作业机器的图像采集装置及其处理方法，其中所采集的图像不仅适于被简化识别，也适于被提取信息，快速地了解作业机器的状态。
本发明的另一个优势在于提供一种用于农割作业机器的图像采集装置及其处理方法，其中所述图像采集装置和所述图像处理方法对作业机器的要求不高，可以作为附加设备配置于机器，促进不同类型的收割作业机器进化。
本发明的另一个优势在于提供一种用于农割作业机器的图像采集装置及其处理方法，其中所述图像采集装置根据不同的作业机器适应性地配置，最优化地解决安装配置，提升农业生产的稳定性。
本发明的另一个优势在于提供一种用于农割作业机器的图像采集装置及其处理方法，其中所述图像采集装置适于被统筹性地管理，使得所采集的图像被协作性地处理，简化处理过程，提升处理速度。
本发明的另一个优势在于提供一种用于农割作业机器的图像采集装置及其处理方法，其中根据预设定的控制程序，结合所述图像采集装置实时地反馈，智能化的无人作业可以被实现并且可以被监管。
本发明的另一个优势在于提供一种用于农割作业机器的图像采集装置及其处理方法，其中所述图像采集装置不仅适应性强，而且不同类型的作业机器效果稳定性高，适于广泛应用。
本发明的另一个优势在于提供一种用于农割作业机器的图像采集装置及其 处理方法，其中对作业机器的改造成本较低，提供更多的作业状态以供参考控制，对人工作业也提供全面性地信息辅助。
本发明的其它优势和特点通过下述的详细说明得以充分体现并可通过所附权利要求中特地指出的手段和装置的组合得以实现。
依本发明的一个方面，能够实现前述目的和其他目的和优势的本发明的一用于农割作业机器的图像采集装置，包括一机体和一摄图组件，其中所述摄图组件被安装于所述机体表面，其中所述机体被启动而进行收割作业中，于所述机体的前部形成一操作区域，其中所述摄图组件的采集范围被覆盖式地安装于所述机体，并朝向所述操作区域进行图像采集，其中所述机体包括一身架、至少一轮架和一收割架，其中所述收割架被置于所述身架的前部，其中所述轮架支撑在所述身架的底部，其中所述轮架***作而带动所述收割架和所述身架向前对前方的作物进行切割和输送。
根据本发明的一个实施例，所述摄图组件包括至少一采集器和一处理器，其中所述摄图组件的多个采集器被预先朝向地安装于所述身架的前部与侧部，其中所述采集器所获得的图像被所述处理器处理汇总并分析。
根据本发明的一个实施例，所述采集器被实施为一个全景式的摄像头。
根据本发明的一个实施例，所述采集器被实施为多个广角式摄像头。
根据本发明的一个实施例，所述采集器为至少三个，其中所述摄图组件的多个采集器被预先朝向地安装于所述身架的前部与侧部，其中所述采集器所获得的图像被所述处理器处理汇总并分析。
根据本发明的一个实施例，所述处理器按照所述采集器所安装的位置进行图像处理。
根据本发明的一个实施例，所述采集器进一步地包括一远景采集器、至少一近景采集器、以及至少两侧景采集器，其中所述远景采集器和所述近景采集器被适于安装于所述身架的前部，且分别地朝向所述收割架上方与下方，其中所述侧景采集器被分别地安装于所述身架的两侧，且分别地朝向所述机体的两侧。
根据本发明的一个实施例，所述远景采集器沿着所述轮架的上方保持朝向，其中所述近景采集器沿着所述轮架与所述轮架之间的位置而保持朝向。
根据本发明的一个实施例，所述远景采集器、所述近景采集器，以及所述侧景采集器分别地根据所安装的位置被编号，其中所述处理器根据每个不同的编号 得到相应的采集器的位置，配合图像所采集的位置进行图像处理。
根据本发明的一个实施例，多个所述远景采集器被以一定间距地安装，其中多个所述近景采集器被以一定间距地安装。
根据本发明的一个实施例，所述处理器按照所述采集器所安装的位置进行图像处理，其中所述近景采集器所采集的图像被优先拼接，进而与其他所述采集器所采集的图像拼接。
根据本发明的一个实施例，所述处理器按照所述采集器所安装的位置进行图像处理，其中所述远景采集器所采集的图像被优先拼接，进而与其他所述采集器所采集的图像拼接。
根据本发明的一个实施例，位置邻近的多个所述采集器所采集图像被优先地拼接，进而与其他所述采集器所采集的图像拼接。
根据本发明的一个实施例，位于所述收割架上方和对应下方的所述远景采集器与所述近景采集器所采集的图像被优先拼接。
根据本发明的一个实施例，位于所述收割架端部下方的所述近景采集器与端部边缘所述侧景采集器所采集的图像被优先拼接。
依本发明的另一个方面，本发明进一步提供一农割作业机器，包括一机体和一图像采集装置，其中所述图像采集装置包括一摄图组件和一通信机，其中所述图像采集装置被安装于所述机体，其中所述通信机被可通信地连接于所述摄图组件，以从所述摄图组件获得所采集图像信息，进而辅助调整所述机体的工作位置和作业速度。
根据本发明的一个实施例，所述通信机将所述摄图组件所提供的信息上传等待被获取。
根据本发明的一个实施例，所述通信机将所述摄图组件所采集的图像的处理后信息传输至远程控制端。
根据本发明的一个实施例，所述摄图组件包括：至少一采集器和一处理器，其中所述摄图组件的多个采集器被预先朝向地安装于所述机体的前部与侧部，其中所述采集器所获得的图像被所述处理器处理汇总并分析。
根据本发明的一个实施例，所述采集器为至少三个，其中所述摄图组件的多个采集器被预先朝向地安装于所述机体的前部与侧部，其中所述采集器所获得的图像被所述处理器处理汇总并传至所述通信机。
根据本发明的一个实施例，所述处理器按照所述采集器所安装的位置进行图像处理。
依本发明的另一个方面，本发明进一步提供一种用于农割作业机器的图像处理方法，其中所述图像处理方法包括以下步骤：
I.采集一机体前部与侧部的多份图像；
II.根据所采集的图像的位置，拼接图像；以及
III.根据所拼接的图像，获得围绕所述机体的状态信息。
根据本发明的一个实施例，在步骤III之后，进一步地包括步骤：
远程地传输所述机体的状态信息。
根据本发明的一个实施例，在步骤II中，按照被安装于所述机体的多个采集器所安装的位置进行图像处理。
根据本发明的一个实施例，在步骤II中，朝向相近的多个所述采集器所采集的图像被优先地拼接。
根据本发明的一个实施例，在步骤II中，位置邻近的多个所述采集器所采集图像被优先地拼接。
通过对随后的描述和附图的理解，本发明进一步的目的和优势将得以充分体现。
本发明的这些和其它目的、特点和优势，通过下述的详细说明，附图和权利要求得以充分体现。
附图说明
图1是一种农业收割作业的方式和场景示意图。
图2是根据本发明的一个优选实施例的用于农割作业机器的图像采集装置的整体示意图。
图3是根据本发明的上述优选实施例的用于农割作业机器的图像采集装置的侧面示意图。
图4是根据本发明的上述优选实施例的用于农割作业机器的图像处理方法的部分示意图。
图5是根据本发明的上述优选实施例的用于农割作业机器的图像处理方法的部分示意图。
图6是根据本发明的上述优选实施例的用于农割作业机器的图像处理方法的部分示意图。
图7是根据本发明的上述优选实施例的用于农割作业机器的图像处理方法的示意图。
图8是根据本发明的上述优选实施例的另一种可行方式的用于农割作业机器的图像处理方法的部分示意图。
图9是根据本发明的上述优选实施例的上述可行方式的用于农割作业机器的图像处理方法的部分示意图。
图10是根据本发明的上述优选实施例的用于农割作业机器的图像采集装置及其处理方法应用示意图。
具体实施方式
以下描述用于揭露本发明以使本领域技术人员能够实现本发明。以下描述中的优选实施例只作为举例，本领域技术人员可以想到其他显而易见的变型。在以下描述中界定的本发明的基本原理可以应用于其他实施方案、变形方案、改进方案、等同方案以及没有背离本发明的精神和范围的其他技术方案。
本领域技术人员应理解的是，在本发明的揭露中，术语“纵向”、“横向”、“上”、“下”、“前”、“后”、“左”、“右”、“竖直”、“水平”、“顶”、“底”、“内”、“外”等指示的方位或位置关系是基于附图所示的方位或位置关系，其仅是为了便于描述本发明和简化描述，而不是指示或暗示所指的装置或元件必须具有特定的方位、以特定的方位构造和操作，因此上述术语不能理解为对本发明的限制。
可以理解的是，术语“一”应理解为“至少一”或“一个或多个”，即在一个实施例中，一个元件的数量可以为一个，而在另外的实施例中，该元件的数量可以为多个，术语“一”不能理解为对数量的限制。
本发明提供一用于农割作业机器的图像采集装置及其处理方法，对于农业收割作业中的状态信息进行实时地采集并处理，为后续的控制与监管提供直接的数据源，为优化作业效率提供基础。
本发明的一个优选实施例如图2至图10所示，所述农割作业机器包括一机体10和一摄图组件20，其中所述摄图组件20被安装于所述机体10表面。所述 机体10被启动而进行收割作业中，于所述机体10的前部形成一操作区域100，所述机体10仅在所述操作区域100进行收割作业。本领域的技术人员可以理解的是，所述机体10的收割作业在前行中对农作物依次地进行切割、输送、脱粒以及储存等等操作。也就是说，所述机体10的所述操作区域100中除了正在收割的作物还包括已经切割的作物和待收割的作物，而且所述操作区域100为移动的。
所述摄图组件20被覆盖式地安装于所述机体10，并朝向所述操作区域100进行图像采集。也就是说，所述摄图组件20的采集范围覆盖所述机体10的前部和侧部。值得一提的是，已经切割的作业已经被所述机体10所保存。所述摄图组件20针对正在收割的作物和待收割的作物的所述操作区域100进行图像采集。优选地，在本优选实施例中，所述摄图组件20从所述机体10的前部和两侧部被安装。也就是说，所述摄图组件20以前半周围的方式对所述操作区域100进行采集。
值得一提的是，所述摄图组件20所采集的图像基本地覆盖所述操作区域100。从所述摄图组件20所采集的图像可以反应所述机体10周围的影像，经过进一步地处理之后，所采集的图像可以表达所述操作区域100的状态信息。
优选地，所述摄图组件20所采集的图像可以为多份，所输出提供的图像为一份。更优选地，所述摄图组件20所输出的为所述操作区域100的状态信息。根据所述摄图组件20所采集的图像而得到所述机体10作业中的状态。
更多地，所述图像采集装置包括所述摄图组件20和一通信机30，其中所述通信机30被可通信地连接于所述摄图组件20，以从所述摄图组件20获得所采集图像信息。所述图像采集装置被安装于所述机体10。当所述机体10执行收割作业时，所述摄图组件20获取所述操作区域100的图像，所述通信机30将所述摄图组件20所提供的信息上传等待被获取。优选地，所述通信机30将所述摄图组件20所提供的一份图像传输至远程监控端，从而保证传输实时性。更优选地，所述通信机30将所述摄图组件20所采集的图像的处理后信息传输至远程控制端，从而保证传输的速度。
在本优选实施例中，所述摄图组件20被分布式置于所述机体10的前部与侧部。所述摄图组件20所采集的多份图像被拼接，进而通过所述通信机30传输。
具体地，如图2和图3所示，所述机体10包括一身架11、至少一轮架12和 一收割架13，其中所述收割架13被置于所述身架11的前部，其中所述轮架12支撑在所述身架11的底部。所述轮架12***作而带动所述收割架13和所述身架11向前对前方的作物进行切割和输送。所述图像采集装置被置于所述农割作业机器的所述身架11。更具体地，所述摄图组件20被置于所述身架11，且所述摄图组件20所朝向的所述操作区域100避免采集所述轮架12和所述收割架13。
更多地，在本优选实施例中，所述摄图组件20包括多个采集器和一处理器24。所述摄图组件20的多个采集器被预先朝向地安装于所述身架11的前部与侧部。所述采集器所获得的图像被所述处理器24处理汇总并分析。优选地，所述处理器24按照所述采集器所安装的位置进行图像处理。
具体地，所述采集器可以被实施为多个广角式摄像头，也可以被实施为一个全景式的摄像头。在本优选实施例中，所述采集器如图3称分布式的安装并分类地称为一远景采集器21、至少一近景采集器22、至少一侧景采集器23。所述远景采集器21和所述近景采集器22被适于所述身架11的前部，分别地朝向所述收割架13上方与下方。更具体地，所述远景采集器21沿着所述轮架12的上方采集，所述近景采集器22沿着所述轮架12与所述轮架12之间的位置进行采集。所述侧景采集器23被分别地安装于所述身架11的两侧，对两侧作物状态进行图像采集。
也就是说，所述远景采集器21所采集的图像与待收割的作物状态相关，所述近景采集器22所采集的图像与正在收割的作物相关。所述侧景采集器23所采集的图像与所述操作区域100的位置相关。
优选地，所述远景采集器21、所述近景采集器22，以及所述侧景采集器23分别地根据所安装的位置被编号。所述处理器24根据每个不同的编号得到相应的采集器的位置，配合图像所采集的位置进行图像处理。由于根据图像的位置来源而处理，处理速度得到提升。尤其是图像拼接处理，基于所述远景采集器21、所述近景采集器22，以及所述侧景采集器23的所在安装位置，可以较快地得到所述机身10周围的图像，从而获得所述操作区域100的状态。尤其，根据所述远景采集器21、所述近景采集器22，以及所述侧景采集器23的不同位置类型，所采集图像中的所述身架11、所述轮架12和所述收割架13可以被既定的忽略。例如，所述远景采集器21所采集的图像的下方边缘可以被认为是所述收割架13，而在处理程度中预先的剪切或者忽略。例如，所述近景采集器22所采集的图像 的上方边缘可以被认为是所述收割架13，而下方边缘可以被认为是轮架12，从而获得正在切割的作物的状态信息。例如，所述侧景采集器23所采集的图像的下方边缘可以被认为是轮架12，从而所述机体10的部分可以被排除。因此，所述远景采集器21、所述近景采集器22，以及所述侧景采集器23所采集的图像被所述处理器24快速地初步处理并拼接后得到所述操作区域100的状态信息。
具体地，本优选实施例的一种图像处理方法如图4至图6所示。用于农割作业机器的所述图像处理方法包括以下步骤：
I.采集一机体10前部与侧部的多份图像；
II.根据所采集的图像的位置，拼接图像；以及
III.根据所拼接的图像，获得围绕所述机体10的状态信息。
具体地，如图4所示，本优选实施例中，多个所述远景采集器21朝向所述收割架13的上方而采集。多个所述远景采集器21被以一定间距地安装，使得所述机体10前部的所述操作区域100被覆盖。也就是说，多个所述远景采集器21针对待收割的作物进行图像采集，即步骤I。
然后，根据多个所述远景采集器21的安装位置，依次地拼接所采集的图像，使得所述收割架13前方的待收割的作物图像被采集。值得一提的是，与现有的图像拼接不同，无需对每个图像进行识别而判断图像内容，而是直接根据安装的位置，或者每个图像中所述收割架13来拼接多个图像，从而获得所述机体10前方的全面性的图像，即步骤II。
而后，根据所述远景采集器21所采集整体性的所述机体10前方的图像，所述机体10前方的所述操作状态100被实时地获取并表征状态信息。优选地，所采集的图像可以被后续进一步地详细分析。
具体地，如图5所示，本优选实施例中，多个所述近景采集器22朝向所述收割架13的下方和所述轮架12之间而采集。多个所述近景采集器22被以一定间距地安装，使得所述机体10中部的所述操作区域100被覆盖。也就是说，多个所述近景采集器22针对正在收割的作物进行图像采集，即步骤I。
然后，根据多个所述近景采集器22的安装位置，依次地拼接所采集的图像，使得所述收割架13的下方的正在收割的作物图像被采集。值得一提的是，与现有的图像拼接不同，无需对每个图像进行识别而判断图像内容，而是直接根据安装的位置，或者每个图像中所述收割架13来拼接多个图像，从而获得所述机体 10正在工作的全面性的图像，即步骤II。
而后，根据所述近景采集器22所采集整体性的所述机体10工作图像，所述机体10的所述操作区域100中的切割状态被实时地获取并表征。优选地，所采集的图像可以被后续进一步地详细分析。
具体地，如图6所示，本优选实施例中，多个所述侧景采集器23朝向所述机体10的两侧而采集。多个所述侧景采集器23被分别安装在所述机体10的左右两侧，使得所述机体10两侧的作物状态被获取。也就是说，一般地，所述侧景采集器23向已经收割结束的场地和待收割的场地分别地进行图像采集，即步骤I。
然后，根据多个所述近侧景采集器23的安装位置，依次地拼接所采集的图像，使得已经收割结束的场地和待收割的场地图像被采集。值得一提的是，与现有的图像拼接不同，无需对每个图像进行识别而判断图像内容，而是直接根据安装的位置，例如安装在所述机体10左侧的侧景采集器23获得已经收割结束的作物状态，安装在所述机体10右侧的侧景采集器23获得待收割的场地的作物状态，从而获得所述机体10正在工作的左右作物的状态被获得，进而可以得到所述机体10所在的位置，即步骤II。
而后，根据所述侧景采集器23所采集的所述机体10侧部图像，所述机体10的前行方向和速度可以被获得，保证收割前行方向的准确，辅助提高收割效率。
更多地，所述远景采集器21、所述近景采集器22，以及所述侧景采集器23所采集的图像被最后地拼接，而得到所述机体10的所述操作区域100的图像。如图7所示，所述机体10的三侧状态被全面性的获得。优选地所述摄图组件20最后得到类似于所述机体10从上向下的观察角度，包括正在收割的作物和待收割的作物以及收割作业在场地中的位置。
更多地，所述图像处理方法，在步骤III之后，进一步地包括步骤：
远程地传输所述机体10的状态信息。
如图10所示，所述机体10的工作状态可以被远程地获取。进一步地，所述机体10被远程地控制收割作业中的参数。例如，所述近景采集器22和所述侧景采集器23多采集到的图像中得到作物与所述收割架13有一定差距，也就是所述机体10的工作较为偏离所需要的收割作物区域，需要将所述收割架13充分地填满，以实现高效率的作物。更具体地，例如，反馈的图像中，得到所述收割架 13中有较左侧80％正在收割作物，那么远程可以获得这一状态信息，并调整所述机体10的前行方向向左，使得所述收割架13充分地覆盖待收割的作物。
当然，对于所述机体10作业的调整可以由远程手动的操作，也可以由既定程序来调整，通过所述图像采集装置所得到的图像，辅助控制从而实现无人化的农业生产。
本发明的上述优选实施例的另一可行方式的用于农割作业机器的图像采集装置被阐述，如图8至图9，其中所述农割作业机器包括一机体10A和一摄图组件20A，其中所述机体10A和所述摄图组件20A的结构与上述优选实施例的所述农割作业机器的所述机体10和所述摄图组件20类似，本发明不再赘述。
与上述实施方式不同的是，图4至图6为根据同一类的安装位置拼接图像，最后三个不同的所述远景采集器21A、所述近景采集器22A，以及所述侧景采集器23A再合成最后的图像。本实施方式中，根据所述远景采集器21A、所述近景采集器22A，以及所述侧景采集器23A的位置，优选地拼接位置相近的所述采集器所采集的图像，再后续地处理为一份图像。
优先地，位于所述收割架13A上方和对应下方的所述远景采集器21A与所述近景采集器22A所采集的图像被先拼接。如图8所示，多个所述远景采集器21A与多个所述近景采集器22A被认为安装位置邻近。也就是说，所述远景采集器21A与所述近景采集器22A针对相对所述收割架13A同位置的待收割的作物和正在收割作物进行图像采集，即步骤I。
然后，根据位于所述收割架13A上方和对应下方的所述远景采集器21A与所述近景采集器22A的安装位置，依次地拼接所采集的图像，使得位于所述收割架13A上方和对应下方图像被采集。值得一提的是，与现有的图像拼接不同，无需对每个图像进行识别而判断图像内容，而是直接根据安装的位置，从而获得所述机体10A某一位置上的作物被切割前后的全面性的图像，即步骤II。
而后，根据所述机体10A某一位置的图像，所述机体10A前方的所述操作状态100A被实时地获取并表征状态信息。优选地，所采集的图像可以被后续进一步地详细分析，类似于传统人工切割的视角。
另外地，位于所述收割架13A端部下方的所述近景采集器22A与端部边缘所述侧景采集器23A所采集的图像被先拼接。如图9所示，一侧所述侧景采集器23A与多个所述近景采集器22A被认为安装位置邻近。也就是说，所述侧景 采集器23A与所述近景采集器22A针对相对所述收割架13A端部的正在收割作物和已经收割的作物进行图像采集，即步骤I。
然后，位于所述收割架13A端部下方的所述近景采集器22A与端部边缘所述侧景采集器23A的安装位置，依次地拼接所采集的图像，使得位于所述收割架13A端部的图像被采集。值得一提的是，与现有的图像拼接不同，无需对每个图像进行识别而判断图像内容，而是直接根据安装的位置，从而获得所述机体10A某一位置上的作物被切割前后的全面性的图像，即步骤II。
而后，根据所述机体10A某一位置的图像，所述机体10A前方的所述操作状态100A被实时地获取并表征状态信息。优选地，所采集的图像可以被后续进一步地详细分析，类似于传统人工切割的视角。本实施方式中，进一步地得到所述机体10A的边缘工作状态，便于辅助调整所述机体10A的工作位置和作业速度。
值得一提的是，所有的所述采集器，包括所述远景采集器21A、所述近景采集器22A，以及所述侧景采集器23A所采集的图像被最后地拼接，而得到所述机体10A的所述操作区域100A的图像。如图7所示，所述机体10A的作业前后状态被全面性的获得。优选地所述摄图组件20A最后得到类似于所述机体10A从上向下的观察角度，包括正在收割的作物和待收割的作物以及收割作业在场地中的位置。
如图10所示，本实施方式中所述机体10A的工作状态可以被远程地获取。进一步地，所述机体10A被远程地控制收割作业中的参数。例如，所述近景采集器22A和所述侧景采集器23A多采集到的图像中得到作物与所述收割架13A有一定差距，也就是所述机体10A的工作较为偏离所需要的收割作物区域，需要将所述收割架13A靠近待收割的作物，以实现高效率的作物。更具体地，例如，反馈的图像中，得到所述收割架13A中有较左侧有未被收割的作物在所述侧景采集器23A中显示，那么远程可以获得这一状态信息，并调整所述机体10A的前行方向向左，使得所述收割架13A充分地覆盖待收割的作物。
本领域的技术人员应理解，上述描述及附图中所示的本发明的实施例只作为举例而并不限制本发明。本发明的目的已经完整并有效地实现。本发明的功能及结构原理已在实施例中展示和说明，在没有背离所述原理下，本发明的实施方式可以有任何变形或修改。

Claims (26)

  1. 一农割作业机器，其特征在于，包括：
    一机体和一摄图组件，其中所述摄图组件被安装于所述机体表面，其中所述机体被启动而进行收割作业中，于所述机体的前部形成一操作区域，其中所述摄图组件的采集范围被覆盖式地安装于所述机体，并朝向所述操作区域进行图像采集，其中所述机体包括一身架、至少一轮架和一收割架，其中所述收割架被置于所述身架的前部，其中所述轮架支撑在所述身架的底部，其中所述轮架***作而带动所述收割架和所述身架向前对前方的作物进行切割和输送。
  2. 根据权利要求1所述的农割作业机器，其中所述摄图组件包括至少一采集器和一处理器，其中所述摄图组件的多个采集器被预先朝向地安装于所述身架的前部与侧部，其中所述采集器所获得的图像被所述处理器处理汇总并分析。
  3. 根据权利要求2所述的农割作业机器，其中所述采集器被实施为一个全景式的摄像头。
  4. 根据权利要求2所述的农割作业机器，其中所述采集器被实施为多个广角式摄像头。
  5. 根据权利要求2所述的农割作业机器，其中所述采集器为至少三个，其中所述摄图组件的多个采集器被预先朝向地安装于所述身架的前部与侧部，其中所述采集器所获得的图像被所述处理器处理汇总并分析。
  6. 根据权利要求4或者5所述述的农割作业机器，其中所述处理器按照所述采集器所安装的位置进行图像处理。
  7. 根据权利要求5所述的农割作业机器，其中所述采集器进一步地包括一远景采集器、至少一近景采集器、以及至少两侧景采集器，其中所述远景采集器和所述近景采集器被适于安装于所述身架的前部，且分别地朝向所述收割架上方与下方，其中所述侧景采集器被分别地安装于所述身架的两侧，且分别地朝向所述机体的两侧。
  8. 根据权利要求7所述的农割作业机器，其中所述远景采集器沿着所述轮架的上方保持朝向，其中所述近景采集器沿着所述轮架与所述轮架之间的位置而保持朝向。
  9. 根据权利要求5所述的农割作业机器，其中所述远景采集器、所述近景采集器，以及所述侧景采集器分别地根据所安装的位置被编号，其中所述处理器根据每个不同的编号得到相应的采集器的位置，配合图像所采集的位置进行图像 处理。
  10. 根据权利要求5所述的农割作业机器，其中多个所述远景采集器被以一定间距地安装，其中多个所述近景采集器被以一定间距地安装。
  11. 根据权利要求5所述的农割作业机器，其中所述处理器按照所述采集器所安装的位置进行图像处理，其中所述近景采集器所采集的图像被优先拼接，进而与其他所述采集器所采集的图像拼接。
  12. 根据权利要求5所述的农割作业机器，其中所述处理器按照所述采集器所安装的位置进行图像处理，其中所述远景采集器所采集的图像被优先拼接，进而与其他所述采集器所采集的图像拼接。
  13. 根据权利要求5所述的农割作业机器，其中位置邻近的多个所述采集器所采集图像被优先地拼接，进而与其他所述采集器所采集的图像拼接。
  14. 根据权利要求13所述的农割作业机器，其中位于所述收割架上方和对应下方的所述远景采集器与所述近景采集器所采集的图像被优先拼接。
  15. 根据权利要求13所述的农割作业机器，其中位于所述收割架端部下方的所述近景采集器与端部边缘所述侧景采集器所采集的图像被优先拼接。
  16. 一农割作业机器，其特征在于，包括：
    一机体和一图像采集装置，其中所述图像采集装置包括一摄图组件和一通信机，其中所述图像采集装置被安装于所述机体，其中所述通信机被可通信地连接于所述摄图组件，以从所述摄图组件获得所采集图像信息，进而辅助调整所述机体的工作位置和作业速度。
  17. 根据权利要求16所述的农割作业机器，其中所述通信机将所述摄图组件所提供的信息上传等待被获取。
  18. 根据权利要求16所述的农割作业机器，其中所述通信机将所述摄图组件所采集的图像的处理后信息传输至远程控制端。
  19. 根据权利要求17所述的农割作业机器，其中所述摄图组件包括至少一采集器和一处理器，其中所述摄图组件的多个采集器被预先朝向地安装于所述机体的前部与侧部，其中所述采集器所获得的图像被所述处理器处理汇总并分析。
  20. 根据权利要求19所述的农割作业机器，其中所述采集器为至少三个，其中所述摄图组件的多个采集器被预先朝向地安装于所述机体的前部与侧部，其中所述采集器所获得的图像被所述处理器处理汇总并传至所述通信机。
  21. 根据权利要求19所述的农割作业机器，其中所述处理器按照所述采集器所安装的位置进行图像处理。
  22. 一种用于农割作业机器的图像处理方法，其特征在于，包括以下步骤：
    I.采集一机体前部与侧部的多份图像；
    II.根据所采集的图像的位置，拼接图像；以及
    III.根据所拼接的图像，获得围绕所述机体的状态信息。
  23. 根据权利要求22所述图像处理方法，其中在步骤III之后，进一步地包括步骤：
    远程地传输所述机体的状态信息。
  24. 根据权利要求22所述图像处理方法，其中在步骤II中，按照被安装于所述机体的多个采集器所安装的位置进行图像处理。
  25. 根据权利要求24所述图像处理方法，其中在步骤II中，朝向相近的多个所述采集器所采集的图像被优先地拼接。
  26. 根据权利要求24所述图像处理方法，其中在步骤II中，位置邻近的多个所述采集器所采集图像被优先地拼接。
PCT/CN2020/080322 2019-04-09 2020-03-20 用于农割作业机器的图像采集装置及其处理方法 WO2020207225A1 (zh)

Priority Applications (5)

Application Number Priority Date Filing Date Title
JP2021559955A JP7397880B2 (ja) 2019-04-09 2020-03-20 農業刈り作業機械のための画像収集装置及びその処理方法
US17/602,726 US20220167555A1 (en) 2019-04-09 2020-03-20 Image acquisition device and processing method for agricultural harvesting operation machine
CA3135646A CA3135646A1 (en) 2019-04-09 2020-03-20 Image acquisition device and processing method for agricultural harvesting operation machine
EP20788625.0A EP3954190A4 (en) 2019-04-09 2020-03-20 IMAGE CAPTURE DEVICE AND PROCESSING METHOD FOR AGRICULTURAL HARVESTER
AU2020256953A AU2020256953B2 (en) 2019-04-09 2020-03-20 Image acquisition device and processing method for agricultural harvesting operation machine

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
CN201920473543.3U CN210466135U (zh) 2019-04-09 2019-04-09 农割作业机器
CN201910280375.0 2019-04-09
CN201910280375.0A CN110209155B (zh) 2019-04-09 2019-04-09 用于农割作业机器的图像采集装置及其处理方法
CN201920473543.3 2019-04-09

Publications (1)

Publication Number Publication Date
WO2020207225A1 true WO2020207225A1 (zh) 2020-10-15

Family

ID=72751525

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/080322 WO2020207225A1 (zh) 2019-04-09 2020-03-20 用于农割作业机器的图像采集装置及其处理方法

Country Status (6)

Country Link
US (1) US20220167555A1 (zh)
EP (1) EP3954190A4 (zh)
JP (1) JP7397880B2 (zh)
AU (1) AU2020256953B2 (zh)
CA (1) CA3135646A1 (zh)
WO (1) WO2020207225A1 (zh)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117765085B (zh) * 2024-02-22 2024-06-07 华南农业大学 一种茶芽叶采摘远近交替定位方法

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2010029135A (ja) * 2008-07-30 2010-02-12 Iseki & Co Ltd コンバインの表示装置
US20110072773A1 (en) * 2009-09-30 2011-03-31 Cnh America Llc Automatic display of remote camera image
CN105682446A (zh) * 2013-10-01 2016-06-15 洋马株式会社 联合收割机
CN105991929A (zh) * 2016-06-21 2016-10-05 浩云科技股份有限公司 全空间摄像机外参标定及全空间视频拼接方法
CN110209155A (zh) * 2019-04-09 2019-09-06 丰疆智能科技股份有限公司 用于农割作业机器的图像采集装置及其处理方法
CN210130123U (zh) * 2019-01-04 2020-03-10 丰疆智能科技股份有限公司 带有云台摄像装置的收割机

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2015070799A (ja) * 2013-10-01 2015-04-16 ヤンマー株式会社 コンバイン
JP6163460B2 (ja) 2014-06-30 2017-07-12 ヤンマー株式会社 随伴作業システム
WO2017221641A1 (ja) 2016-06-22 2017-12-28 コニカミノルタ株式会社 植物生育指標測定装置、該方法および該プログラム
JP2018186728A (ja) * 2017-04-28 2018-11-29 井関農機株式会社 コンバイン
US10757859B2 (en) * 2017-07-20 2020-09-01 Deere & Company System for optimizing platform settings based on crop state classification

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2010029135A (ja) * 2008-07-30 2010-02-12 Iseki & Co Ltd コンバインの表示装置
US20110072773A1 (en) * 2009-09-30 2011-03-31 Cnh America Llc Automatic display of remote camera image
CN105682446A (zh) * 2013-10-01 2016-06-15 洋马株式会社 联合收割机
CN105991929A (zh) * 2016-06-21 2016-10-05 浩云科技股份有限公司 全空间摄像机外参标定及全空间视频拼接方法
CN210130123U (zh) * 2019-01-04 2020-03-10 丰疆智能科技股份有限公司 带有云台摄像装置的收割机
CN110209155A (zh) * 2019-04-09 2019-09-06 丰疆智能科技股份有限公司 用于农割作业机器的图像采集装置及其处理方法

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP3954190A4 *

Also Published As

Publication number Publication date
US20220167555A1 (en) 2022-06-02
AU2020256953A1 (en) 2021-12-02
EP3954190A4 (en) 2023-01-04
CA3135646A1 (en) 2020-10-15
AU2020256953B2 (en) 2024-02-29
EP3954190A1 (en) 2022-02-16
JP7397880B2 (ja) 2023-12-13
JP2022528260A (ja) 2022-06-09

Similar Documents

Publication Publication Date Title
US9137943B2 (en) Trainable autonomous lawn mower
CN110250123B (zh) 基于图像识别的低密度储粮害虫实时监测***
CN203646084U (zh) 智能多功能作业车
WO2020207225A1 (zh) 用于农割作业机器的图像采集装置及其处理方法
EP3841868A1 (en) System and method for windrow path planning
CN101646068A (zh) 植物病虫害信息获取***及方法
JP2019028688A (ja) 自律走行コンバインの収穫システム
US11930726B2 (en) Machine-vision system for tracking and quantifying missed tassel during a detasseling operation
CN207321927U (zh) 基于机器视觉的菠萝采摘装置
CN210466135U (zh) 农割作业机器
WO2023231408A1 (zh) 一种搭载在无人机上的水果自动采收装置及其控制方法
CN110209155B (zh) 用于农割作业机器的图像采集装置及其处理方法
CN107047083A (zh) 一种打顶机器人及其实现方法
EP3939407A1 (en) System and method of assisted or automated grain unload synchronization
US11980135B2 (en) Method and system for harvesting, packaging, and tracking crop material
CN110209154B (zh) 自动收割机的残留收割路径规划***及其方法
CN208079864U (zh) 一种无人机双目视觉定位番茄采摘装置
CN209824456U (zh) 一种对刀拉茎式不对行玉米收获割台装置
CN116938972A (zh) 一种基于5g通信与机器学习的端云协同农情检测***
EP3909417A1 (en) Agricultural raking system and method for automatic settings
EP3939408A1 (en) System and method of assisted or automated grain unload synchronization
CN113607096A (zh) 一种ai应用于农作物倒伏检测的方法
CN105988469A (zh) 自动行走设备
CN116098055B (zh) 玉米去雄除弱机器人及利用其进行去雄除弱的方法
CN208739625U (zh) 一种智能果蔬自动采收装置

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20788625

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 3135646

Country of ref document: CA

ENP Entry into the national phase

Ref document number: 2021559955

Country of ref document: JP

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 2020788625

Country of ref document: EP

Effective date: 20211109

ENP Entry into the national phase

Ref document number: 2020256953

Country of ref document: AU

Date of ref document: 20200320

Kind code of ref document: A