Self-Powered Fish Physiological Tags and Machine Vision for the Application of the Intelligent Net Cage

View Statistics Email Alert RSS Feed

Information

Details

Project title

Code/計畫編號

MOST109-2221-E019-034

Translated Name/計畫中文名

結合魚體自供電生理感測標籤與機器視覺於海上箱網智慧養殖應用

Project Coordinator/計畫主持人

Shih-Hao Huang

Funding Organization/主管機關

National Science and Technology Council

Department/Unit

Department of Mechanical and Mechatronic Engineering

Website

https://www.grb.gov.tw/search/planDetail?id=13535507

Year

2020

Start date/計畫起

01-08-2020

Expected Completion/計畫迄

31-07-2021

Bugetid/研究經費

916千元

ResearchField/研究領域

機械工程

近年來智慧養殖風氣盛行，在海上箱網自動化遠端檢測方面，主要著重在水質與養殖魚體大小和行為的監測，然而，可遠端即時監控魚體健康與異常警示，減少養殖業者因魚類疾病或寒(熱)害造成的損失，卻是產學研仍未進行相關的研發。魚體內的血糖、膽固醇、和乳酸濃度變化是生理異常的重要指標，血糖高低與的營養狀態有關、疾病感染的魚有低於正常膽固醇濃度、乳酸濃度是疲勞的指標。本研究以二年期進行結合魚體自供電生理感測標籤與機器視覺應用於海上箱網智慧養殖，達到遠端即時監控魚體健康的目的，預計開發基於酵素燃料電池原理之探針式魚體自供電生理感測標籤，此生理感測標籤內含三根修飾不同催化酵素的金屬絲作為陽極，以共陰極方式連接於酵素陰極，此生理感測標籤由外部植入魚體內，分別催化葡萄糖、乳酸、膽固醇產生電力，並外接升壓電荷幫浦電路進行充放電過程，分別驅動三顆紅綠藍光LED燈閃爍，閃爍的頻率正比於魚體內葡萄糖(紅光)、膽固醇(綠光)、乳酸(藍光)的濃度，達到自體供電多項魚類生理指標感測，此外，產生的電力額外驅動預先設定固定閃爍頻的黃色LED燈作為辨識用途。在海上箱網養殖的應用上，生理感測標籤可分別植入於魚群中選定的多隻標定魚魚體內，不同的標定魚具有不同的預先設定固定閃爍頻率的黃色LED燈作為辨識，藉由無線水下攝影機拍攝影像，經由機器視覺辨識技術進行標定魚辨識與個別魚體上三顆紅綠藍光LED燈閃爍偵測，同步得知多隻魚體的生理指標，達到遠距即時監控魚體健康。第二年將在大型魚缸模擬寒(熱)害環境與進行病原菌感染過程，利用魚體自供電生理感測標籤與機器視覺，遠端監控測試魚血糖/總膽固醇/乳酸變化之性能測試，也將實際應用於台灣海洋大學貢寮箱網實驗場域，以重要經濟海水養殖魚種海驪作為測試魚種，實際進行外海遠端即時監控魚體健康。本計畫提出的魚體自供電生理感測標籤，不需要電力來源、低成本、可大量製作，達到同步即時偵測魚體內的血糖、膽固醇、和乳酸濃度變化，藉由機器視覺辨識與偵測技術，同時監測多隻標定魚的健康與異常警示，此一構想迄今尚未有相關文獻提出類似之論述，敬請審查委員大力支持。 Taiwan has been developing the net cage for more than 30 years. The remote detection in the net cage focuses on the monitoring of water quality and size and behavior of farmed fish. However, the researches in the physical health and abnormal warnings of farmed fish, which could reduce the losses caused by fish diseases or cold (heat) in aquaculture, have not been carried out. The changes in blood sugar, cholesterol, and lactic acid concentrations in fish are important indicators of physiological abnormalities. The level of blood sugar is related to the nutritional status. The fish infected with diseases have lower than normal cholesterol concentration. Lactic acid concentration is an indicator of fatigue. In this two-year study, we will develop self-powered fish physiological tags and machine vision for the intelligent net cage to achieve the purpose of remote monitoring of fish health. The self-powered fish physiological tag was based on the principle of the enzymatic fuel cell which contains three wires modified with different catalytic enzymes as anodes by a common enzymatic cathode. The self-powered fish physiological tag is externally implanted into the fish to respectively catalyze glucose, lactic acid and cholesterol to produce electricity, and connected to an external boosting charge pump circuit for driving three red, green and blue LED lights to blink. The blinking frequency is proportional to the concentrations of glucose (red light), lactic acid (green light), and cholesterol (blue light) in fish. In addition, the generated electric power additionally drives a yellow LED lamp with a preset fixed blinking frequency for identification purposes. In the application, the self-powered fish physiological tag can be respectively implanted into a plurality of selected fish from the fish group. The wireless underwater camera captures the image and wirelessly transmits it to the remote data processing center. The machine vision technology is used to identify the fish and detect the three red, green and blue LED lights on the individual fish body to know the physiological indexes of the multiple fish. In the second year, we will practically apply the self-powered fish physiological tag for the important economic fish of cobia as test fish species at the experimental site of the Taiwan Ocean University. The self-powered fish physiological tag t does not require electricity source, low cost, and can be mass-produced to achieve simultaneous detection of blood sugar, cholesterol, and lactic acid concentration changes in fish, by machine vision identification and detection. This concept has not yet been published in the relevant literature, and the research results fully have the value of academic publications.

Keyword(s)

自供電
酵素燃料電池
生理標籤
機器視覺
智慧箱網
self-powered
enzymatic fuel cell
physiological tag
machine vision
intelligent net cage

DSpace CRIS

Self-Powered Fish Physiological Tags and Machine Vision for the Application of the Intelligent Net Cage

Details

Description