CN115115548A - Point cloud repairing method and device, electronic equipment and storage equipment - Google Patents

Point cloud repairing method and device, electronic equipment and storage equipment Download PDF

Info

Publication number
CN115115548A
CN115115548A CN202210769414.5A CN202210769414A CN115115548A CN 115115548 A CN115115548 A CN 115115548A CN 202210769414 A CN202210769414 A CN 202210769414A CN 115115548 A CN115115548 A CN 115115548A
Authority
CN
China
Prior art keywords
point cloud
target object
point
contour
missing area
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
Application number
CN202210769414.5A
Other languages
Chinese (zh)
Inventor
李秋凤
司永胜
曹玉凤
王明亚
吴春会
刘敏
冯凡
李建国
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Hebei Agricultural University
Original Assignee
Hebei Agricultural University
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
Application filed by Hebei Agricultural University filed Critical Hebei Agricultural University
Priority to CN202210769414.5A priority Critical patent/CN115115548A/en
Publication of CN115115548A publication Critical patent/CN115115548A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G06T5/77—
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/12—Edge-based segmentation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/13—Edge detection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10028—Range image; Depth image; 3D point clouds

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Image Processing (AREA)

Abstract

The application discloses a point cloud repairing method, a point cloud repairing device, electronic equipment and storage equipment, wherein the method comprises the following steps: acquiring a first point cloud of a target object; extracting a plurality of contour points of a target object from a first point cloud of the target object according to the first point cloud; generating a plurality of first contour curves of the target object according to the plurality of contour points of the target object; point supplementing is carried out on the plurality of first contour curves of the target object, and a plurality of second contour curves after point supplementing are obtained; obtaining second point clouds of the target object according to the point clouds on the plurality of second contour curves after point compensation and the point clouds in the first contour curve in the first point cloud; detecting a point cloud missing region in a second point cloud of the target object, and identifying the boundary of the point cloud missing region; and repairing the point cloud in the point cloud missing area according to the existing point cloud with a preset position relation with the boundary of the point cloud missing area. The method can realize the repair of the large-area non-closed missing point cloud.

Description

Point cloud repairing method and device, electronic equipment and storage equipment
Technical Field
The application relates to the technical field of image processing, in particular to a point cloud repairing method and device, electronic equipment and storage equipment.
Background
The cattle breeding industry plays an important strategic position in national economy of China, in recent years, the large-scale breeding of cattle grows rapidly, and the body type score and the health condition of cattle have important reference significance for the large-scale breeding of cattle.
The three-dimensional point cloud data contains important geometric information such as size and appearance, so that the complete three-dimensional point cloud data of the cattle body is obtained, and the method has very practical application value in scoring the body condition of the cattle body, measuring the body size and weight, evaluating the health and the like. At present, point cloud data can be acquired through a laser scanner, a depth camera and the like, and due to the accuracy of equipment and the complexity of an acquisition environment, the acquired data has the problems of partial deletion and the like. Especially, the ox body field collection environment is very complicated, and when the ox body data is collected, in order to stipulate the activity range of the ox body, most of the data need set up the collection channel that the railing was built, leads to the ox body point cloud data that obtains to have a large tracts of land, and for the disappearance problem of non-enclosed, seriously influences the point cloud registration and the three-dimensional reconstruction effect of the ox body, and then influences the precision of the body chi parameter measurement of follow-up ox body.
Most of the existing point cloud repairing methods aim at repairing closed point cloud holes and cannot solve the problems. Therefore, it is desirable to provide a method for repairing a large area of missing point clouds.
Disclosure of Invention
The application provides a point cloud repairing method, which can repair non-closed large-area missing point clouds.
The application provides a point cloud repairing method, which comprises the following steps: acquiring a first point cloud of a target object; extracting a plurality of contour points of the target object from a first point cloud of the target object according to the first point cloud; generating a plurality of first contour curves of the target object according to the plurality of contour points of the target object; point supplementing is carried out on the plurality of first contour curves of the target object, and a plurality of second contour curves after point supplementing are obtained; obtaining second point clouds of the target object according to the point clouds on the plurality of second contour curves after point compensation and the point clouds in the first contour curve in the first point cloud; detecting a point cloud missing region in a second point cloud of the target object, and identifying a boundary of the point cloud missing region; and repairing the point cloud in the point cloud missing area according to the existing point cloud with a preset position relation with the boundary of the point cloud missing area.
Optionally, the obtaining a first point cloud of the target object includes:
the method comprises the steps of obtaining video data of a target object, and converting the video data into point clouds serving as first point clouds of the target object.
Optionally, the extracting, according to the first point cloud of the target object, a plurality of contour points of the target object from the first point cloud includes:
slicing the first point cloud of the target object to obtain a plurality of first point cloud slices of the target object;
calculating the number of point clouds in each first point cloud slice;
after removing the point cloud slice with the point cloud number of the minimum value, extracting the point cloud in the reserved point cloud slice to be used as a third point cloud of the target object;
extracting a plurality of contour points of the target object from the third point cloud of the target object.
Optionally, the slicing the first point cloud of the target object to obtain a plurality of first point cloud slices of the target object includes:
projecting the first point cloud of the target object onto a plane to obtain a side view point cloud of the target object;
and slicing the side view point cloud of the target object along the longitudinal axis direction to obtain a plurality of first point cloud slices of the target object.
Optionally, the extracting a plurality of contour points of the target object from the third point cloud of the target object includes:
slicing the third point cloud of the target object along a transverse axis to obtain a plurality of second point cloud slices of the target object;
and detecting a maximum value point and a minimum value point of each second point cloud slice in the longitudinal axis direction as a plurality of contour points of the target object.
Optionally, the generating a plurality of first contour curves of the target object according to the plurality of contour points of the target object includes:
and performing curve fitting on the plurality of contour points of the target object by adopting a preset algorithm to generate a plurality of first contour curves of the target object.
Optionally, the performing point supplementation on the plurality of first contour curves of the target object to obtain a plurality of second contour curves after point supplementation includes:
calculating a minimum interval of transverse distances between point clouds in a third point cloud of the target object;
comparing a lateral distance between adjacent point clouds on a first contour curve of the target object to a size of a minimum separation;
and if the transverse distance between the adjacent point clouds on the first contour curve of the target object is greater than the minimum interval, performing point compensation on the plurality of first contour curves of the target object to obtain a plurality of second contour curves after point compensation.
Optionally, if the lateral distance between the adjacent point clouds on the first contour curves of the target object is greater than the minimum interval, performing point compensation on the plurality of first contour curves of the target object, including:
and selecting points on the first contour curve of the target object as datum points, and performing point supplementation on the first contour curve of the target object by taking the minimum interval as a unit.
Optionally, the detecting a point cloud missing region in the second point cloud of the target object, and identifying a boundary of the point cloud missing region includes:
slicing the second point cloud of the target object along a longitudinal axis to obtain a plurality of third point cloud slices of the target object;
calculating the number of point clouds in each third point cloud slice;
selecting a point cloud slice with the point cloud number of the minimum value as a point cloud missing area in a second point cloud of the target object;
and detecting the boundary characteristic points of the point cloud missing area, and identifying the boundary of the point cloud missing area according to the boundary characteristic points of the point cloud missing area.
Optionally, the repairing the point cloud in the point cloud missing area according to an existing point cloud having a preset position relationship with the boundary of the point cloud missing area includes:
and repairing the point cloud in the point cloud missing area according to the existing point cloud which has a preset position relation with the boundary of the point cloud missing area and the minimum interval and a preset algorithm.
Optionally, the repairing the point cloud in the point cloud missing region according to an existing point cloud having a preset position relationship with the boundary of the point cloud missing region and a minimum interval and according to a preset algorithm includes:
dividing the point cloud missing region along the direction of a longitudinal axis and/or a transverse axis according to the minimum interval to obtain m and/or n divided point cloud missing regions, wherein m and n are integers more than or equal to 1;
and selecting the existing point cloud with a preset position relation with the boundary of the point cloud missing area as a reference, and performing point cloud repair on each segmented point cloud missing area by taking the minimum interval as a unit.
Optionally, the number m of the point cloud missing region divided along the longitudinal axis direction is obtained by using the following formula:
Figure BDA0003723415060000031
wherein, X _ max and X _ min are respectively the maximum value and the minimum value of the boundary point of the point cloud missing region in the longitudinal axis direction, d is the minimum interval, and m is the number of parts of the point cloud missing region divided in the longitudinal axis direction;
obtaining the number n of parts of the point cloud missing region segmented along the direction of the transverse axis by adopting the following formula:
Figure BDA0003723415060000041
and Y _ max and Y _ min are respectively the maximum value and the minimum value of the boundary points of the point cloud missing region in the transverse axis direction, d is the minimum interval, and n is the number of parts of the point cloud missing region divided in the transverse axis direction.
Optionally, the selecting an existing point cloud having a preset position relationship with the boundary of the point cloud missing region as a reference, and performing point cloud repairing on each segmented point cloud missing region by using the minimum interval as a unit includes:
obtaining a longitudinal axis coordinate x and a transverse axis coordinate y of the repaired point cloud by adopting the following formula:
x＝X_min+i*d
y＝Y_min+j*d
i∈(1,m-1)
j∈(1,n-1)
wherein X is a longitudinal coordinate value of the repaired point cloud, Y is an abscissa value of the repaired point cloud, the z coordinate of the repaired point cloud is a median value of z coordinates of other point clouds in an ith block area divided by the point cloud missing area along the longitudinal axis direction, X _ min is a minimum value of the boundary point of the point cloud missing area along the longitudinal axis direction, Y _ min is a minimum value of the boundary point of the point cloud missing area along the transverse axis direction, d is a minimum interval, m is the number of parts of the point cloud missing area divided along the longitudinal axis direction, n is the number of parts of the point cloud missing area divided along the transverse axis direction, i is the ith block divided by the point cloud missing area along the longitudinal axis direction, the range of i is greater than or equal to and less than or equal to m-1, j is the jth block divided by the point cloud missing area along the transverse axis direction, and the range of j is greater than or equal to and less than or equal to n-1.
Optionally, the method is applied to a server, and the method further includes:
obtaining a fourth point cloud of the target object according to the repaired point cloud;
analyzing the fourth point cloud to obtain health state data of the target object;
and sending the health state data of the target object to a terminal.
This application provides a some cloud prosthetic devices simultaneously, includes:
the first point cloud obtaining unit is used for obtaining a first point cloud of a target object;
a contour point extracting unit, configured to extract a plurality of contour points of the target object from a first point cloud of the target object according to the first point cloud;
a first contour curve generating unit configured to generate a plurality of first contour curves of the target object according to a plurality of contour points of the target object;
a second contour curve generating unit, configured to perform point compensation on the plurality of first contour curves of the target object, and obtain a plurality of second contour curves after the point compensation;
a second point cloud obtaining unit, configured to obtain a second point cloud of the target object according to the point clouds on the plurality of second contour curves after point supplementation and the point cloud in the first contour curve in the first point cloud;
the identification unit is used for detecting a point cloud missing area in the second point cloud of the target object and identifying the boundary of the point cloud missing area;
and the repairing unit is used for repairing the point cloud in the point cloud missing area according to the existing point cloud with a preset position relation with the boundary of the point cloud missing area.
This application provides an electronic equipment simultaneously, includes:
a processor; and
a memory for storing a computer program for performing any of the above methods when the apparatus is powered on and the computer program is run by the processor.
The present application also provides a storage device storing a computer program, which is executed by a processor to perform any one of the above methods.
Compared with the prior art, the method has the following advantages:
the point cloud repairing method provided by the application comprises the following steps: acquiring a first point cloud of a target object; extracting a plurality of contour points of the target object from a first point cloud of the target object according to the first point cloud; generating a plurality of first contour curves of the target object according to the plurality of contour points of the target object; point supplementing is carried out on the plurality of first contour curves of the target object, and a plurality of second contour curves after point supplementing are obtained; obtaining second point clouds of the target object according to the point clouds on the plurality of second contour curves after point compensation and the point clouds in the first contour curve in the first point cloud; detecting a point cloud missing region in a second point cloud of the target object, and identifying a boundary of the point cloud missing region; and repairing the point cloud in the point cloud missing area according to the existing point cloud with a preset position relation with the boundary of the point cloud missing area.
The point cloud repairing method includes the steps that first point cloud of a target object, namely the point cloud containing a non-closed missing area, is obtained, then a plurality of contour points of the target object are extracted from the first point cloud, and a plurality of first contour curves of the target object are generated according to the contour points, wherein the first contour curves comprise a point cloud dense area and a point cloud sparse area; secondly, point supplementing is carried out on the plurality of first contour curves of the target object, namely point cloud sparse regions on the first contour curves, and a plurality of second contour curves after point supplementing are obtained; obtaining a second point cloud of the target object according to the point clouds on the plurality of second contour curves after point compensation and the point cloud in the first contour curve in the first point cloud, namely obtaining the point cloud of the closed type missing area; then, detecting a point cloud missing area in a second point cloud of the target object, and identifying the boundary of the point cloud missing area; and finally, repairing the point cloud in the point cloud missing area according to the existing point cloud with a preset position relation with the boundary of the point cloud missing area.
According to the point cloud repairing method, the contour curve of the target object is generated through the contour points of the target object, then the point is supplemented to the contour curve of the target object, the non-closed missing area of the point cloud is converted into the closed missing area, and finally the point cloud in the closed missing area is repaired. Therefore, the point cloud repairing method can realize the repairing of the non-closed large-area missing point cloud.
Drawings
Fig. 1 is a scene schematic diagram of a point cloud repairing method provided in the present application;
fig. 2 is a flowchart of a point cloud repairing method according to a first embodiment of the present application;
FIG. 3 is a first point cloud schematic of a target object provided herein;
FIG. 4 is a flow chart of contour point extraction for a target object provided herein;
FIG. 5 is a graph of the number of point clouds in a first point cloud slice of a target object as provided herein;
FIG. 6 is a schematic view of a third point cloud of a target object as provided herein;
FIG. 7 is a schematic view of a second point cloud of a target object provided herein;
FIG. 8 is a schematic illustration of the number of point clouds in a third point cloud slice of the target object provided herein;
FIG. 9 is a schematic diagram of the boundary of the point cloud missing region of the target object provided in the present application;
FIG. 10 is a schematic illustration of point cloud repair in a point cloud missing region of a target object provided herein;
FIG. 11 is a schematic view of a point cloud repair apparatus according to a second embodiment of the present application;
fig. 12 is a schematic diagram of an electronic device provided in a third embodiment of the present application.
Detailed Description
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present application. This application is capable of implementation in many different ways than those herein set forth and of similar import by those skilled in the art without departing from the spirit of this application and is therefore not limited to the specific implementations disclosed below.
First, in order to make those skilled in the art better understand the scheme of the present application, a detailed description is given below of a specific application scenario of an embodiment of the point cloud repairing method based on the point cloud repairing method provided by the present application. Fig. 1 is a schematic view of an application scenario of a point cloud repairing method according to a first embodiment of the present application.
In a specific implementation process, the point cloud restoration method is implemented in the application, the main application scene is an animal body in a farm, such as a cow, a sheep and the like, and the target object is the cow body in the farm in the embodiment of the application. Because the collection environment to the ox body is very complicated in the plant, when gathering the ox body data, in order to prescribe the home range of the ox body, most all need set up the collection passageway that the railing was built, when shooing the ox body, because sheltering from of railing, the ox body data that obtain also can correspond have some deletions, lead to finally the ox body point cloud data that obtain to have a large tracts of land, and be non-closed disappearance. The point cloud repairing method is used for repairing the large-area and non-closed point cloud missing part to complete the point cloud missing area on the body surface of the cattle body, so that complete point cloud data of the cattle body is obtained, and health condition data of the cattle body is obtained through the point cloud data.
Firstly, a first point cloud of a cow body needs to be obtained, specifically: the staff of plant utilizes shooting equipment degree of depth camera to gather the degree of depth video data of the ox body from the passageway side through setting up the one-way walking rail passageway of the ox body with the fixed back of ox body to upload the service end with this degree of depth video data. After the server obtains the depth video data of the cattle body, the depth video data is converted into three-dimensional point cloud data, namely first point cloud of a target object, wherein the target object is the cattle body, and the first point cloud is a point cloud containing a non-closed point cloud missing area.
After the first point cloud of the cattle body is obtained, projecting the first point cloud onto a plane to obtain side view point cloud of the cattle body, and slicing the side view point cloud of the cattle body along the direction of a longitudinal axis to obtain a plurality of first point cloud slices of the cattle body. Calculating the number of point clouds in each first point cloud slice, removing the point cloud slices with the minimum point cloud number, extracting the point clouds in the reserved point cloud slices, namely extracting the body part point clouds of the cattle body, namely the third point clouds of the cattle body, and extracting a plurality of contour points of the cattle body from the body part point clouds of the cattle body.
Then, according to the plurality of contour points of the extracted cow body, performing curve fitting on the plurality of contour points to generate a contour curve, namely a first contour curve, wherein the point clouds on the first contour curve have sparse regions and dense regions, comparing the minimum interval of the transverse distance between the point clouds in the body part point clouds with the transverse distance between the adjacent point clouds on the first contour curve, and then, according to the point clouds on the plurality of second contour curves after point complementing and the point clouds in the first contour curve in the first point cloud, obtaining a second point cloud of the cow body, namely, after the point complementing of the first pair of contour curves, converting the point cloud missing regions in the first point cloud from non-closed type to closed type, namely obtaining the second point cloud, wherein the second point cloud is the point cloud containing the closed point cloud missing regions.
And finally, detecting a point cloud missing area in the second point cloud of the cattle body, identifying the boundary of the point cloud missing area, carrying out grid segmentation on the point cloud based on coordinates according to the existing point cloud with a preset position relation with the boundary of the point cloud missing area, namely according to the three-dimensional coordinate data of the point cloud near the point cloud missing area, adding point rows according to the minimum interval, and further completing the repair of the point cloud in the point cloud missing area.
The point cloud repairing method comprises the steps of analyzing the whole process of the point cloud repairing method, generating a contour curve of a target object according to contour points in the point cloud of the target object after the point cloud of the target object is obtained, then, performing point repairing on the contour curve of the target object, converting a non-closed missing area of the point cloud into a closed missing area, and finally, repairing the point cloud in the closed missing area. Therefore, the point cloud repairing method can realize the repairing of the non-closed large-area missing point cloud.
The present application is described in detail below with reference to a number of embodiments and the accompanying drawings.
First embodiment
A first embodiment of the present application provides a point cloud repairing method, fig. 2 is a flowchart of the point cloud repairing method provided in the first embodiment of the present application, and the information service method is described in detail below with reference to fig. 2.
Step S201: a first point cloud of a target object is obtained.
The step is applied to a server, and is used for acquiring a first point cloud of a target object, wherein the acquiring of the first point cloud of the target object comprises the following steps: the method comprises the steps of obtaining video data of a target object, and converting the video data into point clouds serving as first point clouds of the target object.
In the embodiment of the application, also taking the target object as an example of a cow, a worker in a farm collects video data of the cow by using a Kinect (Kinect is a new vocabulary created by connecting two characters) depth camera to obtain depth video data of the cow, wherein the depth data is an image formed by emitting infrared rays by using an infrared emission camera and receiving the infrared rays by using an infrared receiving camera, a depth distance is calculated by using an emission-reception difference, and the infrared image has a far-near gradient as can be seen from the captured image. And after the terminal collects the depth video data of the target object, the depth video data is sent to the server side.
After the server side obtains the depth video data of the target object, the depth video data is converted into point clouds serving as first point clouds of the target object. Referring to fig. 3, fig. 3 is a schematic diagram of a first point cloud of a target object provided in the present application, wherein the first point cloud includes a non-closed point cloud missing region.
In the embodiment of the present application, a direction perpendicular to the ground from the back to the foot of the cow body is an X-axis direction of the three-dimensional point cloud coordinate system, a direction parallel to the ground from the head to the tail of the cow body is a Y-axis direction of the three-dimensional point cloud coordinate system, and a body width direction of the cow body is a Z-axis direction of the three-dimensional point cloud coordinate system.
In the above process of obtaining the first point cloud of the target object for the server, a process of extracting the contour point of the target object from the first point cloud is described next.
Step S202: extracting a plurality of contour points of the target object from the first point cloud according to the first point cloud of the target object.
This step is used to extract a plurality of contour points of the target object from the first point cloud. Fig. 4 is a flowchart of the contour point extraction of the target object provided in the present application, and the following describes in detail the process of extracting the contour point of the target object with reference to fig. 4.
The method for extracting a plurality of contour points of the target object from the first point cloud according to the first point cloud of the target object comprises the following steps:
step S401: and carrying out slicing processing on the first point cloud of the target object to obtain a plurality of first point cloud slices of the target object.
In the embodiment of the application, after the first point cloud of the target object is obtained, slicing needs to be performed on the first point cloud of the target object, so that a plurality of first point cloud slices of the target object are obtained. Wherein the slicing the first point cloud of the target object to obtain a plurality of first point cloud slices of the target object includes: projecting the first point cloud of the target object onto a plane to obtain a side view point cloud of the target object; and slicing the side view point cloud of the target object along the longitudinal axis direction to obtain a plurality of first point cloud slices of the target object.
It should be noted that before slicing the first point cloud of the target object, the first point cloud of the target object needs to be projected onto the xoy two-dimensional plane, so that the three-dimensional point cloud data can be converted into two-dimensional point cloud data, that is, a side view point cloud of the target object is obtained, and then the side view point cloud of the target object is sliced along the longitudinal axis direction (X axis), so that a plurality of first point cloud slices of the target object can be obtained.
Step S402: and calculating the number of point clouds in each first point cloud slice.
And slicing the side-looking point cloud of the target object along the longitudinal axis direction to obtain a plurality of first point cloud slices. The step is to calculate the number of point clouds in each first point cloud slice, please refer to fig. 5, where fig. 5 is a graph illustrating the number of point clouds in the first point cloud slice of the target object provided in the present application, and it can be seen from the graph that the minimum number of point clouds in the boundary area between the limbs and the trunk of the cow body occurs.
Step S403: and after the point cloud slices with the point cloud number of the minimum value are removed, extracting the point clouds in the reserved point cloud slices to be used as a third point cloud of the target object.
And after the number of the point clouds in each first point cloud slice is calculated, removing the point cloud slices with the minimum point cloud number, and extracting the point clouds in the reserved point cloud slices to obtain a third point cloud of the target object.
It should be noted here that, because the number of point clouds in a boundary region between a limb and a trunk of a cow body may have a minimum value, a point cloud slice with the minimum point cloud number is removed, that is, a leg of the cow body is removed, and the remaining point clouds are extracted as trunk point clouds, please refer to fig. 6, where fig. 6 is a third point cloud schematic diagram of a target object provided in the present application, and the finally obtained third point cloud is the trunk point cloud of the cow body.
Step S404: extracting a plurality of contour points of the target object from the third point cloud of the target object.
The method is used for extracting contour points of the cattle body from the trunk point cloud of the cattle body.
The extracting a plurality of contour points of the target object from the third point cloud of the target object comprises: slicing the third point cloud of the target object along a transverse axis to obtain a plurality of second point cloud slices of the target object; and detecting a maximum value point and a minimum value point of each second point cloud slice in the longitudinal axis direction as a plurality of contour points of the target object.
It should be noted here that, extracting contour points of the cow body from the point cloud of the trunk of the cow body, slicing the point cloud of the trunk of the cow body again along the horizontal axis (Y axis) to obtain a plurality of second point cloud slices, and then detecting a maximum point and a minimum point of each second point cloud slice in the direction of the vertical axis (X axis) as the contour points of the cow body.
Step S203: and generating a plurality of first contour curves of the target object according to the plurality of contour points of the target object.
The step is used for generating a first contour curve of the cattle body according to the extracted contour points of the cattle body.
Generating a plurality of first contour curves of the target object according to the plurality of contour points of the target object, including: and performing curve fitting on the plurality of contour points of the target object by adopting a preset algorithm to generate a plurality of first contour curves of the target object. The preset algorithm includes a least square method, and the least square method (also called a least square method) is a mathematical optimization technology. It finds the best functional match of the data by minimizing the sum of the squares of the errors. Unknown data can be easily obtained by the least square method, and the sum of squares of errors between these obtained data and actual data is minimized. Least squares methods are also commonly used for curve fitting. According to the plurality of contour points of the cattle body, curve fitting is performed by using a least square method, and a plurality of first contour curves of the cattle body are generated.
Step S204: and performing point compensation on the plurality of first contour curves of the target object to obtain a plurality of second contour curves after point compensation.
The method comprises the following steps of performing point supplementing on a point cloud sparse region on the first contour curve, and further obtaining a second contour curve after point supplementing.
Performing point compensation on the plurality of first contour curves of the target object to obtain a plurality of second contour curves after point compensation, including: calculating a minimum interval of transverse distances between point clouds in a third point cloud of the target object; comparing a lateral distance between adjacent point clouds on a first contour curve of the target object to a size of a minimum separation; and if the transverse distance between the adjacent point clouds on the first contour curve of the target object is greater than the minimum interval, performing point compensation on the plurality of first contour curves of the target object to obtain a plurality of second contour curves after point compensation.
Here, since the first contour curve is obtained by curve-fitting a plurality of contour points, the point cloud on the first contour curve has a sparse region and also has a dense region. After obtaining trunk point clouds of a cattle body, firstly calculating the minimum interval of the transverse distance between the point clouds in the trunk point clouds of the lower cattle body, then comparing the transverse distance between the adjacent point clouds on a first contour curve of the cattle body with the minimum interval, if the transverse distance between the adjacent point clouds on the first contour curve is larger than the minimum interval, proving that the area is a point cloud sparse area, and performing point supplementation on the first contour curve to obtain a second contour curve after point supplementation; on the contrary, if the region is proved to be a dense region, the first contour curve does not need to be subjected to point filling.
If the lateral distance between the adjacent point clouds on the first contour curve of the target object is greater than the minimum interval, performing point compensation on the plurality of first contour curves of the target object, including: and selecting points on the first contour curve of the target object as datum points, and performing point supplementation on the first contour curve of the target object by taking the minimum interval as a unit. The method specifically comprises the following steps: selecting any one existing point in the point cloud sparse region on the first contour curve as a reference point, determining the position of the point at the position with the minimum distance in the horizontal direction, wherein the intersection point of the position and the first contour curve in the vertical direction is the point supplementing position, and according to the method, supplementing points to the first contour curve to obtain a new point adding row, and simultaneously converting the point cloud sparse region on the first contour curve into a point cloud dense region, namely obtaining a second contour curve.
It should be noted that, after the position of the new point array is determined, the abscissa and ordinate values of the new point array may also be determined, and the Z-coordinate value of the new point array is an average value of Z-coordinates of all point clouds in the second point cloud slice where the point is located.
Step S205: and obtaining the second point cloud of the target object according to the point clouds on the plurality of second contour curves after point compensation and the point cloud in the first contour curve in the first point cloud.
The step is used for obtaining a second point cloud of the target object, namely the point cloud containing the closed point cloud missing area.
Referring to fig. 7, fig. 7 is a schematic diagram of a second point cloud of a target object provided in the present application, a plurality of first contour curves in a first point cloud are subjected to point compensation to obtain a second contour curve after point compensation, and the second point cloud of the target object can be obtained according to the point clouds on the plurality of second contour curves after point compensation and the point cloud in the first contour curve in the first point cloud.
It should be noted that, through the above steps, the point cloud missing area in the first point cloud of the bovine body is converted from the non-closed type to the closed type, that is, a second point cloud is obtained, where the second point cloud is the point cloud including the closed point cloud missing area, the contour point is shown in fig. 7, and through point supplementation on the plurality of first contour curves, the bovine body surface point cloud data including the non-closed point cloud missing area can be converted into the bovine body surface point cloud data including the closed point cloud missing area.
Step S206: and detecting a point cloud missing area in the second point cloud of the target object, and identifying the boundary of the point cloud missing area.
The method comprises the steps of detecting a point cloud missing area in a second point cloud of the target object, and then identifying the boundary of the point cloud missing area.
Firstly, a point cloud missing region in a second point cloud of a target object is detected. The detecting a point cloud missing region in a second point cloud of the target object, identifying a boundary of the point cloud missing region, includes: slicing the second point cloud of the target object along a longitudinal axis to obtain a plurality of third point cloud slices of the target object; calculating the number of point clouds in each third point cloud slice; and selecting the point cloud slice with the point cloud number of the minimum value as a point cloud missing area in the second point cloud of the target object.
Referring to fig. 8, fig. 8 is a schematic diagram illustrating the number of point clouds in a third point cloud slice of a target object provided in the present application, and since the slice includes a point cloud missing region, the point cloud number is the minimum, and therefore, the point cloud missing region in a second point cloud of the target object can be identified by selecting the point cloud slice whose point cloud number is the minimum.
And after a point cloud missing area in a second point cloud of the target object is detected, detecting boundary characteristic points of the point cloud missing area, and then identifying the boundary of the point cloud missing area according to the boundary characteristic points of the point cloud missing area. Referring to fig. 9, fig. 9 is a schematic diagram of a boundary of a point cloud missing region of a target object provided in the present application, a k-d _ tree algorithm is used to detect a boundary feature point of a slice including the point cloud missing region, where the k-d _ tree (short for k-dimensional tree) is a data structure for partitioning a k-dimensional data space. The method is mainly applied to searching of multidimensional space key data (such as range searching and nearest neighbor searching). The k-d _ tree is a special case of a binary spatial partition tree.
Step S207: and repairing the point cloud in the point cloud missing area according to the existing point cloud with a preset position relation with the boundary of the point cloud missing area.
The method is used for repairing the point cloud in the point cloud missing area. And after the point cloud missing area is obtained according to the steps, repairing the point cloud in the point cloud missing area according to the position characteristics of the point cloud near the boundary of the point cloud missing area.
The repairing of the point cloud in the point cloud missing area according to the existing point cloud with a preset position relation with the boundary of the point cloud missing area comprises the following steps: and repairing the point cloud in the point cloud missing area according to the existing point cloud which has a preset position relation with the boundary of the point cloud missing area and the minimum interval and a preset algorithm. The preset position relation refers to the distance between the preset position relation and the boundary of the point cloud missing area, wherein the preset position relation refers to the distance between the preset position relation and the boundary of the point cloud missing area, and the distance is the shortest.
The repairing of the point cloud in the point cloud missing area according to the existing point cloud having a preset position relation with the boundary of the point cloud missing area and the minimum interval and according to a preset algorithm comprises the following steps: dividing the point cloud missing region along the direction of a longitudinal axis and/or a transverse axis according to the minimum interval to obtain m and/or n divided point cloud missing regions, wherein m and n are integers more than or equal to 1; and selecting the existing point cloud with a preset position relation with the boundary of the point cloud missing area as a reference, and performing point cloud repair on each segmented point cloud missing area by taking the minimum interval as a unit.
Specifically, the number m of the point cloud missing region divided along the longitudinal axis direction is obtained by adopting the following formula:
Figure BDA0003723415060000131
wherein, X _ max and X _ min are respectively the maximum value and the minimum value of the boundary point of the point cloud missing region in the longitudinal axis direction, d is the minimum interval, and m is the number of parts of the point cloud missing region divided in the longitudinal axis direction.
Obtaining the number n of parts of the point cloud missing region segmented along the direction of the transverse axis by adopting the following formula:
Figure BDA0003723415060000132
y _ max and Y _ min are respectively the maximum value and the minimum value of the boundary points of the point cloud missing area in the transverse axis direction, d is the minimum interval, and n is the number of parts of the point cloud missing area divided in the transverse axis direction.
Selecting existing point clouds having a preset position relation with the boundary of the point cloud missing area as a reference, and performing point cloud repair on each segmented point cloud missing area by taking the minimum interval as a unit, wherein the method comprises the following steps: obtaining a longitudinal axis coordinate x and a transverse axis coordinate y of the repaired point cloud by adopting the following formula:
x＝X_min+i*d
y＝Y_min+j*d
i∈(1,m-1)
j∈(1,n-1)
wherein X is a longitudinal coordinate value of the repaired point cloud, Y is an abscissa value of the repaired point cloud, the z coordinate of the repaired point cloud is a median value of z coordinates of other point clouds in an ith block area divided by the point cloud missing area along the longitudinal axis direction, X _ min is a minimum value of the boundary point of the point cloud missing area along the longitudinal axis direction, Y _ min is a minimum value of the boundary point of the point cloud missing area along the transverse axis direction, d is a minimum interval, m is the number of parts of the point cloud missing area divided along the longitudinal axis direction, n is the number of parts of the point cloud missing area divided along the transverse axis direction, i is the ith block divided by the point cloud missing area along the longitudinal axis direction, the range of i is greater than or equal to and less than or equal to m-1, j is the jth block divided by the point cloud missing area along the transverse axis direction, and the range of j is greater than or equal to and less than or equal to n-1. Referring to fig. 10, fig. 10 is a schematic view of point cloud repairing in a point cloud missing area of a target object provided in the present application, where points at a newly added point row in the diagram are points after the point cloud missing area is repaired.
The point cloud repairing method is applied to a server side, and the method further comprises the following steps: obtaining a fourth point cloud of the target object according to the repaired point cloud; analyzing the fourth point cloud to obtain health state data of the target object; and sending the health state data of the target object to a terminal. And the fourth point cloud is the point cloud obtained by repairing the point cloud in the non-closed missing area, namely the complete point cloud of the cattle body. The health state data of the cattle body can be obtained by analyzing the complete point cloud data of the cattle body. The health state data comprises body condition scores, body size data, weight data, health grades and the like of the cattle.
The point cloud repairing method comprises the following steps: acquiring a first point cloud of a target object; extracting a plurality of contour points of the target object from a first point cloud of the target object according to the first point cloud; generating a plurality of first contour curves of the target object according to the plurality of contour points of the target object; point supplementing is carried out on the plurality of first contour curves of the target object, and a plurality of second contour curves after point supplementing are obtained; obtaining second point clouds of the target object according to the point clouds on the plurality of second contour curves after point compensation and the point clouds in the first contour curve in the first point cloud; detecting a point cloud missing region in a second point cloud of the target object, and identifying a boundary of the point cloud missing region; and repairing the point cloud in the point cloud missing area according to the existing point cloud with a preset position relation with the boundary of the point cloud missing area.
The point cloud repairing method includes the steps that first point cloud of a target object, namely the point cloud containing a non-closed missing area, is obtained, then a plurality of contour points of the target object are extracted from the first point cloud, and a plurality of first contour curves of the target object are generated according to the contour points, wherein the first contour curves comprise a point cloud dense area and a point cloud sparse area; secondly, point supplementing is carried out on the plurality of first contour curves of the target object, namely point cloud sparse regions on the first contour curves, and a plurality of second contour curves after point supplementing are obtained; obtaining a second point cloud of the target object according to the point clouds on the plurality of second contour curves after point compensation and the point cloud in the first contour curve in the first point cloud, namely obtaining the point cloud of the closed type missing area; then, detecting a point cloud missing area in a second point cloud of the target object, and identifying the boundary of the point cloud missing area; and finally, repairing the point cloud in the point cloud missing area according to the existing point cloud with a preset position relation with the boundary of the point cloud missing area.
According to the point cloud repairing method, the contour curve of the target object is generated through the contour points of the target object, then the point is supplemented to the contour curve of the target object, the non-closed missing area of the point cloud is converted into the closed missing area, and finally the point cloud in the closed missing area is repaired. Therefore, the point cloud repairing method can realize the repairing of the non-closed large-area missing point cloud.
Second embodiment
In the first embodiment, a point cloud repairing method is provided, and correspondingly, a second embodiment of the present application provides a point cloud repairing apparatus. Since the apparatus embodiment is substantially similar to the method embodiment, the description is relatively simple, and reference may be made to some description of the method embodiment for relevant points. The device embodiments described below are merely illustrative.
Fig. 11 is a schematic view of a point cloud repairing apparatus according to a second embodiment of the present application.
The device comprises:
a first point cloud obtaining unit 1101 configured to obtain a first point cloud of a target object;
a contour point extracting unit 1102, configured to extract a plurality of contour points of the target object from a first point cloud of the target object according to the first point cloud;
a first contour curve generating unit 1103 configured to generate a plurality of first contour curves of the target object according to the plurality of contour points of the target object;
a second contour curve generating unit 1104, configured to perform point compensation on the plurality of first contour curves of the target object, and obtain a plurality of second contour curves after the point compensation;
a second point cloud obtaining unit 1105, configured to obtain a second point cloud of the target object according to the point clouds on the plurality of second contour curves after point compensation and the point cloud in the first contour curve in the first point cloud;
an identifying unit 1106, configured to detect a point cloud missing region in the second point cloud of the target object, and identify a boundary of the point cloud missing region;
the repairing unit 1107 is configured to repair the point cloud in the point cloud missing area according to an existing point cloud having a preset position relationship with the boundary of the point cloud missing area.
Third embodiment
Corresponding to the above method embodiments provided by the present application, a third embodiment of the present application further provides an electronic device. Since the third embodiment is substantially similar to the above method embodiment provided in this application, it is described relatively simply, and reference may be made to some descriptions of the above method embodiment provided in this application for relevant points. The third embodiment described below is merely illustrative.
Fig. 12 is a schematic diagram of an electronic device provided in an embodiment of the present application.
The electronic device includes: a processor 1201;
and a memory 1202 for storing a computer program which, when the apparatus is powered on and the computer program is run by the processor, performs the method provided in the above-described embodiments of the present application.
It should be noted that, for the detailed description of the electronic device provided in the third embodiment of the present application, reference may be made to the related description of the foregoing method embodiment provided in the present application, and details are not repeated here.
Fourth embodiment
Corresponding to the above method embodiments provided by the present application, a fourth embodiment of the present application further provides a storage device. Since the fourth embodiment is substantially similar to the above method embodiment provided in this application, the description is relatively simple, and for the relevant points, reference may be made to part of the description of the above method embodiment provided in this application. The fourth embodiment described below is merely illustrative.
The storage device stores a computer program that is executed by a processor to perform the methods provided in the embodiments of the present application.
It should be noted that, for the detailed description of the storage device provided in the fourth embodiment of the present application, reference may be made to the related description of the foregoing method embodiment provided in the present application, and details are not repeated here.
Although the present application has been described with reference to the preferred embodiments, it is not intended to limit the present application, and those skilled in the art can make variations and modifications without departing from the spirit and scope of the present application, therefore, the scope of the present application should be determined by the claims that follow.
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). Memory is an example of a computer-readable medium.
1. 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, computer readable media does not include non-transitory computer readable media (transient media), such as modulated data signals and carrier waves.
2. 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.

Claims (10)

1. A point cloud repairing method is characterized by comprising the following steps:
acquiring a first point cloud of a target object;
extracting a plurality of contour points of the target object from a first point cloud of the target object according to the first point cloud;
generating a plurality of first contour curves of the target object according to the plurality of contour points of the target object;
point supplementing is carried out on the plurality of first contour curves of the target object, and a plurality of second contour curves after point supplementing are obtained;
obtaining second point clouds of the target object according to the point clouds on the plurality of second contour curves after point compensation and the point clouds in the first contour curve in the first point cloud;
detecting a point cloud missing region in a second point cloud of the target object, and identifying a boundary of the point cloud missing region;
and repairing the point cloud in the point cloud missing area according to the existing point cloud with a preset position relation with the boundary of the point cloud missing area.
2. The point cloud repair method of claim 1, wherein the extracting, from the first point cloud of the target object, a plurality of contour points of the target object according to the first point cloud comprises:
slicing the first point cloud of the target object to obtain a plurality of first point cloud slices of the target object;
calculating the number of point clouds in each first point cloud slice;
after removing the point cloud slice with the point cloud number of the minimum value, extracting the point cloud in the reserved point cloud slice to be used as a third point cloud of the target object;
extracting a plurality of contour points of the target object from the third point cloud of the target object.
3. The point cloud repair method of claim 1, wherein generating a plurality of first contour curves for the target object from a plurality of contour points for the target object comprises:
and performing curve fitting on the plurality of contour points of the target object by adopting a preset algorithm to generate a plurality of first contour curves of the target object.
4. The point cloud repairing method according to claim 2, wherein the point supplementing is performed on the plurality of first contour curves of the target object, and a plurality of second contour curves after point supplementing are obtained, and the point cloud repairing method includes:
calculating a minimum interval of a transverse distance between point clouds in a third point cloud of the target object;
comparing a lateral distance between adjacent point clouds on a first contour curve of the target object to a size of a minimum separation;
and if the transverse distance between the adjacent point clouds on the first contour curve of the target object is greater than the minimum interval, performing point compensation on the plurality of first contour curves of the target object to obtain a plurality of second contour curves after point compensation.
5. The point cloud repair method of claim 1, wherein the detecting a point cloud missing region in a second point cloud of the target object, identifying a boundary of the point cloud missing region, comprises:
slicing the second point cloud of the target object along a longitudinal axis to obtain a plurality of third point cloud slices of the target object;
calculating the number of point clouds in each third point cloud slice;
selecting a point cloud slice with the point cloud number of the minimum value as a point cloud missing area in a second point cloud of the target object;
and detecting the boundary characteristic points of the point cloud missing area, and identifying the boundary of the point cloud missing area according to the boundary characteristic points of the point cloud missing area.
6. The point cloud repairing method according to claim 4, wherein repairing the point cloud in the point cloud missing area according to an existing point cloud having a preset positional relationship with a boundary of the point cloud missing area comprises:
and repairing the point cloud in the point cloud missing area according to the existing point cloud which has a preset position relation with the boundary of the point cloud missing area and the minimum interval and a preset algorithm.
7. The point cloud repair method of claim 1, applied to a server, the method further comprising:
obtaining a fourth point cloud of the target object according to the repaired point cloud;
analyzing the fourth point cloud to obtain health state data of the target object;
and sending the health state data of the target object to a terminal.
8. A point cloud repair device, comprising:
the first point cloud obtaining unit is used for obtaining a first point cloud of a target object;
a contour point extracting unit, configured to extract a plurality of contour points of the target object from a first point cloud of the target object according to the first point cloud;
a first contour curve generating unit configured to generate a plurality of first contour curves of the target object according to a plurality of contour points of the target object;
a second contour curve generating unit, configured to perform point compensation on the plurality of first contour curves of the target object, and obtain a plurality of second contour curves after the point compensation;
a second point cloud obtaining unit, configured to obtain a second point cloud of the target object according to the point clouds on the plurality of second contour curves after point supplementation and the point cloud in the first contour curve in the first point cloud;
the identification unit is used for detecting a point cloud missing area in the second point cloud of the target object and identifying the boundary of the point cloud missing area;
and the repairing unit is used for repairing the point cloud in the point cloud missing area according to the existing point cloud with a preset position relation with the boundary of the point cloud missing area.
9. An electronic device, comprising:
a processor; and
a memory for storing a computer program which, when the apparatus is powered on and the computer program is run by the processor, performs the method of any one of claims 1 to 7.
10. A storage device, characterized in that the storage device stores a computer program which is executed by a processor for performing the method according to any of claims 1-7.
CN202210769414.5A 2022-06-30 2022-06-30 Point cloud repairing method and device, electronic equipment and storage equipment Pending CN115115548A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210769414.5A CN115115548A (en) 2022-06-30 2022-06-30 Point cloud repairing method and device, electronic equipment and storage equipment

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210769414.5A CN115115548A (en) 2022-06-30 2022-06-30 Point cloud repairing method and device, electronic equipment and storage equipment

Publications (1)

Publication Number Publication Date
CN115115548A true CN115115548A (en) 2022-09-27

Family

ID=83329702

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210769414.5A Pending CN115115548A (en) 2022-06-30 2022-06-30 Point cloud repairing method and device, electronic equipment and storage equipment

Country Status (1)

Country Link
CN (1) CN115115548A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115601272A (en) * 2022-12-16 2023-01-13 海纳云物联科技有限公司（Cn） Point cloud data processing method, device and equipment

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115601272A (en) * 2022-12-16 2023-01-13 海纳云物联科技有限公司（Cn） Point cloud data processing method, device and equipment
CN115601272B (en) * 2022-12-16 2023-04-11 海纳云物联科技有限公司 Point cloud data processing method, device and equipment

Similar Documents

Publication Publication Date Title
Koreň et al. Accuracy of tree diameter estimation from terrestrial laser scanning by circle-fitting methods
CN110221311B (en) Method for automatically extracting tree height of high-canopy-closure forest stand based on TLS and UAV
Kattenborn et al. Automatic single tree detection in plantations using UAV-based photogrammetric point clouds
US11481972B2 (en) Method of performing dendrometry and forest mapping
Ye et al. Analysing the potential of UAV point cloud as input in quantitative structure modelling for assessment of woody biomass of single trees
Su et al. Extracting wood point cloud of individual trees based on geometric features
Roy et al. Surveying apple orchards with a monocular vision system
CN115115548A (en) Point cloud repairing method and device, electronic equipment and storage equipment
CN115854895A (en) Non-contact stumpage breast diameter measurement method based on target stumpage form
Novotny et al. Comparison of spruce and beech tree attributes from field data, airborne and terrestrial laser scanning using manual and automatic methods
Wu et al. Plant 3D reconstruction based on LiDAR and multi-view sequence images
CN114241217A (en) Trunk point cloud efficient extraction method based on cylindrical features
CN112906719A (en) Standing tree factor measuring method based on consumption-level depth camera
Xiong et al. Automatic three-dimensional reconstruction based on four-view stereo vision using checkerboard pattern
Han et al. A non-destructive approach to estimate buttress volume using 3D point cloud data
CN115937466A (en) Three-dimensional model generation method, system and storage medium integrating GIS
KR101255022B1 (en) Detecting method of building crack using point group
CN113724381B (en) Dynamic three-dimensional scene rapid reconstruction method based on high-resolution remote sensing image
Chai et al. A novel solution for extracting individual tree crown parameters in high-density plantation considering inter-tree growth competition using terrestrial close-range scanning and photogrammetry technology
Marinelli et al. Fusion of high and very high density lidar data for 3d forest change detection
Wang et al. Kinect-based 3D reconstruction of leaf lettuce
CN117036944B (en) Tree carbon sink amount calculating method and system based on point cloud data and image recognition
Harikumar et al. An effective approach to 3D stem modeling and branch-knot localization in multiscan TLS data
CN115620171B (en) Forest storage amount evaluation method, device, equipment and storage medium
Li et al. Automatic reconstruction and modeling of dormant jujube trees using three-view image constraints for intelligent pruning applications

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