BD-Freshwater-Fish: An Image Dataset from Bangladesh for AI-Powered Automatic Fish Species Classification and Detection toward Smart Aquaculture

1) BD-Freshwater-Fish: an image dataset contain 4389 images of 12 different species. The Species are- (i) Rohu (ii) Katla (iii) Mrigal (iv) Grass Carp (v) Common Carp (vi) Mirror Carp (vii) Black Rohu (viii) Silver Carp (ix) Striped Catfish (x) Nile Tilapia (xi) Long-whiskered Catfish (xii) Freshwater Shark 2) Images are captured from local fish market of Sylhet and Jessore district of Bangladesh using HD mobile camera. 3) This dataset can be used in various computer vision and deep learning based models for automatic identification and detection of fish species. 4) This image based dataset can be used by fish biologist to teach ichthyology students about fish species, also.1) 孟加拉国淡水鱼类数据集（BD-Freshwater-Fish）：该图像数据集包含12个不同鱼种共计4389张图像，所涵盖的鱼种依次为： (i) 罗氏野鲮 (Rohu) (ii) 卡特拉鲃 (Katla) (iii) 麦瑞加拉鲮 (Mrigal) (iv) 草鱼 (Grass Carp) (v) 鲤 (Common Carp) (vi) 镜鲤 (Mirror Carp) (vii) 黑罗氏野鲮 (Black Rohu) (viii) 鲢鱼 (Silver Carp) (ix) 条纹鲶鱼 (Striped Catfish) (x) 尼罗罗非鱼 (Nile Tilapia) (xi) 长须鲶 (Long-whiskered Catfish) (xii) 淡水鲨 (Freshwater Shark) 2) 该数据集的图像均采用高清手机摄像头，在孟加拉国锡尔赫特县与杰索尔县的当地鱼类市场采集获取。 3) 本数据集可应用于各类基于计算机视觉与深度学习的模型，用于实现鱼种的自动识别与检测任务。 4) 此外，该图像数据集还可供鱼类生物学家向鱼类学专业学生讲授鱼种相关知识。