SE-FPN-based target detection model training method and target detection method and device

A technology of SE-FPN and target detection, which is applied in the field of computer vision, can solve problems affecting detection accuracy, waste of resources, uneven distribution of targets, etc., and achieve the effect of enriching semantic information

Pending Publication Date: 2021-03-26
北京轩宇空间科技有限公司
View PDF0 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, in the data set obtained by the actual project, the scale of the target is often unevenly distributed. Most of the targets are concentrated on a certain...

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
  • SE-FPN-based target detection model training method and target detection method and device
  • SE-FPN-based target detection model training method and target detection method and device
  • SE-FPN-based target detection model training method and target detection method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0063] This example provides a SE-FPN-based target detection model training method, such as figure 1 As shown, the process includes the following steps:

[0064] S100: First randomly select four training pictures from the data set, scale them according to different scaling factors, and then splicing them into a new picture whose resolution is the input resolution set by the network model. The new picture includes multiple different sizes Target.

[0065] The target labels corresponding to the four pictures are scaled and spliced ​​in the same way as the picture, so as to maintain the consistency between the new image and the target label.

[0066] When the target is multi-category, first count the number of samples of each category, calculate the probability of each sample being drawn according to the statistical results, generate the sampling probability of the data set, and draw samples according to the sampling probability each time.

[0067] S200: Allocate multiple obje...

Embodiment 2

[0089] This example provides a SE-FPN-based target detection model training device, such as Figure 7 As shown, it includes image preprocessing module, feature extraction module, target distribution module, sample selection module, and calculation module.

[0090] The image preprocessing module scales the multiple training pictures obtained from the data set according to different scaling factors, and stitches them into a new picture, and performs the same scaling and splicing processing on the target labels corresponding to the multiple training pictures, and the new The picture includes multiple objects of different sizes.

[0091] The target allocation module allocates multiple targets of different sizes to different pyramid feature layers of the SE-FPN target detection model according to a predetermined allocation strategy.

[0092] In each pyramid feature layer, the sample selection module finds the nearest m positions, respectively m positions of all anchors and groun...

Embodiment 3

[0095] This example provides a target detection method based on the SE-FPN target detection model, such as Figure 8 shown, including the following steps:

[0096] S100: Acquire an image to be detected.

[0097] S200: Input the image to be detected into the pre-trained SE-FPN target detection model; wherein, the SE-FPN target detection model is trained by using the SE-FPN-based target detection model training method in Embodiment 1.

[0098] S300: Detect the image to be detected by using the SE-FPN target detection model to obtain a target detection result; the target detection result includes position information of the target object in the image to be detected.

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 SE-FPN-based target detection model training method and device and a target detection method and device, and the training method comprises the steps: zooming a plurality oftraining pictures according to different zooming coefficients, and splicing the training pictures into a new picture which comprises a plurality of targets of different sizes; distributing the plurality of targets with different sizes to different pyramid feature layers of an SE-FPN target detection model according to a predetermined distribution strategy; in each pyramid feature layer, finding mpositions closest to a central point according to true values of training samples of the pyramid feature layer, calculating DIoUDg of all anchors of the m positions and the true values, calculating aDg mean value mg and a standard deviation vg, obtaining a threshold tg, and selecting the central position which is larger than the tg and located in a target frame and anchor output; and calculatinga classification loss function and a position regression function, and training a model through a back propagation algorithm. The SEFPN-based target detection network model is constructed, an image preprocessing mode and a sample selection strategy are improved, the model is trained, the model is applied to target detection, and the target detection efficiency is improved.

Description

technical field [0001] The invention relates to the field of computer vision, in particular to a SE-FPN-based target detection model training method, target detection method and device. Background technique [0002] Object detection is a basic research topic in the field of computer vision, and it is widely used in areas such as unmanned driving, intelligent monitoring, and automatic object recognition. Traditional object detection methods mainly perform region selection and localization through sliding windows, and then classify them through support vector machines (SVM) and other classifiers. With the development of deep learning, convolutional neural network has achieved a series of results in target detection. Compared with traditional target detection methods, the detection method based on deep convolutional network model has the advantages of independent feature extraction and strong generalization ability. , has become one of the important research topics in the fiel...

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
IPC IPC(8): G06T3/40G06N3/08G06N3/04G06K9/62
CPCG06T3/40G06T3/4038G06N3/084G06V2201/07G06N3/045G06F18/214
Inventor 谷晓琳杨敏张燚刘科
Owner 北京轩宇空间科技有限公司
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