Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Rapid large-scale point-cloud data reading method based on memory pre-distribution and multi-point writing technology

A technology of memory pre-allocation and point cloud data, which is applied in the direction of memory address/allocation/relocation, concurrent instruction execution, machine execution device, etc. Improve user experience, increase reading speed, and improve the effect of reading speed

Inactive Publication Date: 2015-01-07
SOUTHWEAT UNIV OF SCI & TECH
View PDF2 Cites 17 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0008] The purpose of the present invention is to provide a large-scale point cloud data fast reading method based on memory pre-allocation and multi-point parallel writing technology, to solve the existing high-scale point cloud data file reading time delay, reading speed (loading) slow) problem

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Rapid large-scale point-cloud data reading method based on memory pre-distribution and multi-point writing technology
  • Rapid large-scale point-cloud data reading method based on memory pre-distribution and multi-point writing technology
  • Rapid large-scale point-cloud data reading method based on memory pre-distribution and multi-point writing technology

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0020] Specific implementation mode one: as figure 2 with 3 As shown, a method for quickly reading large-scale point cloud data based on memory pre-allocation and multi-point parallel writing technology described in this embodiment is characterized in that the method is:

[0021] Step A, memory pre-allocation process: first determine the number of points in the point cloud data file, so as to determine the memory size that all points in the point cloud data file need to occupy, and then pre-allocate memory of the corresponding size for the point cloud data;

[0022] Step B, multi-point writing process: through the memory-mapped file mechanism, map the point cloud data file to the mapped memory and create a thread pool with a specified number of threads. Each thread is responsible for parsing part of the point data information in the mapped memory, and The parsing results are written to the previously pre-allocated memory to achieve multi-point and write.

[0023] Step A is ...

specific Embodiment approach 2

[0024] Specific embodiment two: as shown in Figure 2, this embodiment is the specific implementation process of the large-scale point cloud data fast reading method based on memory pre-allocation and multi-point parallel writing technology described in specific embodiment one:

[0025] Step 1. Select the point cloud data file,

[0026] Step 2, determine the number of points in the point cloud data file, and determine the number of points in the point cloud data file by means of data block,

[0027] Step 3. Allocate memory of corresponding size: determine the memory size that all points in the point cloud data file need to occupy according to the number of points in the point cloud data file, and then pre-allocate memory of the corresponding size for the point cloud data;

[0028] Step 4: Map the point cloud data file into the mapped memory,

[0029] Step 5. Create a thread pool with a specified number of threads,

[0030] Step 6. Divide the mapped memory file into blocks, an...

specific Embodiment approach 3

[0033] Specific implementation mode three: in this implementation mode, the specific operation process of the data block mode described in step 2 is: allocate a memory buffer (such as 64KB) of a specified size, and use the file I / O function to read the memory buffer of this size File data, and find the number of newline characters ("\n") from the data to determine the number of points in the block data; if the last character in the block data is not a newline character, it indicates the last character of the block data A point data is incomplete, move the file pointer forward at this time, so that the pointer position points to the starting position of this point data; then re-use the file I / O function to read a new piece of file data, and repeat this process until the middle point of the file The data calculation is completed. Other compositions and connections are the same as those in Embodiment 1 or 2.

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a rapid large-scale point-cloud data reading method based on memory pre-distribution and multi-point writing technology, belongs to the technical field of point-cloud data file reading, and aims to solve the problems that reading time of an existing high and large-scale point-cloud data file is delayed and the existing high and large-scale point-cloud data file is slowly red. The method includes a memory pre-distribution process and a multi-point writing process and includes the steps: firstly, determining the number of points in a point-cloud data file, determining the memory size occupied by all points in the point-cloud data file and pre-distributing memories with the corresponding sizes for point-cloud data; secondly, mapping the point-cloud data file to a mapped memory through a memory mapping file mechanism, then building a thread pool containing a designated number of threads, enabling each thread to be responsible for analyzing parts of point-cloud data information in the mapped memory, and writing analyzed results into the pre-distributed memories to realize multi-point writing. Test results indicate that by the aid of the reading method based on the memory pre-distribution and multi-point writing technology, the reading speed of the point-cloud data file and particularly the large-scale point-cloud data file is increased by 220%-300%.

Description

technical field [0001] The invention relates to a method for quickly reading large-scale point cloud data, and belongs to the technical field of point cloud data file reading. Background technique [0002] With the development of reverse engineering technology, 3D scanning equipment has greatly improved in scanning scale and performance. On the one hand, it is very easy to collect 3D point cloud data expressing the surface shape of objects. The scale grows geometrically. According to different applications, the processing scale of point cloud data can range from hundreds of points to tens of millions of points, and even reach the level of hundreds of millions of points. [0003] The concept of large-scale point cloud data in the narrow sense is that the number of points on the macroscopic scale is very large, such as tens of millions, or even hundreds of millions; while the concept of large-scale point cloud data in the broad sense is related to the system memory for proces...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06F12/02G06F9/38
Inventor 张建生
Owner SOUTHWEAT UNIV OF SCI & TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products