Voxel51/BarkVN-50

BarkVN-50是一个全面的树皮纹理分类数据集，包含50种不同树种的5,678张图像，每张图像的分辨率为303×404像素。该数据集专为使用计算机视觉和深度学习技术进行树种自动识别而设计。树皮纹理为树种识别提供了可靠的特征，因为它不受季节变化的影响，这使得该数据集对于包括森林保护、疾病诊断和植物生产在内的全年林业应用特别有价值。数据集通常分为训练集（80%的样本）和测试集（20%的样本），并已使用K-Fold交叉验证（k=5）进行稳健的性能评估。

BarkVN-50 is a comprehensive tree bark texture classification dataset containing 5,678 images of 303×404 pixels across 50 different tree species. This dataset is specifically designed for automated tree species identification using computer vision and deep learning techniques. Bark texture provides a reliable characteristic for tree identification as it remains present despite seasonal variations, making the dataset particularly valuable for year-round forestry applications including forest conservation, disease diagnosis, and plant production. The dataset is typically split into a training set (80% of samples) and a testing set (20% of samples), with K-Fold Cross Validation (k=5) used for robust performance evaluation.