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

A fully convolutional network-based oil spill detection method on the sea surface and its system and application

A fully convolutional network, sea surface oil spill technology, applied in the field of sea surface oil spill detection, can solve the problems of difficult segmentation of oil spill and sea water background, inability to take into account detection accuracy and speed, low segmentation accuracy, etc., to reduce memory and Complex calculation, solving scale change problem, the effect of less calculation

Active Publication Date: 2022-08-09
CHINA UNIV OF PETROLEUM (EAST CHINA)
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0007] In order to solve the problems in the prior art that the segmentation of oil spill and seawater background is difficult, the segmentation accuracy is low, and the detection accuracy and speed cannot be taken into account, the present invention provides a sea surface oil spill detection method based on a full convolutional network, which reduces memory In the case of complex calculations, the detection accuracy of the oil film is improved, while maintaining the real-time detection speed

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
  • A fully convolutional network-based oil spill detection method on the sea surface and its system and application
  • A fully convolutional network-based oil spill detection method on the sea surface and its system and application
  • A fully convolutional network-based oil spill detection method on the sea surface and its system and application

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0100] like figure 1 As shown, a method flow chart of a fully convolutional network-based sea surface oil spill detection method is disclosed, and the specific operations are as follows:

[0101] 1) First, obtain an image of oil spill on the sea surface, and perform a preprocessing operation on the oil spill image to obtain an image data set. The oil spill image is input into a fully convolutional network to perform binary segmentation of the oil film area and the background image, and then the spectrogram of the oil film area is extracted.

[0102] 2) combining the image data set with the spectrogram of the oil film as a training sample set; labeling the training sample set to obtain label information of the training sample set;

[0103] then input the label information of the training sample set into the pre-built initial sea surface oil spill detection network;

[0104] 3) The parameters of the initial sea surface oil spill detection network are optimized by the back prop...

Embodiment 2

[0108] like figure 2 As shown, it is a flow chart of a sea surface oil spill detection method based on a fully convolutional network:

[0109] Step S01: simulate a sea surface oil spill site in the laboratory and take an oil spill image as a data training set, perform preprocessing on the oil spill image and segment it to obtain an oil film area, and obtain a spectrogram obtained by two-dimensional Fourier transform and annotated images together as a training sample set.

[0110] The preprocessing operation mainly includes: (1) The adaptive histogram equalization method for limiting contrast solves the problem of unsatisfactory contrast while suppressing the enhancement of noise. (2) The high-frequency enhancement filtering method sharpens and enhances the edge of the image, and stretches the grayscale distribution of the image to a certain extent. The main purpose is to make the oil spill image clearer and enhance the contrast of seawater and oil film to improve segmentati...

Embodiment 3

[0118] like Figure 4 As shown, this embodiment provides a sea surface oil spill detection system based on a fully convolutional network, which is an end-to-end one-stage anchor-free detection network, including a backbone network, a Transformer module, a Transformed feature pyramid, and a head network.

[0119] The detection network is based on full convolution and is used to predict the class information and regression information of each pixel in the image.

[0120] The backbone network adopts ResNet-50 network, which is used to extract and save the feature information of oil spill images.

[0121] The Transformer module mainly includes single-scale transformation, cross-scale transformation and Gateblock, and increasing the global information of features is beneficial to the detection of large-area oil films. ;

[0122] The Transformed feature pyramid solves the problem of scale change, and at the same time obtains multi-scale context information to establish the connect...

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 sea surface oil spill detection method and system, comprising: inputting a sea surface oil spill image as a data set into an initial sea surface oil spill detection network, extracting multi-scale feature information of the image, and converting the image multi-scale feature information through a Transformer module. The information is fused to obtain the Transformed feature pyramid, and the prediction information of each pixel of the head network image of the feature pyramid is used; the error between the prediction information and the label is calculated by using the loss function; the network parameters are optimized by the back propagation algorithm until the error reaches the expected value, and the sea surface overflow is obtained. Oil detection network, output detection information. If there is an oil film in the environmental photos, it is determined that there is an oil spill on the sea surface, and an early warning signal is issued. The invention extracts the multi-scale features of the image, obtains the Transformed feature pyramid, solves the problem of scale change, and can quickly and accurately identify the oil film in the field photos.

Description

technical field [0001] The invention belongs to the field of computer vision, in particular to a sea surface oil spill detection method based on a fully convolutional network, a system and an application thereof. Background technique [0002] The information disclosed in this Background section is only for enhancement of understanding of the general background of the invention and should not necessarily be taken as an acknowledgement or any form of suggestion that this information forms the prior art already known to a person of ordinary skill in the art. [0003] In the global marine pollution, the most common is oil spill pollution. Oil spills destroy marine ecosystems and affect human health and sea surface activities. It is a key research direction to detect oil spills in a timely and accurate manner and issue an alert at the first time. Therefore, the rapid and accurate detection of oil spills on the sea surface is the most important task at present, and the complex s...

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): G06T7/00G06T7/12G06T7/13G06V10/44G06V10/80G06V10/774G06V10/82G06K9/62G06N3/04G06N3/08
CPCG06T7/0002G06T7/12G06T7/13G06N3/08G06T2207/10004G06T2207/20081G06T2207/20084G06T2207/20016G06V10/44G06N3/045G06F18/253G06F18/214Y02A20/204
Inventor 梁鸿杨莹巩亚明魏学成张千
Owner CHINA UNIV OF PETROLEUM (EAST CHINA)
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