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

Multi-target detection method based on improved VGG16 network

A detection method and multi-target technology, applied in biological neural network models, image data processing, instruments, etc., can solve the problems of low recognition accuracy, slow recognition, cumbersome operation, etc., and achieve the effect of accurate classification probability

Active Publication Date: 2020-11-24
ZHEJIANG UNIV OF TECH
View PDF8 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] In order to overcome the above-mentioned deficiencies of the prior art, the present invention provides a multi-target detection method based on the improved VGG16 network for the problems of traditional detection methods such as cumbersome operation, low recognition accuracy, and slow recognition.

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-target detection method based on improved VGG16 network
  • Multi-target detection method based on improved VGG16 network
  • Multi-target detection method based on improved VGG16 network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0111] The technical solution of the present invention will be further described below in conjunction with the accompanying drawings.

[0112] In order to overcome the above-mentioned deficiencies of the prior art, the present invention provides a multi-target detection method based on the improved VGG16 network to solve the problems of traditional detection methods such as cumbersome operation, low recognition accuracy, and slow recognition. First, image enhancement is performed on the collected sample images to make the foreground and background of the sample images more distinct; then, the feature extraction model is constructed using the improved VGG16, and the model parameters are designed reasonably; The target is positioned to frame the candidate boundary; finally, the loss of the candidate bounding box is calculated to obtain a more accurate bounding box and the corresponding classification probability.

[0113] To achieve the above object, the present invention adopts...

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 multi-target detection method based on an improved VGG16 network. The multi-target detection method comprises the steps of 1, sample image enhancement in which a histogram equalization and histogram matching method is selected, and the display effect of the image is changed by changing the gray histogram of the image; 2, feature extraction model construction in whicha feature extraction network model is established and composed of a truncated VGGNet-16 network and an enhanced network layer, feature maps of different scales corresponding to parameters are generated ineach layer, target object detection is carried out on the feature maps of the different scales at the same time, and the feature maps of the different scales are used for predicting target objects ofdifferent scales; 3, setting a feature extraction model correlation function; 4, positioning a target on the extracted feature map; and 5, target positioning and setting of a feature classification loss function. According to the method, the recognition efficiency can be improved while the recognition precision is improved, so that the problems of difficult detection and difficult classification are solved.

Description

technical field [0001] The invention relates to a multi-target detection method based on an improved VGG16 network. [0002] technical background [0003] In recent years, with the rapid development of computer science and technology, image processing and image target detection based on computer technology have also achieved unprecedented rapid development. Among them, deep learning extracts key target features by learning massive digital image features. It has surpassed humans in detection and brought one surprise after another to the industry. With the re-emergence of neural networks, the video image method based on convolutional neural networks has become the mainstream technology of image segmentation and recognition, using template matching, edge feature extraction, gradient histogram and other means to achieve accurate recognition of images. Although image feature recognition based on neural network can effectively identify features for targets in complex scenes, and i...

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): G06T5/40G06T7/90G06T7/11G06N3/04G06K9/62G06K9/46
CPCG06T5/40G06T7/90G06T7/11G06V10/44G06V2201/07G06N3/045G06F18/24
Inventor 张烨樊一超陈威慧
Owner ZHEJIANG UNIV OF TECH
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