Oreochromis niloticus Fingerlings WhiteTray Dataset (50 Fish per Image Polygon annotation)

The Oreochromis niloticus Fingerlings WhiteTray Dataset is a curated repository of annotated images designed for the development and benchmarking of Artificial Intelligence models in precision aquaculture, particularly for fish detection and counting tasks.

The dataset contains images of Nile tilapia (Oreochromis niloticus) fingerlings captured in a controlled environment using white trays (white tray setup), which enhances visual contrast and facilitates object detection and segmentation processes.

Each image contains exactly 50 fish, ensuring a fixed and controlled density across the entire dataset, annotation method Polygon (Segments). This characteristic makes the dataset especially suitable for evaluating counting algorithms, density estimation methods, and object detection models under standardized conditions.

The images were obtained under consistent lighting and background conditions, minimizing environmental variability while still preserving common computer vision challenges such as:

Variations in orientation and positioning

Subtle scale differences among fingerlings

Visual similarity between instances (low inter-class variance)

All images are annotated with bounding boxes in YOLO format, supporting direct use in state-of-the-art object detection frameworks.

This dataset is highly recommended for training and evaluating deep learning architectures, including the YOLO family (v8 to v11) and Faster R-CNN, particularly in scenarios requiring precise counting and detection under controlled densities.

Folder structure:

ON_Fingerlings_WhiteTray_50_polygon_fish.zip/
├── data.yaml
├── train/
│ ├── images/ (160 jpg files)
│ └── labels/ (160 txt files)
├── valid/
│ ├── images/ (20 jpg files)
│ └── labels/ (20 txt files)
└── test/
├── images/ (20 jpg files)
└── labels/ (20 txt files)

尼罗罗非鱼幼鱼白盘数据集（Oreochromis niloticus Fingerlings WhiteTray Dataset）是一套经精心整理的标注图像库，旨在为精准水产养殖领域的人工智能模型开发与基准测试提供支撑，尤其适用于鱼类检测与计数任务。

该数据集包含在受控环境下采用白色托盘（白盘配置）拍摄的尼罗罗非鱼（Oreochromis niloticus）幼鱼图像，该配置可有效提升视觉对比度，助力目标检测与分割任务的开展。

每张图像恰好包含50尾幼鱼，确保整个数据集内的养殖密度固定且可控，所有标注均采用多边形（Polygon）分割（Segments）标注方式。这一特性使得该数据集特别适合在标准化条件下评估计数算法、密度估计方法以及目标检测模型。

所有图像均在一致的光照与背景条件下采集，尽可能降低环境变量带来的干扰，同时仍保留了计算机视觉领域常见的典型挑战，包括：幼鱼的朝向与位置差异、个体间细微的尺度差异、实例间视觉相似度较高（类间方差较低）。

所有图像均附带边界框（bounding boxes）的YOLO格式标注，可直接兼容当前主流的目标检测框架。

该数据集强烈推荐用于训练与评估深度学习架构，包括YOLO系列（v8至v11）以及Faster R-CNN，尤其适用于需要在可控密度下实现精准计数与检测的场景。

数据集文件夹结构如下：
ON_Fingerlings_WhiteTray_50_polygon_fish.zip/
├── data.yaml
├── train/
│ ├── images/（含160个jpg文件）
│ └── labels/（含160个txt文件）
├── valid/
│ ├── images/（含20个jpg文件）
│ └── labels/（含20个txt文件）
└── test/
├── images/（含20个jpg文件）
└── labels/（含20个txt文件）