Fishing ground forecast method through adoption of multi-scale environment characteristics

Abstract

The present invention discloses fishing ground forecast method through adoption of multi-scale environment characteristics. The method comprises the following steps: 1) performing preprocessing of original marine environment remote sensing data of a target region; 2) constructing a multi-scale environment characteristic data set base on the pre-processing marine environment remote sensing data; 3) calculating the catch per unit effort (CPUE) according to the fishery production data, setting a threshold according to the CPUE, and generating fishing ground sample labels; 4) retrieving the marine environment remote sensing data approaching or satisfying the threshold in the multi-scale environment characteristic data set, and building a multiple temporal-spatial scales fishing ground marine environment characteristic sample set; and 5) performing fishing ground forecast according to the multiple temporal-spatial scales fishing ground marine environment characteristic sample set, and obtaining a fishing ground probability forecast result. The fishing ground forecast method through adoption of multi-scale environment characteristics builds an effective mapping form a multi-dimensional marine environment characteristic space to a marine fishing ground characteristic space so as to allow the mapping relation to more steadily and comprehensively reflect the marine fishing ground distribution rule and improve the fishing ground forecast accuracy.

Description

technical field [0001] The invention relates to the field of marine fishery forecasting, in particular to a fishery forecasting method utilizing multi-scale environmental features. Background technique [0002] Providing marine environment information such as the abundance distribution of marine fishery resources and marine hydrometeorology to fishermen or fishing fleets engaged in marine fishing production can effectively reduce the time for fishing at sea, save fuel and reduce labor costs, and improve fishing efficiency. Economic value and social benefits. Existing technologies focus on the entire fishing area as an object, using single-resolution marine environmental parameters and fishery production data with a fixed grid size, and performing marine environment analysis and fishery forecasting through traditional statistical learning models. [0003] The distribution and changes of marine fisheries are closely related to various marine environmental factors at different...

AI Technical Summary

Problems solved by technology

The mapping model established by using a few marine environmental parameters with a fixed grid size and a single temporal-spatial resolution can only reflect a partial spatial subset of the distribution characteristics of marine fisheries, and cannot achieve effective forecasting for the characteristic spatial areas of fisheries that are not covered.

Images