A combination of traditional fish sampling methods(hand-line and gill net) and modern hydroacoustic techniques were used to study fish community structure,distribution,and diel movements of fish on Zhubi Reef to enhan...A combination of traditional fish sampling methods(hand-line and gill net) and modern hydroacoustic techniques were used to study fish community structure,distribution,and diel movements of fish on Zhubi Reef to enhance understanding of the ecosystem.We collected 126 individuals from 29 species,20 genera,17 families,three orders,and two classes using traditional gear.Perciforms were the dominant group in terms of species richness and biomass.The acoustic data indicated that very small(target strength[TS],dB) <-60 dB) and small(-60 dB≤TS<-45 dB) fish contributed the most to abundance and species richness on the coral reef,and that the proportion of medium-sized(-45 dB≤TS<-35 dB) and largesized(-35 dB≤TS) fish increased gradually as depth increased.The single-target detection method revealed two distinct size classes during the day in the 12-16 and 16-20-m layers.One group consisted of very smallsized fish(TS<-60 dB) and the other consisted of medium and large-sized fish(TS>-55 dB).The number of single-target detections was significantly higher during the night than during the day(P<0.05).The singletarget TS frequency distribution during the day was significantly different than during the night at depths of4-8,8-12,12-16,and 16-20 m.Significant differences were observed among the 4-8,8-12,12-16,and16-20-m-depth layers during day and night.Diel vertical movement was evidenced as fish began to spread and move upward just before sunset and began to assemble and descend shortly(15 min) after sunrise.展开更多
为提高鱼类行为学数据的提取效率,实验提出了一种基于单摄像机的鱼类三维观测方法,将防水镜面安装在实验用鱼缸上方,模拟一台由上向下拍摄的摄像机,实现了单摄像机三维成像。同时运用多目标跟踪的IMMJPDA(interacting multiple model jo...为提高鱼类行为学数据的提取效率,实验提出了一种基于单摄像机的鱼类三维观测方法,将防水镜面安装在实验用鱼缸上方,模拟一台由上向下拍摄的摄像机,实现了单摄像机三维成像。同时运用多目标跟踪的IMMJPDA(interacting multiple model joint probabilistic data association)算法,实现了鱼类运动的三维实时自动跟踪,并通过摄像机倾斜矫正和摄像机标定提高了测量精度。通过对6条红鼻剪刀鱼的跟踪,实验结果显示:本方法可正确区分、提取和跟踪鱼群个体以及它们的镜像,自动输出鱼的三维坐标、实时速度、方向等参数,并生成完整的鱼类行为三维轨迹图。展开更多
基金Supported by the National Key Technology Research and Development Program(No.2013BAD13B06)the Financial Project of the Ministry of Agriculture(No.NFZX2013)+1 种基金the Special Scientific Research Funds for Central Non-profit Institutes,Chinese Academy of Fishery Sciences(Nos.2014TS18,2013ZD03)the National Agricultural Finance Project "South China Sea Fishing Information Dynamic Acquisition"
文摘A combination of traditional fish sampling methods(hand-line and gill net) and modern hydroacoustic techniques were used to study fish community structure,distribution,and diel movements of fish on Zhubi Reef to enhance understanding of the ecosystem.We collected 126 individuals from 29 species,20 genera,17 families,three orders,and two classes using traditional gear.Perciforms were the dominant group in terms of species richness and biomass.The acoustic data indicated that very small(target strength[TS],dB) <-60 dB) and small(-60 dB≤TS<-45 dB) fish contributed the most to abundance and species richness on the coral reef,and that the proportion of medium-sized(-45 dB≤TS<-35 dB) and largesized(-35 dB≤TS) fish increased gradually as depth increased.The single-target detection method revealed two distinct size classes during the day in the 12-16 and 16-20-m layers.One group consisted of very smallsized fish(TS<-60 dB) and the other consisted of medium and large-sized fish(TS>-55 dB).The number of single-target detections was significantly higher during the night than during the day(P<0.05).The singletarget TS frequency distribution during the day was significantly different than during the night at depths of4-8,8-12,12-16,and 16-20 m.Significant differences were observed among the 4-8,8-12,12-16,and16-20-m-depth layers during day and night.Diel vertical movement was evidenced as fish began to spread and move upward just before sunset and began to assemble and descend shortly(15 min) after sunrise.
文摘为提高鱼类行为学数据的提取效率,实验提出了一种基于单摄像机的鱼类三维观测方法,将防水镜面安装在实验用鱼缸上方,模拟一台由上向下拍摄的摄像机,实现了单摄像机三维成像。同时运用多目标跟踪的IMMJPDA(interacting multiple model joint probabilistic data association)算法,实现了鱼类运动的三维实时自动跟踪,并通过摄像机倾斜矫正和摄像机标定提高了测量精度。通过对6条红鼻剪刀鱼的跟踪,实验结果显示:本方法可正确区分、提取和跟踪鱼群个体以及它们的镜像,自动输出鱼的三维坐标、实时速度、方向等参数,并生成完整的鱼类行为三维轨迹图。
