A method and device for identifying seawater polluted areas based on high-resolution remote sensing images

Abstract

The invention discloses a seawater polluted area identification method and equipment based on high-resolution remote sensing images, belonging to the field of digital image processing. The method first uses a supervised learning algorithm to automatically classify remote sensing images of sea and land, and through process-based iterative clustering, the classification results can reach a higher level of accuracy. Then, using the chlorophyll concentration difference between the marine polluted area and the surrounding seawater and the brightness difference of the pollutant shadow, the normalized vegetation index related to chlorophyll and the normalized water shadow index related to brightness in remote sensing interpretation are based on Combining the image interpretation idea of ​​segmentation with the saliency mechanism based on human vision, the extraction of marine polluted areas is realized through threshold segmentation, and the areas with good water quality and severely polluted areas are extracted respectively, and the transitional areas of pollution are further obtained. The method of the invention provides a convenient and accurate reference for marine pollution prevention and treatment.

Description

technical field [0001] The invention belongs to the technical field of digital image processing, and in particular relates to a method and equipment for identifying seawater polluted areas based on high-resolution remote sensing images. Background technique [0002] Today, as the global land resources are increasingly tense and the environment is deteriorating, countries all over the world are turning their attention to the ocean. The development of marine resources and the development of marine economy have become an important pillar of the national economy of coastal countries and the forefront of sustainable development strategies. While modern ocean development brings huge economic benefits, it also brings a series of resources and ecological environment problems. At present, the pollution of the marine environment is becoming more and more serious, and the ecological environment is also becoming worse and worse. Marine economic industries such as marine fishery and mar...

AI Technical Summary

Problems solved by technology

However, in these methods, there are constraints such as the large amount of calculation of the convolutional layer of the deep learning convolutional neural network model, high requirements for computing equipment, huge capital investment, and the need for guidance from professional and technical personnel; and the Grab-Cut image segmentation algorithm focuses on solving the original There are problems such as the accuracy of the remote sensing image algorithm segmentation scale and level, the lack of the "segmentation-classification" link, and the lack of discussion on the classification and identification of specific pollutants

Images