Small target pedestrian detection method based on improved YOLO algorithm

A pedestrian detection and small target technology, which is applied in the field of computer vision technology and intelligent transportation, can solve the problems that the detection accuracy and real-time performance cannot meet the actual needs, the proportion of pedestrians occupying the image is small, and the recognition rate is low.

Pending Publication Date: 2021-05-07
ZHEJIANG UNIVERSITY OF SCIENCE AND TECHNOLOGY
View PDF0 Cites 10 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, its detection accuracy and real-time performance still cannot meet the actual needs. In actual traffic detection tasks, pedestrians occupy a small proportion of the image, and the recognition rate is low in dark environments and occlusions.

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
  • Small target pedestrian detection method based on improved YOLO algorithm
  • Small target pedestrian detection method based on improved YOLO algorithm
  • Small target pedestrian detection method based on improved YOLO algorithm

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0049] The platform configuration for deep learning training of the model in this embodiment is as follows:

[0050]

[0051] Table 1 Training platform configuration

[0052]The deep learning data set of this embodiment is produced on the basis of open source data sets: KITTI data set and INRIA data set. The KITTI data set is the world's largest computer vision algorithm evaluation data set in autonomous driving scenarios, and the training set contains 7481 images. Car camera photos of vehicles, pedestrians, etc. The INRIA dataset contains 902 pedestrian photos.

[0053] The steps of the present invention are:

[0054] 1. Data processing

[0055] In this embodiment, it is necessary to filter, fill, and integrate the pictures in each training set. The specific operation is: select a total of 1223 pictures in the KITTI dataset with pedestrian targets, and fill a single picture with gray pixels to a size of 1248×416, and then splicing Three photos, the effect is shown in Fi...

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 relates to the technical field of computer vision and intelligent traffic, and discloses a small target pedestrian detection method based on an improved YOLO algorithm. The method comprises the following steps of: firstly, making a small target pedestrian data set by using KITTI and INRIA data sets; secondly, re-clustering a pre-selected box by adopting a k-means algorithm based on the data set; thirdly, on the basis of a YOLO-V3 model, using a Mish activation function for replacing ReLU, simplifying a feature extraction network , and using a PANet structure for feature fusion; and finally, optimizing a loss function, and calculating a coordinate error by using CIoU. Compared with a YOLO-V3 model, the improved algorithm has the advantages that the network reasoning speed is increased by 3.2 AP and 20.8%, and the algorithm has certain practicability in small-target pedestrian detection tasks.

Description

technical field [0001] The invention relates to the fields of computer vision technology and intelligent transportation technology, in particular to a small target pedestrian detection method based on an improved YOLO algorithm. Background technique [0002] At present, in the development of intelligent transportation and smart cities, intelligent driving technology is a research hotspot in the industry. During the driving process, the intelligent system needs to detect the objects in the surrounding environment of the vehicle, such as vehicles, traffic signs, pedestrians, etc. Small target pedestrians Due to factors such as low pixel ratio and easy occlusion, in actual detection tasks, the recognition accuracy is usually not high. Therefore, improving the detection accuracy of small target pedestrians and reducing the detection delay are the goals that the industry is constantly pursuing. [0003] As one of the important deep models, CNN (convolutional neural network) can e...

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): G06K9/00G06K9/62G06N3/04G06N3/08
CPCG06N3/04G06N3/08G06V40/10G06F18/23213G06F18/25
Inventor 徐兴王凯耀赵芸
Owner ZHEJIANG UNIVERSITY OF SCIENCE AND TECHNOLOGY
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap