CN113641773A - Agricultural machinery behavior visualization marking method for driving track - Google Patents
Agricultural machinery behavior visualization marking method for driving track Download PDFInfo
- Publication number
- CN113641773A CN113641773A CN202110928338.3A CN202110928338A CN113641773A CN 113641773 A CN113641773 A CN 113641773A CN 202110928338 A CN202110928338 A CN 202110928338A CN 113641773 A CN113641773 A CN 113641773A
- Authority
- CN
- China
- Prior art keywords
- data
- agricultural machinery
- track
- point
- marking
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 22
- 238000012800 visualization Methods 0.000 title claims description 6
- 230000006399 behavior Effects 0.000 claims abstract description 40
- 238000004140 cleaning Methods 0.000 claims abstract description 19
- 230000011218 segmentation Effects 0.000 claims abstract description 13
- 238000002372 labelling Methods 0.000 claims abstract description 10
- 230000000007 visual effect Effects 0.000 claims abstract description 7
- 230000002159 abnormal effect Effects 0.000 claims description 3
- 230000000694 effects Effects 0.000 description 4
- 238000012952 Resampling Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000007418 data mining Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 230000028838 turning behavior Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000009193 crawling Effects 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- DMBHHRLKUKUOEG-UHFFFAOYSA-N diphenylamine Chemical compound C=1C=CC=CC=1NC1=CC=CC=C1 DMBHHRLKUKUOEG-UHFFFAOYSA-N 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/29—Geographical information databases
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/02—Agriculture; Fishing; Forestry; Mining
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Mining & Mineral Resources (AREA)
- Economics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Agronomy & Crop Science (AREA)
- Animal Husbandry (AREA)
- Marine Sciences & Fisheries (AREA)
- Data Mining & Analysis (AREA)
- Remote Sensing (AREA)
- Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Guiding Agricultural Machines (AREA)
Abstract
The invention discloses a visual marking method of agricultural machinery behaviors for a driving track, belonging to the technical field of computers. The method comprises the following steps: step 1: acquiring data; acquiring agricultural machinery data according to a data format of an agricultural machinery track; step 2: data cleaning; cleaning wrong data sampled by the Beidou terminal; and step 3: dividing the field and preliminarily marking; carrying out primary labeling on the field segmentation aiming at the cleaned data in the step 2; and 4, step 4: amplifying the data which are preliminarily marked with inaccuracy in the step 3 and carrying out fine adjustment; and 5: and uploading the field division data with accurate marking and marking the driving behavior of the agricultural machinery in the field. The invention can carry out the manual marking of the offline agricultural machinery behaviors, effectively solves the problems of fussy and inaccurate manual recording on site and achieves the purpose of saving time and labor.
Description
Technical Field
The invention relates to the technical field of computers, in particular to a visual marking method for agricultural machinery behaviors aiming at a driving track.
Background
The agricultural machine travel track data is one of the most basic data in agricultural machine operation management, and is agricultural machine positioning information (e.g., longitude, latitude, direction, etc.) obtained by using a global positioning system. The driving behavior of the agricultural machine can be analyzed according to the travel track of the agricultural machine, and relevant information of certain behaviors (for example, time spent on turning around in the field) can be obtained. The method has great propulsion effect on designing agricultural machinery management decision service models and promoting agricultural machinery informatization.
Agricultural machine behavior analysis aiming at massive driving tracks needs a data mining technology, and the agricultural machine driving tracks can be automatically segmented and agricultural machine behaviors of all segments can be given. However, these agricultural machine trajectory mining techniques often employ fully supervised machine learning or deep learning techniques for model training, which require a certain amount of manual annotation data. At present, manual marking of agricultural machinery driving behaviors mainly depends on-site manual recording. Namely, in the driving process of the agricultural machinery, the position information and the driving behavior of the current agricultural machinery are recorded on site in real time by a driver or an assistant. The field manual labeling method can directly utilize the information of the agricultural machinery operation scene, but is time-consuming and labor-consuming. Particularly, when the agricultural machinery is in a state of long-time running, problems such as data error and leakage or inaccuracy are easily caused. Therefore, the invention provides an off-site artificial marking method for agricultural machinery driving behaviors, which is characterized in that a Beidou terminal installed on an agricultural machinery is used for collecting agricultural machinery track data, an agricultural machinery operation scene is restored through a visual UI interface, a corresponding agricultural machinery driving track is drawn, and then the artificial marking of the agricultural machinery behaviors is carried out.
Disclosure of Invention
The invention aims to provide an agricultural machinery behavior visualization marking method aiming at a driving track, which is characterized by comprising the following steps of:
step 1: acquiring data; acquiring agricultural machinery data according to a data format of an agricultural machinery track;
step 2: data cleaning; cleaning wrong data sampled by the Beidou terminal;
and step 3: dividing the field and preliminarily marking; carrying out primary labeling on the field segmentation aiming at the cleaned data in the step 2;
and 4, step 4: amplifying the data which are preliminarily marked with inaccuracy in the step 3 and carrying out fine adjustment;
and 5: and uploading the field division data with accurate marking and marking the driving behavior of the agricultural machinery in the field.
The agricultural data includes longitude, latitude, and date.
The step 2 specifically comprises the following substeps:
step 21: cleaning off a point with the time interval of 0S between two points, and keeping the first point;
step 22: cleaning a point with the same longitude and latitude and the speed of 0, and keeping the point as the first point;
step 23: cleaning continuous points with the same longitude and latitude, the same speed and different speeds from 0, and keeping the point as the first point;
step 24: cleaning out continuous points with different longitudes and latitudes and the speed of 0, and keeping the point as the first point;
step 25: directly deleting the points with abnormal longitude and latitude;
step 26: and calculating the direction included angle of the agricultural machinery track data with the missing direction angle according to the longitude and latitude between the front point and the rear point, and taking the direction included angle as the direction data of the front point.
The step 3 specifically comprises the following substeps:
step 31: checking the sorted and cleaned data to determine that the data contains longitude, latitude and date fields;
step 32: uploading track data, drawing track data points into a Google image map, and connecting lines according to time sequence;
step 33: and selecting a plurality of points to form a closed polygon around the range of the interested track data, wherein the track points in the polygon are preliminarily marked farmland track data.
The invention has the beneficial effects that:
the visual agricultural machinery behavior marking method combined with the Beidou terminal furthest restores the agricultural machinery operation scene by utilizing the visual technology, so that the offline manual marking of the agricultural machinery behavior can be performed, the problems of complexity and inaccuracy of field manual recording are effectively solved, and the purpose of saving time and labor is achieved.
Drawings
FIG. 1 is a flow chart of an agricultural machinery behavior visualization marking method for a driving track according to the present invention;
FIG. 2 is a diagram of the results of rough segmentation of the field;
FIGS. 3(a) and 3(b) are graphs showing the results of fine-tuning the preliminary field segmentation;
fig. 4 is a diagram of the effect of segmentation of different agricultural machinery driving behaviors.
Detailed Description
The invention provides a visual marking method of agricultural machinery behaviors for a driving track, which is further explained by combining the accompanying drawings and specific embodiments.
The invention can draw a path and mark data according to the input agricultural machinery track data, thereby distinguishing different driving behaviors of the agricultural machinery. The work flow diagram is shown in figure 1:
step one, data acquisition:
the method comprises the steps of firstly compiling a data crawling program, and acquiring agricultural machinery data according to a data format of an agricultural machinery track, wherein fields required by the most basic data are longitude, latitude and date.
Step two, data cleaning:
in the acquisition process of agricultural machinery driving track, the big dipper terminal often can appear resampling, static orbit etc. sampling error condition, in order to avoid sampling error to cause the influence to follow-up classification method, need carry out corresponding clearance to data in advance, and the data clearance flow is:
(1) the resampling type: and cleaning the point with the time interval of 0S between the two points, and keeping the first point.
(2) Type of stationary trajectory: the same longitude and latitude are cleared, and the point reserved point with the speed of 0 is still the first one.
(3) The resampling type: and cleaning continuous points with the same longitude and latitude, the same speed and different speeds from 0, and keeping the point as the first one.
(4) Static drift type: and (4) cleaning out continuous points with different longitudes and latitudes and 0 speed, and keeping the first point.
(5) The longitude and latitude abnormity type is as follows: because the data acquisition area is a plurality of provinces and cities in China, when the latitude and longitude range exceeds the range of China, the condition that the acquired point is abnormal in latitude and longitude needs to be cleaned, and the cleaning mode is to directly delete the data.
(6) Directional data miss type: because the acquisition requirements of the Beidou terminals are different, the acquisition platforms of agricultural machine data are different, and direction angle data of certain agricultural machine trajectory data are lost, the direction included angle of the agricultural machine trajectory data is calculated according to longitude and latitude data between a front point and a rear point and is used as the direction data of the front point.
Step three, field division and preliminary labeling:
the sorted and cleaned data is checked to determine whether the data contains the required longitude and latitude, date and other fields. Secondly, uploading track data through a program, drawing the track data points into a Google image map, and connecting the track data points according to the time sequence. And finally, clicking a plurality of points around the range of the track data to be marked according to the range of the track data to be marked by the user after clicking to start circle selection, and enclosing a closed polygon, wherein the track points in the polygon are marked farmland track data. The agricultural machinery driving track display method can display the agricultural machinery driving track by combining the Google image map according to the input track data, and carry out preliminary rough segmentation to separate the track of the farmland from the track of the road, wherein the segmentation result is shown in figure 2.
Step four, fine adjustment of the primary labeling data:
the third field segmentation marking program can roughly distinguish which of the track data are farmlands and which are roads, but because the data volume of one track data is too large or special driving behaviors of some agricultural machinery, such as turning of the agricultural machinery, waiting for low-speed driving areas such as traffic lights and the like can cause the situation that the track point density is too high, the range defined in the third step is easy to calibrate the areas with too high track point density, so that the field segmentation can not be marked very accurately, and the subsequent data mining effect can be directly influenced. Therefore, on the basis of the data roughly labeled by dividing the field in the third step, the data parts belonging to the field and the road are respectively extracted by the fine-tuning labeling program in the fourth step, and are displayed by pictures to be finely adjusted. Firstly, the program reads the data preliminarily marked in the third step, and each track point (the label value is 1) marked as the field is read and displayed, so that the track data in the field can be amplified and marked easily compared with the third step. Similar to the previous step, labeling is performed in a circle selection mode, 4 points are clicked around the range of the track data which the user wants to label, a similar rectangle is formed, and the track points in the rectangle are points corrected to be roads (the label value is 0). Clicking a mouse wheel key and pressing any key of a keyboard to store and switch to the next farmland track.
In the case of inaccurate preliminary field segmentation caused by excessive track data or special driving behaviors of agricultural machinery, preliminary labeled data is extracted and micro-adjustment is performed in a circle selection mode to ensure that segmentation labeling of the field and the road is as accurate as possible, and the segmentation result is shown in fig. 3(a) and 3 (b).
Step five, marking the driving behaviors of farm machinery in the field:
as the main operation and the movement range of the agricultural machine are in the farmland, the driving behavior of the agricultural machine in the farmland has great research significance, and the driving behavior in the agricultural machine farmland needs to be labeled. The driving behaviors in the agricultural machinery field can be roughly divided into: field work behavior, field turn around behavior, and field turn behavior. Marking the driving behavior of the agricultural machine in the field, taking the distinguishing and marking of the operation behavior of the agricultural machine in the field and the turning behavior of the agricultural machine in the field as an example (the marking of the turning behavior in the agricultural machine field is similar to the marking), and the specific method comprises the following steps: uploading the relatively accurate field dividing data marked in the fourth step, reading and displaying each track point (label value is 1) marked as a field in the data by the program, marking in a click mode, namely marking a section of track, clicking the start point and the stop point of the track section, changing the color of the track points of the track section including the start point and the stop point, automatically marking the label value as 2 by the data, and storing and switching to the next track in the field by any key of a keyboard.
Marking the track data of the farm machinery in the field aiming at the complex and changeable farm machinery driving behaviors in the field, and calibrating the three driving behaviors of the farm machinery in the field by adopting a click selection mode to segment different farm machinery driving behaviors in the field, wherein the effect is shown in figure 4.
The present invention is not limited to the above embodiments, and any changes or substitutions that can be easily made by those skilled in the art within the technical scope of the present invention are also within the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (4)
1. A visual marking method of agricultural machinery behaviors aiming at a driving track is characterized by comprising the following steps:
step 1: acquiring data; acquiring agricultural machinery data according to a data format of an agricultural machinery track;
step 2: data cleaning; cleaning wrong data sampled by the Beidou terminal;
and step 3: dividing the field and preliminarily marking; carrying out primary labeling on the field segmentation aiming at the cleaned data in the step 2;
and 4, step 4: amplifying the data which are preliminarily marked with inaccuracy in the step 3 and carrying out fine adjustment;
and 5: and uploading the field division data with accurate marking and marking the driving behavior of the agricultural machinery in the field.
2. The agricultural machinery behavior visualization labeling method for the driving track according to claim 1, wherein the agricultural machinery data comprises longitude, latitude and date.
3. The agricultural machinery behavior visualization marking method for the driving track according to claim 1, wherein the step 2 specifically comprises the following substeps:
step 21: cleaning off a point with the time interval of 0S between two points, and keeping the first point;
step 22: cleaning a point with the same longitude and latitude and the speed of 0, and keeping the point as the first point;
step 23: cleaning continuous points with the same longitude and latitude, the same speed and different speeds from 0, and keeping the point as the first point;
step 24: cleaning out continuous points with different longitudes and latitudes and the speed of 0, and keeping the point as the first point;
step 25: directly deleting the points with abnormal longitude and latitude;
step 26: and calculating the direction included angle of the agricultural machinery track data with the missing direction angle according to the longitude and latitude between the front point and the rear point, and taking the direction included angle as the direction data of the front point.
4. The method for visually labeling agricultural machinery behaviors for a driving track according to claim 1, wherein the step 3 specifically comprises the following substeps:
step 31: checking the sorted and cleaned data to determine that the data contains longitude, latitude and date fields;
step 32: uploading track data, drawing track data points into a Google image map, and connecting lines according to time sequence;
step 33: and selecting a plurality of points to form a closed polygon around the range of the interested track data, wherein the track points in the polygon are preliminarily marked farmland track data.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110928338.3A CN113641773A (en) | 2021-08-13 | 2021-08-13 | Agricultural machinery behavior visualization marking method for driving track |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110928338.3A CN113641773A (en) | 2021-08-13 | 2021-08-13 | Agricultural machinery behavior visualization marking method for driving track |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113641773A true CN113641773A (en) | 2021-11-12 |
Family
ID=78421377
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110928338.3A Pending CN113641773A (en) | 2021-08-13 | 2021-08-13 | Agricultural machinery behavior visualization marking method for driving track |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113641773A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114442623A (en) * | 2022-01-20 | 2022-05-06 | 中国农业大学 | Agricultural machinery operation track field segmentation method based on space-time diagram neural network |
CN114998744A (en) * | 2022-07-18 | 2022-09-02 | 中国农业大学 | Agricultural machinery track field segmentation method based on motion and vision dual-feature fusion |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104240134A (en) * | 2014-09-04 | 2014-12-24 | 奇瑞重工股份有限公司 | Agricultural machine service method and system based on farmland states |
CN107657637A (en) * | 2017-09-25 | 2018-02-02 | 中国农业大学 | A kind of agricultural machinery working area acquisition methods |
CN109684384A (en) * | 2018-12-29 | 2019-04-26 | 太原科技大学 | A kind of track data space-time density Analysis System and its analysis method |
CN110622200A (en) * | 2017-06-26 | 2019-12-27 | 株式会社久保田 | Farmland map generation system |
CN112905576A (en) * | 2021-03-02 | 2021-06-04 | 中国农业大学 | Method and system for determining farmland and road based on agricultural machinery operation track |
-
2021
- 2021-08-13 CN CN202110928338.3A patent/CN113641773A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104240134A (en) * | 2014-09-04 | 2014-12-24 | 奇瑞重工股份有限公司 | Agricultural machine service method and system based on farmland states |
CN110622200A (en) * | 2017-06-26 | 2019-12-27 | 株式会社久保田 | Farmland map generation system |
CN107657637A (en) * | 2017-09-25 | 2018-02-02 | 中国农业大学 | A kind of agricultural machinery working area acquisition methods |
CN109684384A (en) * | 2018-12-29 | 2019-04-26 | 太原科技大学 | A kind of track data space-time density Analysis System and its analysis method |
CN112905576A (en) * | 2021-03-02 | 2021-06-04 | 中国农业大学 | Method and system for determining farmland and road based on agricultural machinery operation track |
Non-Patent Citations (2)
Title |
---|
王培等: "基于农机空间运行轨迹的作业状态自动识别试验", 《农业工程学报》, vol. 31, no. 3, pages 56 * |
肖敬等: "基于时空立方体的农机运动轨迹分割方法", 《江苏农业科学》, vol. 46, no. 20, pages 244 * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114442623A (en) * | 2022-01-20 | 2022-05-06 | 中国农业大学 | Agricultural machinery operation track field segmentation method based on space-time diagram neural network |
CN114442623B (en) * | 2022-01-20 | 2023-10-24 | 中国农业大学 | Agricultural machinery operation track Tian Lu segmentation method based on space-time diagram neural network |
CN114998744A (en) * | 2022-07-18 | 2022-09-02 | 中国农业大学 | Agricultural machinery track field segmentation method based on motion and vision dual-feature fusion |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109446369B (en) | Interaction method and system for semi-automatic image annotation | |
CN113641773A (en) | Agricultural machinery behavior visualization marking method for driving track | |
US10281284B2 (en) | Hybrid road network and grid based spatial-temporal indexing under missing road links | |
Lowe | Night lights and ArcGis: A brief guide | |
US20030120421A1 (en) | High resolution tracking of mobile assets | |
US20210001891A1 (en) | Training data generation for dynamic objects using high definition map data | |
CN109993064B (en) | Method for extracting connection path between road network nodes in picture | |
CN111695488A (en) | Interest plane identification method, device, equipment and storage medium | |
CN113052369B (en) | Intelligent agricultural machinery operation management method and system | |
Bosurgi et al. | Pavement condition information modelling in an I-BIM environment | |
CN111858814A (en) | Method, device and equipment for repairing motion trail and storage medium | |
CN111582518A (en) | Automatic generation method and device for power inspection report and terminal equipment | |
CN111854766B (en) | Road information determination method and device, electronic equipment and storage medium | |
CN113656477A (en) | Method for verifying and fusing multi-source heterogeneous data of homeland space | |
Graser | An exploratory data analysis protocol for identifying problems in continuous movement data | |
Gu et al. | A road extraction method for intelligent dispatching based on MD-LinkNeSt network in open-pit mine | |
Ding et al. | Mit-avt clustered driving scene dataset: Evaluating perception systems in real-world naturalistic driving scenarios | |
Cho et al. | Web-based agricultural infrastructure digital twin system integrated with GIS and BIM concepts | |
CN111581306B (en) | Driving track simulation method and device | |
Sehra et al. | Quality assessment of crowdsourced data against custom recorded map data | |
Bhatnagar et al. | Mapping wheel-ruts from timber harvesting operations using deep learning techniques in drone imagery | |
Yang et al. | Detecting road network errors from trajectory data with partial map matching and bidirectional recurrent neural network model | |
CN109902139B (en) | R-tree-based track data compression method | |
CN110647605B (en) | Method and device for mining traffic light data based on trajectory data | |
Braga-Neto et al. | Spatial data for fungal specimens: retrospective georeferencing and practical recommendations for mycologists |
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 |