A method and a system for constructing an underwater image data set

Abstract

The invention belongs to the field of image processing, and discloses an underwater image data set construction method and system, and the method comprises the steps: constructing underwater fuzzy image data sets of various types, and building a background light data set based on a manual marking method; meanwhile, evaluzting the underwater image quality based on four-dimensional area evaluation indexes of image chromaticity, contrast ratio, gradient and sharpness, so that constructing an underwater high-definition image data set. Based on human subjective discrimination, the accuracy of output underwater background light can be ensured; the constructed underwater blurred image data set and the background light data set thereof can be used as a data source and background light estimation reference in underwater image restoration research; the underwater high-definition image data set construction method is simple and rapid, the constructed underwater image data set can provide trainingsamples for an underwater image quality enhancement algorithm based on deep learning, and the underwater image quality evaluation method can be used for underwater image quality evaluation.

Description

technical field [0001] The invention belongs to the field of image processing, and in particular relates to a method and system for constructing an underwater image data set. Background technique [0002] At present, the existing technologies commonly used in the industry are as follows: [0003] Image datasets are constructed by crowdsourcing. For example, the ImageNet dataset (http: / / image-net.org / index) is currently one of the most widely used datasets in the field of deep learning images, including more than 14 million images, covering more than 20,000 categories. Similar data sets also include CIFAR (http: / / www.cs.toronto.edu / ~kriz / cifar.html) of the Canadian Institute of Advanced Technology, Google's Open Images (https: / / storage.googleapis.com / openimages / web / index.html), and LFW (http: / / vis-www.cs.umass.edu / lfw / ) for face recognition, etc. Crowdsourcing methods can build very large data sets, but may require huge financial support and maintenance personnel, and hav...

AI Technical Summary

Problems solved by technology

Lack of effective background light construction method

[0013] (3) In recent years, more and more deep learning methods have been used to deal with attenuated underwater images. However, since there is currently no public underwater image dataset as a training sample for deep learning methods, the research uses artificial A synthetic underwater image dataset or an image set enhanced with an existing algorithm as a reference label

[0017] Due to the absorption and scattering of underwater light, the uneven light of artificial light sources, and the characteristics of water itself, the obtained underwater images generally have problems such as chromatic aberration, low contrast, and poor clarity.

To establish training samples that can be used for deep learning underwater image processing, high-quality underwater images are needed as labels, but there is no public high-quality underwater image data set so far, from collecting a large number of underwater images to how to judge underwater Image quality is the difficulty in establishing high-definition underwater image datasets

Images