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

Intelligent automobile pedestrian detection algorithm

A pedestrian detection and smart car technology, applied in the field of smart car pedestrian detection algorithms, can solve problems such as the inability to widely use accurate detection of various target objects, low bus detection accuracy, and low motorcycle detection accuracy, and achieve both accuracy. and real-time performance, good robustness and generalization ability, detection accuracy and detection speed improvement effect

Pending Publication Date: 2022-05-17
JILIN UNIV
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0002] The detection accuracy of the original YOLOv3 network model for pedestrian targets is only 79.20%, and the detection performance is relatively low. The detection accuracy of buses is low, and the detection accuracy of motorcycles is the lowest. The detection accuracy of different types of objects has declined to varying degrees. The average detection accuracy of the YOLOv3 network model for all target objects is 84.00%, and it cannot be widely used in the accurate detection of various target objects

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
  • Intelligent automobile pedestrian detection algorithm
  • Intelligent automobile pedestrian detection algorithm
  • Intelligent automobile pedestrian detection algorithm

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0030] Such as figure 1 Shown, a kind of intelligent automobile pedestrian detection algorithm comprises improved k-means clustering algorithm, is characterized in that, improved k-means clustering algorithm comprises the following steps:

[0031] Step 1: Clear the invalid labeled data in the training data set;

[0032] Step 2: Write the coordinate data from the data file corresponding to the training data set into the array;

[0033] Step 3: Read the array data sequentially, and define the projection coordinate of the vertex in the lower left corner of the label box on the x-axis as x min , the projected coordinate on the y-axis is y min , define the projection coordinate of the vertex in the upper right corner of the label box on the x-axis as x max , the projected coordinate on the y-axis is y max ;

[0034] Step 4: Calculate x max with x min the difference, calculate y max with y min the difference;

[0035] Step 5: Calculate x d and y d business;

[0036] Ste...

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 an intelligent automobile pedestrian detection algorithm, which comprises an improved k-means clustering algorithm, and is used for clearing invalid labeled data in a training data set; writing the coordinate data into an array from a data file corresponding to the training data set; obtaining all effective labeled data in the training data set; k clusters are manually selected, and IOU values of all effective annotation data and a clustering center are calculated; selecting the center of all data points in each cluster as a new clustering center; the clustering center does not move any more; and taking the final clustering result as a prior frame obtained by unsupervised learning of the YOLOv3 network model. The method has the beneficial effects that the division size of the grid unit is improved, the pedestrian detection effect under various complex scenes is tested, the robustness and generalization capability are good, the network stability is higher, the anti-interference capability is stronger, and the detection precision and the detection speed are remarkably improved.

Description

technical field [0001] The invention relates to the technical field of pedestrian detection algorithms, in particular to a pedestrian detection algorithm for intelligent vehicles. Background technique [0002] The detection accuracy of the original YOLOv3 network model for pedestrian targets is only 79.20%, and the detection performance is relatively low. The detection accuracy of buses is low, and the detection accuracy of motorcycles is the lowest. The detection accuracy of different types of objects has declined to varying degrees. The average detection accuracy of the YOLOv3 network model for all target objects is 84.00%, and it cannot be widely used in the accurate detection of various target objects. Contents of the invention [0003] The object of the present invention is to provide a kind of intelligent automobile pedestrian detection algorithm, to solve the problem that proposes in the above-mentioned background technology. [0004] For achieving the above object...

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): G06V20/56G06K9/62G06V10/774
CPCG06F18/23213G06F18/214
Inventor 肖峰宋传学曹景伟安靖宇孙发荣
Owner JILIN UNIV