SAIOD Sikkim Aerial Images dataset for Object Detection

The SAIOD dataset is an aerial image dataset prepared by collecting images from Sikkim state located in the Himalayas, India, using a DJI Mavic Air drone. The terrain is diverse, including undulating surfaces, slopes, gorges, rivers, forests, and settlements. Ten classes of objects were identified for image classification, including buildings, cars, debris, footpaths, metalled roads, open fields, shadows, tanks, trees, and roofs.

搜集汇总

背景与挑战

背景概述

SAIOD数据集是一个用于对象检测的航空图像数据集，采集自印度喜马拉雅山脉锡金州的多样地形，包括起伏地表和自然及人造特征。数据集包含10个预定义对象类别，如建筑物、汽车和道路，适用于航空图像分析和目标识别任务。