FGSCM-52

该数据集在扩展现有数据集的基础上，大幅增加了更广泛的数据和更详细的注释，特别是对于不太常见的船舶类别。它不仅增强了模型对各类船舶的泛化能力，还提高了模型在辨识复杂背景和学习船舶判别特征方面的性能。该数据集的任务是进行细粒度的船舶分类。

This dataset, developed by expanding existing datasets, substantially incorporates more comprehensive data and more detailed annotations, especially for less common ship categories. It not only enhances the model's generalization ability across various ship types, but also improves the model's performance in identifying complex backgrounds and learning discriminative features of ships. The task of this dataset is fine-grained ship classification.

搜集汇总

背景与挑战

背景概述

FGSCM-52是一个包含52个类别的细粒度船舶分类数据集，图像尺寸范围广泛，并配有层次化标签，适用于复杂的船舶分类任务。