Marine water surface garbage rapid identification method based on multi-feature YOLOV3

A technology for surface garbage and identification methods, which is applied in neural learning methods, character and pattern recognition, image data processing, etc. It can prevent the influence of water surface light and image noise, solve the inaccurate garbage classification, and ensure the effect of continuous tracking.

Pending Publication Date: 2020-11-17
航天时代(青岛)海洋装备科技发展有限公司
View PDF0 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

At present, most of the water surface garbage detection algorithms are based on the improved method of the mixed Gaussian model and the double background model, which are easily affected by light and noise in complex water surface scenes.
There are problems such as high target false alarm rate and inaccurate object positioning
At the same time, the parameter adjustment process of the above method is complicated, and it is difficult to meet the requirements of rapid identification of water surface garbage

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
  • Marine water surface garbage rapid identification method based on multi-feature YOLOV3
  • Marine water surface garbage rapid identification method based on multi-feature YOLOV3
  • Marine water surface garbage rapid identification method based on multi-feature YOLOV3

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0031] The present invention will be further elaborated below in conjunction with embodiment.

[0032] Such as figure 1 , the implementation steps of this method are as follows:

[0033] A. Read in the image data, perform preprocessing such as histogram equalization and feature learning on the image

[0034] figure 2 , use the ship-mounted camera to obtain real-time 1080P RGB images and compress them into grayscale images, extract the gradient features and morphological feature maps of the grayscale images, use the grayscale images as the first band, and the gradient feature images as the second band and morphological features As the third band, the processed image is used to construct an image feature description subgraph, and its resolution is converted to 416*416.

[0035] B. Establish a target recognition model to detect water surface garbage in real time

[0036] image 3 , first initialize the parameters of the YOLOV3-tiny algorithm, read the configuration file, an...

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 a marine water surface garbage rapid identification method based on multi-feature YOLOV3, and the method is realized through the following steps: S1, photographing an image through a shipborne photographing device, carrying out the histogram equalization preprocessing of the image, and constructing an image feature description subgraph; s2, constructing a target recognition model based on a YOLOV3-tiny target detection method; s3, identifying a target existing in the image feature description sub-graph processed in the step S1 by using a target identification model; s4, judging the distance between the garbage and the ship according to the detected target coordinates, and when the distance is smaller than a set threshold value, collecting the garbage; and S5, carrying out image augmentation processing and model training on an unrecognized image collected from the land base to obtain a new target recognition model, and returning to S3 to execute the steps againuntil the accuracy of the constructed target recognition model meets the requirements.

Description

technical field [0001] The invention relates to the fields of computer vision, image processing, deep learning and artificial intelligence. It specifically relates to a fast identification method for marine garbage based on multi-feature YOLOV3. Background technique [0002] Intelligent garbage collection is a new field that includes multiple technologies such as image processing, artificial intelligence and automatic control. The intelligence of environmental protection equipment is closely related to our life. With the development of economy, the pollution of water surface is getting more and more serious. [0003] There are more and more researches on the rapid identification method of water surface litter. At present, most of the water surface garbage detection algorithms are based on the improved method of the mixed Gaussian model and the double background model, which are easily affected by light and noise in complex water surface scenes. There are problems such as ...

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): G06K9/00G06K9/34G06T5/40G06N3/04G06N3/08
CPCG06T5/40G06N3/08G06V20/46G06V20/41G06V10/267G06V2201/07G06N3/045
Inventor 李威蔡立明高永发谢家文戴智航符浩胡常青
Owner 航天时代(青岛)海洋装备科技发展有限公司
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