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

Garage pedestrian detection method based on improved efficientdet model

A pedestrian detection and model technology, applied in biological neural network models, character and pattern recognition, image enhancement, etc., can solve the problems of complex and changeable garage environment, and achieve the effect of enhancing learning ability, reducing memory cost, and reducing computing bottleneck.

Active Publication Date: 2021-07-02
XUZHOU RITMAN EQUIP CO LTD
View PDF6 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the garage environment is complex and changeable, and the detection target has its uniqueness. Directly using EfficientDet to train the target detector, although the effect is good, there is still a lot of room for improvement, mainly in the positioning accuracy, detection speed and misjudgment rate. Improve

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
  • Garage pedestrian detection method based on improved efficientdet model
  • Garage pedestrian detection method based on improved efficientdet model
  • Garage pedestrian detection method based on improved efficientdet model

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0071] This specific embodiment discloses the garage pedestrian detection method based on the improved EfficientDet model, such as Figure 1 to Figure 7 shown, including the following steps:

[0072] S1: Collect images of garage pedestrians in different time periods and lighting environments;

[0073] S2: if figure 1 As shown, the sample input network needs to be preprocessed and data enhanced before training. For the garage pedestrian image, first cut it into a uniform size, then perform horizontal flip (50% probability) and standardization, and finally use the mosaic data enhancement method. Randomly extract 4 images to generate a composite image, and convert the corresponding label data to generate training samples (such as figure 2 shown);

[0074] S3: This article takes EfficientDet-D0 as an example, in the backbone network EfficientNet-b0 (such as image 3 As shown), the feature distribution network CSPNet is introduced to enhance the learning ability of CNN, while ...

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 garage pedestrian detection method based on the improved EfficientDet model, which belongs to the technical field of target detection in image processing. The invention uses a mosaic data enhancement method to enrich the background information of pedestrian detection, and calculates four times at a time during batch normalization calculation. image data; introduce the feature splitting network CSPNet in the backbone network EfficientNet, enhance the learning ability of CNN, can maintain the accuracy of detection while lightweighting the model, and reduce computing bottlenecks and memory costs; introduce space at the top of the feature extraction network The pyramid pooling module SPP increases the receptive field of the network, and can accurately and quickly complete pedestrian detection in complex and changeable garage environments.

Description

technical field [0001] The invention belongs to the technical field of target detection in image processing, in particular to a garage pedestrian detection method based on an improved EfficientDet model. Background technique [0002] The smart three-dimensional garage is an important part of the process of intelligent city construction. It integrates garage parking space reservation, license plate recognition, automatic parking, and pedestrian detection. Among them, the pedestrian detection in the garage is to ensure the safety of pedestrians in the garage. The environment in the garage is complex and changeable, and the pedestrians in the garage must be taken into account when taking off and landing the parking space, so as to ensure that it can only be lifted and lowered without pedestrians. Therefore, the real-time and accuracy of garage pedestrian detection are very important for the deployment of smart three-dimensional garages. [0003] Garage pedestrian detection is...

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 Patents(China)
IPC IPC(8): G06K9/00G06K9/54G06K9/62G06N3/04G06T5/50
CPCG06T5/50G06T2207/20221G06V40/20G06V10/20G06V2201/07G06N3/045G06F18/24G06F18/253G06F18/214
Inventor 牛丹李永胜陈夕松许翠红陈善龙刘子璇
Owner XUZHOU RITMAN EQUIP CO LTD
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

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com