CN117953232A - Three-dimensional point cloud simplifying method based on boundary point dimension reduction extraction - Google Patents
Three-dimensional point cloud simplifying method based on boundary point dimension reduction extraction Download PDFInfo
- Publication number
- CN117953232A CN117953232A CN202410006289.1A CN202410006289A CN117953232A CN 117953232 A CN117953232 A CN 117953232A CN 202410006289 A CN202410006289 A CN 202410006289A CN 117953232 A CN117953232 A CN 117953232A
- Authority
- CN
- China
- Prior art keywords
- model
- point cloud
- dimensional
- original
- slice
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three-dimensional [3D] modelling for computer graphics
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Graphics (AREA)
- Geometry (AREA)
- Software Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- Processing Or Creating Images (AREA)
Abstract
A three-dimensional point cloud simplifying method based on boundary point dimension reduction extraction mainly comprises the following steps: projecting the original three-dimensional point cloud model and the rotated updated model on an xy coordinate plane, an xz coordinate plane and a yz coordinate plane respectively, and extracting projection boundary points respectively to obtain an original projection boundary point set and a rotated projection boundary point set; calculating the size of a bounding box of the three-dimensional point cloud model, carrying out equidistant slicing on the model along a specified axial direction according to a specified thickness and a specified interval, and extracting slice boundary points to obtain a slice point set; and fusing and de-duplicating the original projection boundary point set, the rotation projection boundary point set and the slicing point set to obtain the simplified three-dimensional point cloud model. The method can effectively reserve the boundary and characteristic information of the original three-dimensional point cloud model while simplifying the model scale, has no special requirement on the structure of the model to be processed, has strong universality, and can greatly improve the precision and efficiency of the point cloud model processing.
Description
Technical Field
The invention relates to the field of computer vision and image processing, in particular to a three-dimensional point cloud simplifying method based on boundary point dimension reduction extraction.
Background
A three-dimensional point cloud is a data structure for representing an object or scene in a three-dimensional space, and is a set of a plurality of three-dimensional coordinate points obtained by performing discrete scanning on the surface of an actual object or scene. The three-dimensional point cloud acquisition device can be generally acquired by adopting devices such as a laser radar, a stereoscopic vision camera and a structured light scanner, and has higher precision, faster efficiency, lower cost and wider application scene along with the continuous development of optical sensor technology at present. Currently, three-dimensional point cloud models play an increasingly important role in the fields of environmental awareness, path planning, industrial detection, virtual reality, medical imaging and the like, and as the point cloud is obtained through discrete sampling, the geometric characteristics of a complex surface can be captured very accurately. To obtain a more accurate, complete point cloud data model, which typically contains tens of thousands of points, even up to millions of points, it is often necessary to increase the sampling frequency of the device and perform repeated scans, such large-scale point clouds present a significant challenge to model computation, storage, and processing.
The point cloud simplification is also called point cloud sampling, and is an effective method for the light weight processing of the three-dimensional point cloud model. The key point of the point cloud simplifying operation is to effectively identify and accurately extract the characteristics and boundaries of the original model, and on one hand, enough points are reserved at key parts such as the characteristics, the boundaries and the like in the simplifying process so as to ensure the integrity of the model; on the other hand, the quantity of the point clouds on the whole scale is reduced as much as possible so as to ensure the aim of light weight. At present, the most applied simplified methods comprise a random sampling method and a uniform sampling method, and the two methods have simple structures and high calculation efficiency, but lack of a model key point identification and selection process leads to a larger gap between the simplified model size and an original model and loss of part of key information, thereby reducing the model processing precision.
In addition, further searching finds that CN112270746a discloses an aluminum alloy 3D printing point cloud reduction algorithm based on a domain covariance feature parameter threshold, feature parameters are calculated through a covariance matrix, and model contours are reserved in a feature region with large curvature, but a standard model which is not suitable for feature consistency and lacks reservation operation on boundary points, such as a sphere model, is not available. CN115294272a discloses a self-adaptive point cloud simplifying method based on point cloud feature partition, which defines a contour line as a strong feature without simplifying, but does not define a contour line selecting method, because the point cloud model is a result obtained by scanning the contour of the target object, the selection of the contour line needs to define an angle and a coordinate plane. CN115546372a discloses a three-dimensional point cloud simplifying method based on voxel filtering, and the 3D-SIFT algorithm needs to extract strong feature points of the original point cloud image, but the method has larger recognition error on the strong feature points when processing the symmetrical structure model.
Disclosure of Invention
Aiming at the defects or improvement demands of the prior art, the three-dimensional point cloud simplifying method based on the dimension reduction extraction of boundary points is provided, and an original three-dimensional point cloud model and a rotated updated model are respectively projected on an xy coordinate plane, an xz coordinate plane and a yz coordinate plane, projection boundary points are respectively extracted, so that an original projection boundary point set and a rotation projection boundary point set are obtained; then, calculating the size of a bounding box of the three-dimensional point cloud model, carrying out equidistant slicing on the model along a specified axial direction according to a specified thickness and a specified interval, and obtaining a slice point set by extracting slice boundary points; and finally, fusing and de-duplicating the original projection boundary point set, the rotation projection boundary point set and the slice point set to obtain the simplified three-dimensional point cloud model. Practical tests show that the scheme can effectively solve the key problems of low simplifying precision, poor integrity, large post-processing error and the like in the prior art.
The three-dimensional point cloud simplifying method based on boundary point dimension reduction extraction mainly comprises the following steps:
S1, respectively projecting an original three-dimensional point cloud model and a rotated updated model on an xy coordinate plane, an xz coordinate plane and a yz coordinate plane, respectively extracting projection boundary points, and obtaining an original projection boundary point set and a rotated projection boundary point set;
S2, calculating the size of a bounding box of the three-dimensional point cloud model, carrying out equidistant slicing on the model along a specified axial direction according to a specified thickness and a specified interval, and obtaining a slice point set by extracting slice boundary points and fusing multiple layers of slices;
And S3, fusing and de-duplicating the original projection boundary point set, the rotation projection boundary point set and the slice point set to obtain the simplified three-dimensional point cloud model.
Further, the step S1 includes:
S11, respectively calculating the projection of an original three-dimensional point cloud model P { P 1,p2,p3,...,pn } on xy, xz and yz planes, and extracting projection boundary points, wherein n is the number of three-dimensional points contained in the point cloud P;
S12, fusing the extracted three boundary points, and removing repeated points to obtain an original projection boundary point set;
S13, the original three-dimensional point cloud model performs rotation operation around a designated coordinate axis according to a set angle;
And S14, sequentially executing the new model after the rotation operation in the step S11 and the step S12 to obtain a rotation projection boundary point set.
Further, the step S2 includes:
S21, calculating the size of an original three-dimensional point cloud model bounding box, and setting specific numerical values of variable slice thickness a and spacing d;
s22, executing equidistant slicing on the original three-dimensional point cloud model along the designated axial direction;
s23, extracting slice boundary points, and fusing the obtained multi-layer slice points to obtain a slice point set.
Further, in the step S11, the three-dimensional point cloud model is represented by a three-dimensional matrix, and only three-dimensional coordinate data information of each point is included in the model;
And extracting boundary points of the projection model in the two-dimensional space by adopting a maximum point distance method.
Further, in the step S12, a unique function is used to search and screen the fusion point set according to the arrangement sequence number, and duplicate points are deleted.
Further, the method of step S13 includes:
Given specific numerical values of rotation variables rx, ry and rz in an radian system, calculating rotation results of the three-dimensional model around x, y and z axes according to a formula (1), a formula (2) and a formula (3) respectively:
and then performing a rotation operation according to the formula (4):
PR＝(Rx*Ry*Rz)*P (4)
Wherein Rx, ry and Rz respectively represent rotation matrixes of the original model around x, y and z axes, and P R represents a new point cloud model obtained by executing rotation operation on the original model.
Further, in step S14, a rotation operation is performed on the original three-dimensional point cloud model for a specified number of times according to the actual requirement, the projections on three coordinate planes are sequentially calculated after the model is updated, and a fusion deduplication operation is performed to obtain a rotation projection boundary point set.
Further, in the step S21, the maximum value and the minimum value of each coordinate axis are adopted to determine the length L, the width W and the height H of the bounding box of the point cloud model, and the slice layer number Nc is calculated according to the formula (5);
Wherein max [ L, W, H ] represents the maximum value of the length, width and height of the bounding box.
Further, the step S23 includes:
Along the axial direction of the slice, constructing a space envelope by the thickness of the slice, selecting three-dimensional points in the space of the envelope according to the axial coordinate values of each point, and constructing a subset q i of slice points, i=1, 2,3,..,
And performing slice point subset fusion to obtain a slice point set Q.
Further, in the step S3, the point set fusion order is unlimited, and the deduplication operation is performed by retrieving the serial number index value of each point after fusion.
Compared with the prior art, the beneficial effects of the present disclosure are: (1) The two-dimensional boundary point set is constructed by adopting the original projection and the multi-angle rotation projection, so that the outline of the three-dimensional point cloud model can be effectively extracted, and the three-dimensional convex structure point cloud model is particularly high in simplifying efficiency and precision; (2) The provided directional equidistant slicing method has no slicing dead angle in the point cloud simplifying process, can realize the accurate extraction of the concave structural characteristics of the three-dimensional model, and is particularly suitable for the three-dimensional point cloud model processing of a complex structure; (3) simple calculation and less input variables; (4) The modularized operation is convenient, and one or more steps can be selected for operation treatment according to actual requirements; (5) The three-dimensional point cloud models with different structures can obtain more accurate simplifying results; (6) The method is strong in universality and suitable for the fields of three-dimensional point cloud classification, machine vision, pattern recognition and the like.
Drawings
The foregoing and other objects, features and advantages of the disclosure will be apparent from the following more particular descriptions of exemplary embodiments of the disclosure as illustrated in the accompanying drawings wherein like reference numbers generally represent like parts throughout the exemplary embodiments of the disclosure.
FIG. 1 is a basic flow chart of a three-dimensional point cloud reduction method based on boundary point dimension reduction extraction according to the present disclosure;
FIG. 2 is a schematic diagram of an exemplary original three-dimensional point cloud model;
FIG. 3 is a two-dimensional view of an original model projected in the xyz plane;
FIG. 4 is a two-dimensional view of a rotation model projected in the xyz plane;
FIG. 5 is an operational view of equidistant slicing of a model along a specified axis;
fig. 6 is a simplified result diagram of the obtained three-dimensional point cloud model.
Detailed Description
Preferred embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While the preferred embodiments of the present disclosure are illustrated in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.
The three-dimensional point cloud simplifying method based on boundary point dimension reduction extraction is used for carrying out multi-view rotation projection and equidistant slicing operation on a three-dimensional point cloud model, and carrying out boundary coordinate dimension reduction extraction in a two-dimensional space, so that efficient simplification of a large-scale three-dimensional point cloud model is realized.
A basic flow diagram in accordance with an exemplary embodiment of the present disclosure is shown in fig. 1. A detailed description will be given by taking the Bunny standard model in this field as an example with reference to fig. 1.
The selected model is shown in fig. 2, the three-dimensional point size of the selected model is 7935, the double-precision matrix with the representation mode of 7935×3 is shown, and the size of the model is 100 times amplified for more visual observation effect because the size of the original model is undersized.
The simplifying process of the model mainly comprises the following steps:
(1) The original point cloud model is represented by a three-dimensional matrix, the model only comprises coordinate data information of each point, fig. 3 (a), 3 (b) and 3 (c) respectively show two-dimensional projections of the original model on an xy plane, an xz plane and a yz plane, and marking points in a "+" shape in the figure show boundary points of a projection model in a two-dimensional space extracted by a maximum point distance method;
(2) Fusing boundary points of the original model in an xy plane, an xz plane and a yz plane, adopting a unique function to search according to sequence numbers, and deleting repeated points;
(3) Setting the values of variables rx, ry and rz of the model around x, y and z axes, in this example, setting the original model to rotate 60 degrees around three coordinate axes in the anticlockwise direction respectively, and converting the values into the values of the radian post-rotation variables rx= 1.0472, ry= 1.0472 and rz= 1.0472, wherein in this example, the original model is only rotated once, and in order to simplify the description, the original three-dimensional point cloud model can be rotated for a specified number of times as required in practical application;
(4) Respectively calculating projections of the new rotated model on an xy plane, an xz plane and a yz plane, respectively as shown in fig. 4 (a), fig. 4 (b) and fig. 4 (c), extracting boundary points of the projection model in a two-dimensional space by adopting a maximum point distance method, and fusing and de-duplicating multi-plane boundary points;
(5) Determining the length L, the width W and the height H of a point cloud model bounding box by adopting each axial extreme value, defining L, W, H in the example to respectively correspond to the directions of x, y and z, and calculating L=15.5515, W=15.3709 and H= 12.0448, so that slicing operation is selected along the direction of the x axis, wherein a dotted line box in fig. 5 is the model bounding box, and a right lower dotted line arrow in fig. 5 indicates the slicing selection axial direction;
(6) Specific values of the slice variable thickness a and the slice interval d are set, in this example, a=0.03, d=0.5, the slice layer number nc=30 is calculated according to the formula (fifth), and a gray section plane in fig. 5 is an exemplary 5 planes showing the slicing effect along the x-axis direction;
(7) Calculating a space envelope body constructed by the thickness of the slice, selecting a point construction slice point subset Q 1-q30 in the envelope body according to the axial coordinate value of the three-dimensional point cloud model slice, and obtaining a slice subset Q after the slice point subset fusion is executed;
Finally, the original projection boundary point set, the rotation projection boundary point set and the slice point set are fused, a simplified model obtained after the duplication removal is shown in fig. 6, and the scale of the point cloud after the simplification is reduced to 2628 from the original 7935.
In summary, compared with the existing point cloud simplifying algorithm, the method provided by the invention has the advantages of simple structure, good universality and high calculation precision, and can effectively reserve the boundary and the characteristics of the original model after simplifying, so that the method is suitable for dividing and classifying the three-dimensional point cloud model. In addition, the model simplifying technology can be used in the related fields of pattern recognition, point cloud matching and the like.
The foregoing technical solutions are merely exemplary embodiments of the present invention, and various modifications and variations can be easily made by those skilled in the art based on the application methods and principles disclosed in the present invention, not limited to the methods described in the foregoing specific embodiments of the present invention, so that the foregoing description is only preferred and not in a limiting sense.
Claims (10)
1. A three-dimensional point cloud simplifying method based on boundary point dimension reduction extraction comprises the following steps:
S1, respectively projecting an original three-dimensional point cloud model and a rotated updated model on an xy coordinate plane, an xz coordinate plane and a yz coordinate plane, respectively extracting projection boundary points, and obtaining an original projection boundary point set and a rotated projection boundary point set;
S2, calculating the size of a bounding box of the three-dimensional point cloud model, carrying out equidistant slicing on the model along a specified axial direction according to a specified thickness and a specified interval, and obtaining a slice point set by extracting slice boundary points and fusing multiple layers of slices;
And S3, fusing and de-duplicating the original projection boundary point set, the rotation projection boundary point set and the slice point set to obtain the simplified three-dimensional point cloud model.
2. The method according to claim 1, wherein the step S1 comprises:
S11, respectively calculating the projection of an original three-dimensional point cloud model P { P 1,p2,p3,...,pn } on xy, xz and yz planes, and extracting projection boundary points, wherein n is the number of three-dimensional points contained in the point cloud P;
S12, fusing the extracted three boundary points, and removing repeated points to obtain an original projection boundary point set;
S13, the original three-dimensional point cloud model performs rotation operation around a designated coordinate axis according to a set angle;
And S14, sequentially executing the new model after the rotation operation in the step S11 and the step S12 to obtain a rotation projection boundary point set.
3. The method according to claim 1 or 2, wherein the step S2 comprises:
S21, calculating the size of an original three-dimensional point cloud model bounding box, and setting specific numerical values of variable slice thickness a and spacing d;
s22, executing equidistant slicing on the original three-dimensional point cloud model along the designated axial direction;
s23, extracting slice boundary points, and fusing the obtained multi-layer slice points to obtain a slice point set.
4. The method according to claim 2, wherein in the step S11, the three-dimensional point cloud model is represented by a three-dimensional matrix, and only three-dimensional coordinate data information of each point is included in the model;
And extracting boundary points of the projection model in the two-dimensional space by adopting a maximum point distance method.
5. The method according to claim 2, wherein in step S12, the unique function is used to search and screen the fusion point set by the permutation number, and delete the duplicate points.
6. The method according to any one of claims 2, 4, 5, wherein the method of step S13 comprises:
Given specific numerical values of rotation variables rx, ry and rz in an radian system, calculating rotation results of the three-dimensional model around x, y and z axes according to a formula (1), a formula (2) and a formula (3) respectively:
and then performing a rotation operation according to the formula (4):
PR＝(Rx*Ry*Rz)*P (4)
Wherein Rx, ry and Rz respectively represent rotation matrixes of the original model around x, y and z axes, and P R represents a new point cloud model obtained by executing rotation operation on the original model.
7. The method according to claim 2, wherein in step S14, a rotation operation is performed on the original three-dimensional point cloud model for a specified number of times according to actual requirements, projections on three coordinate planes are sequentially calculated after model updating, and a fusion deduplication operation is performed to obtain a rotation projection boundary point set.
8. The method according to claim 3, wherein in the step S21, the maximum value and the minimum value of each coordinate axis are used to determine the length L, the width W, and the height H of the bounding box of the point cloud model, and the slice layer number Nc is calculated according to the formula (5);
Wherein max [ L, W, H ] represents the maximum value of the length, width and height of the bounding box.
9. The method according to claim 8, wherein the step S23 includes:
Along the axial direction of the slice, constructing a space envelope by the thickness of the slice, selecting three-dimensional points in the space of the envelope according to the axial coordinate values of each point, and constructing a subset q i of slice points, i=1, 2,3,..,
And performing slice point subset fusion to obtain a slice point set Q.
10. The method according to claim 1, wherein in the step S3, the point set fusion order is unlimited, and the deduplication operation is performed by retrieving the index value of the sequence number of each point after the fusion.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410006289.1A CN117953232A (en) | 2024-01-03 | 2024-01-03 | Three-dimensional point cloud simplifying method based on boundary point dimension reduction extraction |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410006289.1A CN117953232A (en) | 2024-01-03 | 2024-01-03 | Three-dimensional point cloud simplifying method based on boundary point dimension reduction extraction |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117953232A true CN117953232A (en) | 2024-04-30 |
Family
ID=90804171
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410006289.1A Pending CN117953232A (en) | 2024-01-03 | 2024-01-03 | Three-dimensional point cloud simplifying method based on boundary point dimension reduction extraction |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117953232A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118616984A (en) * | 2024-08-05 | 2024-09-10 | 上海船舶工艺研究所(中国船舶集团有限公司第十一研究所) | A method, system, device and storage medium for identifying a small ship weld |
| CN119884401A (en) * | 2025-03-28 | 2025-04-25 | 山东华云三维科技有限公司 | Method, equipment and medium for removing duplication of three-dimensional point set |
| CN120708212A (en) * | 2025-08-27 | 2025-09-26 | 浙江托普云农科技股份有限公司 | Method, system and device for extracting taproot point cloud of taproot system based on three-dimensional point cloud |
-
2024
- 2024-01-03 CN CN202410006289.1A patent/CN117953232A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118616984A (en) * | 2024-08-05 | 2024-09-10 | 上海船舶工艺研究所(中国船舶集团有限公司第十一研究所) | A method, system, device and storage medium for identifying a small ship weld |
| CN119884401A (en) * | 2025-03-28 | 2025-04-25 | 山东华云三维科技有限公司 | Method, equipment and medium for removing duplication of three-dimensional point set |
| CN120708212A (en) * | 2025-08-27 | 2025-09-26 | 浙江托普云农科技股份有限公司 | Method, system and device for extracting taproot point cloud of taproot system based on three-dimensional point cloud |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN119904592B (en) | News scene three-dimensional reconstruction and visualization method based on multi-source remote sensing data | |
| CN112651944B (en) | 3C component high-precision six-dimensional pose estimation method and system based on CAD model | |
| WO2024077812A1 (en) | Single building three-dimensional reconstruction method based on point cloud semantic segmentation and structure fitting | |
| CN117953232A (en) | Three-dimensional point cloud simplifying method based on boundary point dimension reduction extraction | |
| CN108090960A (en) | A kind of Object reconstruction method based on geometrical constraint | |
| EP1694821A2 (en) | Probable reconstruction of surfaces in occluded regions by computed symmetry | |
| CN111798453B (en) | Point cloud registration method and system for unmanned driving assisted positioning | |
| CN107680168B (en) | Grid simplifying method based on plane fitting in three-dimensional reconstruction | |
| CN113111741A (en) | Assembly state identification method based on three-dimensional feature points | |
| CN119273868A (en) | A real-time 3D fusion reconstruction method for ship comprehensive test scenes | |
| KR102129060B1 (en) | Content-based 3d model retrieval method using a single depth image, 3d model retrieval server for performing the methods and computer readable recording medium thereof | |
| CN112465984A (en) | Monocular camera sequence image three-dimensional reconstruction method based on double-layer filtering | |
| CN120014178A (en) | A 3D reconstruction method for unbounded scenes based on drone aerial photography | |
| CN117853656A (en) | Method and system for constructing three-dimensional model by fusing laser point cloud and single-lens image | |
| CN115423854A (en) | Multi-view point cloud registration and point cloud fusion method based on multi-scale feature extraction | |
| CN117710603B (en) | Unmanned aerial vehicle image three-dimensional building modeling method under constraint of linear geometry | |
| CN119206106A (en) | A three-dimensional mapping method and system | |
| CN112200879B (en) | A Map Lightweight Compression Transmission Method | |
| Jisen | A study on target recognition algorithm based on 3D point cloud and feature fusion | |
| CN117726772A (en) | A partitioned quadrilateral mesh reconstruction method for 3D point cloud models | |
| Yin et al. | [Retracted] Virtual Reconstruction Method of Regional 3D Image Based on Visual Transmission Effect | |
| CN120339543A (en) | Method and system for rapid reconstruction of digital twin model of industrial equipment based on laser point cloud | |
| CN119579693A (en) | Two-stage object six-degree-of-freedom pose estimation method based on monocular camera | |
| CN119963766A (en) | Method, device, equipment, medium and program product for three-dimensional reconstruction | |
| CN116704123A (en) | Three-dimensional reconstruction method combined with image main body extraction technology |
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 |