Multi-scale target detection method and system for smoothly transmitting semantic information

A technology of target detection and semantic information, applied in the field of computer vision, can solve the problems of insufficient robustness and low generalization of the detector, and achieve the effect of solving the limited use, reducing the number of parameters, and improving the operation efficiency.

Pending Publication Date: 2022-03-25
YANGTZE DELTA REGION INST OF UNIV OF ELECTRONICS SCI & TECH OF CHINE HUZHOU
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Therefore, the detector is not robust enough and does not generalize well

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
  • Multi-scale target detection method and system for smoothly transmitting semantic information
  • Multi-scale target detection method and system for smoothly transmitting semantic information
  • Multi-scale target detection method and system for smoothly transmitting semantic information

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0106] Step 1, image data preparation.

[0107] The dataset used for training, validation and testing in this paper is the public dataset Pascal VOC 2007. It consists of 21 classes (20 foreground and 1 background). The images in the dataset are RGB three-channel, and each channel has 8-bit depth, so each image has a bit depth of 24. Considering that objects in real life are often subject to various disturbances, it is not as simple as the samples in the dataset. Models trained with simple datasets are difficult to apply to complex and varied real-world scenarios, that is, the generalization ability is not strong. To overcome such problems, offline data augmentation operations are introduced before training. We performed flipping, rotating, cropping, scaling, shifting, edge filling, color space conversion, noise, blurring, and random erasing without changing the relevant information of the labels to enhance the generalization ability of the model.

[0108] Step 2. Network t...

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 provides a multi-scale target detection method and system for smoothly transmitting semantic information, relates to the technical field of computer vision, and aims to improve or prevent degradation of a deep network, completely and efficiently extract image features and make pixel classification more accurate. The method comprises the following steps: S1, constructing a residual block for placing an ReLU layer in front of a convolutional layer; s2, constructing a first-stage backbone network by adopting the residual blocks in the S1; s3, performing feature extraction on the image by using the first-stage backbone network to obtain a feature map; s4, inputting the feature map obtained in the S3 into a second-stage target detection network for processing to obtain a region suggestion; s5, mapping the region suggestion to the feature map to obtain a region of interest, and performing ROI Align processing to obtain a target detection result; in the second stage, the target detection network integrates a cross entropy loss function which introduces a balance coefficient; and the convolution layer of the residual block is subjected to feature image cutting and recombination operation in the convolution process.

Description

technical field [0001] The invention relates to the technical field of computer vision, in particular to a multi-scale target detection method and system for smoothly transmitting semantic information. Background technique [0002] In recent years, computer vision has flourished with the support of deep learning, and object detection has shown great practical and research value in many fields of computer vision. Many object detection methods proposed in the early years are based on artificially designed feature construction. The disadvantage of these detectors is their lack of versatility. Currently, more and more detection models based on convolutional neural network (CNN) have been proposed. By combining with CNNs, object detectors show powerful capabilities by handling complex tasks. There are two types of detectors: (1) one-stage method and (2) two-stage method. For object detectors using a two-stage approach, in the first stage, the detector tries to find a set of c...

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): G06V10/25G06V10/44G06V10/764G06V10/82G06K9/62G06N3/04G06N3/08
CPCG06N3/08G06N3/047G06N3/048G06N3/045G06F18/2415G06F18/241
Inventor 饶云波牟洪雨曾少宁杨泽雨王发新郑伟斌
Owner YANGTZE DELTA REGION INST OF UNIV OF ELECTRONICS SCI & TECH OF CHINE HUZHOU
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