Abstract
Information, especially spatial distribution data, related to coastal raft aquaculture is critical to the sustainable development of marine resources and environmental protection. Commercial high spatial resolution satellite imagery can accurately locate raft aquaculture. However, this type of analysis using this expensive imagery requires a large number of images. In contrast, medium resolution satellite imagery, such as Landsat 8 images, are available at no cost, cover large areas with less data volume, and provide acceptable results. Therefore, we used Landsat 8 images to extract the presence of coastal raft aquaculture. Because the high chlorophyll concentration of coastal raft aquaculture areas cause the Normalized Difference Vegetation Index (NDVI) and the edge features to be salient for the water background, we integrated these features into the proposed method. Three sites from north to south in Eastern China were used to validate the method and compare it with our former proposed method using only object-based visually salient NDVI (OBVS-NDVI) features. The new proposed method not only maintains the true positive results of OBVS-NDVI, but also eliminates most false negative results of OBVS-NDVI. Thus, the new proposed method has potential for use in rapid monitoring of coastal raft aquaculture on a large scale.
Highlights
Aquaculture is one of the fastest growing food production sectors worldwide, an important source of food in many countries, the main protein source for hundreds of millions of people, and has been in the spotlight for its potential to support future food security at a global scale [1]
A fast and efficient extraction method that can employ low resolution remote sensing imagery is urgently needed for monitoring coastal raft aquaculture areas, which is of great significance for the rational development of marine resources and protection of the marine environment
The main principle of this method is based on the visual saliency calculation methods proposed by Itti [29] and Sun [30], which allows the highlighting of raft aquaculture areas by Normalized Difference Vegetation Index (NDVI) feature enhancement of segmentation objects
Summary
Aquaculture is one of the fastest growing food production sectors worldwide, an important source of food in many countries, the main protein source for hundreds of millions of people, and has been in the spotlight for its potential to support future food security at a global scale [1]. A fast and efficient extraction method that can employ low resolution remote sensing imagery is urgently needed for monitoring coastal raft aquaculture areas, which is of great significance for the rational development of marine resources and protection of the marine environment. The main principle of this method is based on the visual saliency calculation methods proposed by Itti [29] and Sun [30], which allows the highlighting of raft aquaculture areas by NDVI feature enhancement of segmentation objects These methods still have a relatively low level of accuracy when only considering spectral information in some spectrally similar non-aquaculture zones, such as coastal shoals and so on.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
