长江江豚声影像监测自动识别分析数据集

通过在目标监测水域选择合适位置布放水声记录仪对水下声音进行连续采集并提取长江江豚声信号进行分析处理，同时联合岸基摄像头拍摄长江江豚出水视频并进行在线自动智能识别分析，最后将识别分析的结果实时回传至监测平台进行直观展示，该数据集未涉及内部数据采集。

This dataset is established via the following procedures: firstly, deploy hydroacoustic recorders at appropriate positions within the target monitoring waters to continuously collect underwater acoustic data, extract the acoustic signals of Yangtze finless porpoises, and conduct analysis and processing on the extracted signals; secondly, collaborate with shore-based cameras to capture videos of surfacing Yangtze finless porpoises and perform online automatic intelligent recognition and analysis on these videos; finally, transmit the results of recognition and analysis back to the monitoring platform in real time for intuitive visualization. This dataset does not involve internal data collection.

搜集汇总

背景与挑战

背景概述

该数据集通过水声记录仪和岸基摄像头采集长江江豚的声信号和视频影像，利用YOLOV7算法进行自动识别分析，包含2000万条数据，为长江江豚研究和保护提供重要数据支持。