Unlock instant, AI-driven research and patent intelligence for your innovation.

Vehicle target identification method and storage medium

A target recognition and vehicle technology, applied in the field of data processing, can solve the problems of reduced recognition accuracy and inability to provide 3D geometric information, and achieve the effects of ensuring accuracy, reducing data processing volume, and reducing quantity

Active Publication Date: 2022-04-15
成都奥伦达科技有限公司 +1
View PDF11 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] Traditional vehicle recognition methods mainly rely on images acquired by visual sensors, using machine learning or deep learning methods for recognition (Han Xiaofeng et al., 2019); however, images are easily affected by different lighting and shadows, and cannot provide reliable 3D geometric information, leading to a reduction in recognition accuracy

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
  • Vehicle target identification method and storage medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0050] It should be noted that, in the coordinate system of this embodiment, the forward direction of the scanner is taken as the X-axis, and the elevation direction is taken as the Z-axis.

[0051] Such as figure 1 As shown, a vehicle target recognition method includes:

[0052] Step 1: Build a feature database, use the feature database to train the support vector machine model, and obtain a classification model; wherein, the feature vectors in the feature database include global features, position features, eigenvalue features, and multi-view projection features.

[0053] The global feature is used to describe the shape, geometric size and other attributes of the entire independent ground object. For vehicles, its geometric shape features are obviously different from other ground objects, and its global feature design includes length L, width W, and height H features and the ratio between length, width, and height, volume, relative density, minimum height, and height diffe...

Embodiment 2

[0088] The difference from Embodiment 1 is that in this embodiment, in step 3 of the vehicle target recognition method, when the point cloud data is scanned repeatedly and the overlap between the front and rear frames is higher than the preset threshold, the offline frame point cloud can be automatically selected. The processing method is calculated; when the overlapping degree of the front and rear frames is lower than the set threshold, this scheme automatically determines that it is real-time frame data; when switching between the implementation frame and the offline frame, the overlapping area is automatically calculated to avoid repeated identification of the overlapping area.

[0089] Through such settings, the application can automatically distinguish real-time frames from offline frames. If the driving speed is slow, such as traffic jams, etc., the point cloud data is scanned repeatedly, and the overlap between the front and rear frames is higher than the set threshold,...

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 belongs to the technical field of data processing, and particularly relates to a vehicle target recognition method and a storage medium, and the method comprises the steps: 1, constructing a feature database, and training a support vector machine model through the feature database, and obtaining a classification model; 2, preprocessing the laser radar data to obtain collected data, removing ground points in the collected data through ground filtering to obtain non-ground points, and removing apogees in the non-ground points by using elevation information of the non-ground points to obtain a candidate point set; 3, according to the types of the candidate points, after a clustering interval is determined in a corresponding mode, a target object is obtained through a density clustering method; wherein the types of the candidate points comprise real-time frame point clouds and offline point clouds; and 4, after extracting the feature vector of the target object, predicting the target object by using the classification model to complete vehicle target identification. By using the method, the accuracy, efficiency and applicability of vehicle target extraction can be improved.

Description

technical field [0001] The invention belongs to the technical field of data processing, and in particular relates to a vehicle target recognition method and a storage medium. Background technique [0002] According to the latest statistics from the Ministry of Public Security in 2020, there are 372 million motor vehicles in China, accounting for about 75% of the total number of vehicles, totaling 281 million. The ever-increasing number of cars has put enormous pressure on road traffic, the incidence of traffic accidents has increased year by year, and problems such as traffic congestion have intensified, which have a direct impact on people's life safety and production and life. In order to make people's driving safer and more worry-free, intelligent driving technology has emerged. Intelligent driving technology can directly control the vehicle or assist the driver to control the vehicle, avoiding dangerous accidents caused by driver fatigue, distraction, speeding, and othe...

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): G06V20/58G06V10/40G06V10/762G06V10/77G06V10/774G06V10/764G06K9/62
CPCY02T10/40
Inventor 刘健飞束子贤江亮亮余建乐魏新元
Owner 成都奥伦达科技有限公司