Apparatus and method for estimating octopus resources in tidal flat fishery based on artificial neural network
An apparatus for estimating octopus resources in a tidal flat fishery according to the present invention includes: an identification part that, when a tidal flat image vector derived from a tidal flat image obtained by capturing a tidal flat is input, uses an identification model, which is an artifi...
Gespeichert in:
Hauptverfasser: | , , , , , , , , |
---|---|
Format: | Patent |
Sprache: | eng ; kor |
Schlagworte: | |
Online-Zugang: | Volltext bestellen |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | An apparatus for estimating octopus resources in a tidal flat fishery according to the present invention includes: an identification part that, when a tidal flat image vector derived from a tidal flat image obtained by capturing a tidal flat is input, uses an identification model, which is an artificial neural network with learned weights, to derive one or more area boxes that specify the area where a buoy exists from the tidal flat image vector and derives an output image vector including the derived area box; and an estimation part that estimates octopus resources based on the output image vector.
본 발명의 갯벌어장의 낙지 자원량을 산정하기 위한 장치는 갯벌을 촬영한 갯벌 영상으로부터 도출된 갯벌영상벡터가 입력되면, 학습된 가중치를 가지는 인공신경망인 식별모델을 이용하여 갯벌영상벡터로부터 부럿이 존재하는 영역을 특정하는 하나 이상의 영역상자를 도출하고, 도출된 영역상자를 포함하는 출력영상벡터를 도출하는 식별부와, 상기 출력영상벡터를 기초로 낙지 자원량을 추정하는 추정부를 포함한다. |
---|