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

Underwater image recognition method based on Lab domain enhancement, classification and contrast improvement

An underwater image and recognition method technology, which is applied in image enhancement, image analysis, image data processing, etc., can solve the problem of poor enhancement effect, and achieve the effect of retaining image clarity, good restoration effect, and low complexity

Active Publication Date: 2021-06-04
BEIJING INSTITUTE OF TECHNOLOGYGY +1
View PDF7 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to aim at the situation that the pictures or videos collected under different lighting conditions in the underwater environment have poor enhancement effects, and propose an underwater image recognition method based on Lab domain enhancement, classification and contrast enhancement, and adjust the image color mode In the Lab mode, the white balance of the Lab color gamut is performed; then, the histogram equalization is performed based on CLAHE to obtain an enhanced image with increased illumination; finally, the Faster RCNN model is used to identify the underwater organisms in the image

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
  • Underwater image recognition method based on Lab domain enhancement, classification and contrast improvement
  • Underwater image recognition method based on Lab domain enhancement, classification and contrast improvement

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0061] This method is suitable for underwater bluish, greenish, and low-illumination scenes caused by light absorption. It is dominated by the atmospheric imaging model, combined with traditional image enhancement methods, and can be effectively corrected according to the color attributes of the picture. , and adaptively improve the contrast, which has a good enhancement effect on the Bubble Vision website and the SQUID dataset (Stereo Quantitative Underwater Image Dataset), and can restore a clearer visual scene. Randomly enhance the underwater image that needs to be identified, and to a certain extent, it can achieve a better recognition effect on underwater organisms. In specific implementation, such as figure 1 As shown, Boakai follows the steps:

[0062] Step 1. Randomly select 30% of the underwater images; during specific implementation, collect underwater image information, and select images from the Real-world Underwater Image Enhancement (RUIE) dataset during specific...

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 an underwater image recognition method based on Lab domain enhancement, classification and contrast improvement, and belongs to the technical field of image enhancement and recovery. The method comprises the following steps: 1) randomly selecting fixed-dimensional images with the proportion of x% from fixed-dimensional images, performing color correction and enhancement on each image, and forming a training set with another fixed-dimensional image with the proportion of 1-x%; 2) performing feature extraction on the images in the training set to obtain a feature map; (3) the feature map is input into the RPN to generate region proposals; 4) inputting the region proposals and the feature map into the ROI pooling layer to generate a feature map with a fixed size; and 5) carrying out a group of full connection operations on the proposal feature map with the fixed size, and then respectively carrying out binary classification and frame regression to generate a classification result and an accurate position of the object. The image definition is reserved to the maximum extent, and a better restoration effect is ensured; the method has high advantages in the aspects of identification precision, complexity and operation power consumption, and can be used for real-time high-precision detection.

Description

technical field [0001] The invention relates to an underwater image recognition method based on Lab domain enhancement, classification and contrast enhancement, and belongs to the technical field of image enhancement and restoration. Background technique [0002] There are abundant resources in the ocean. As an important carrier of ocean information, underwater images have received more and more attention and attention. At the same time, underwater image enhancement technology has also been widely used, such as underwater target detection and tracking, underwater Infrastructure inspections and marine life research, among others. Unlike general images, underwater images have poor visibility due to light absorption and scattering effects. Among them, the absorption of light reduces the light intensity, and the color of the underwater image is distorted due to the different propagation distances of the spectrum of different wavelengths in water; the scattering of light causes ...

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): G06T5/00G06T7/90G06K9/32G06K9/62
CPCG06T7/90G06V10/25G06F18/241G06T5/90
Inventor 卢继华张延军刘国栋王欢尹鹏许欣尤信群谢民
Owner BEIJING INSTITUTE OF TECHNOLOGYGY
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