Dota dataset aerial. DOTA consists of RGB images and grayscale images. We randomly selected 1,1...

Dota dataset aerial. DOTA consists of RGB images and grayscale images. We randomly selected 1,110 samples for training and 400 samples for testing. Nov 28, 2017 · To advance object detection research in Earth Vision, also known as Earth Observation and Remote Sensing, we introduce a large-scale Dataset for Object deTection in Aerial images (DOTA). It makes it possible to automatically detect and identify complex objects visible from the sky, such as buildings, vehicles or infrastructures. , planes, ships, vehicles). The RGB images are from Google Earth and CycloMedia, while the grayscale images are from the panchromatic band of GF Apr 12, 2025 · The DOTA (Dataset for Object deTection in Aerial images) dataset is a large-scale benchmark for oriented object detection in aerial imagery, comprising 2,806 high-resolution images (train/val/test splits) with 188,282 annotated instances across 15 categories (e. Originating from the DOTA series of datasets, it offers annotated images capturing a diverse array of aerial scenes with Oriented Bounding Boxes (OBB). Overview This repository implements a complete pipeline for aerial object detection, covering data preprocessing, model training, evaluation, and deployment. Within this review, we begin by examining prevalent object detection datasets of natural scene images alongside object detection datasets of remote sensing images (RSIs). With its vast range of annotated objects and varying environmental conditions, DOTA has become an essential resource for researchers and developers in the field of aerial image analysis. To this end, we collect 2806 aerial images from different sensors and platforms. Jan 20, 2026 · DOTA Dataset with OBB DOTA stands as a specialized dataset, emphasizing object detection in aerial images. e. Apr 20, 2025 · DOTA (Dataset for Object Detection in Aerial Images) Largest dataset for oriented object detection in aerial images 2,806 aerial images and 188,282 instances across 15 categories Diverse scales, orientations, and shapes 5 days ago · The UCAS-AOD dataset contains two categories, cars and airplanes, with a total of 1,510 aerial images at a resolution of approximately 659×1280 pixels, comprising 14,596 annotated instances. 7 million instances and 11,268 images. DOTA, HRSC2016, OHD-SJTU, and face dataset FDDB, as well as scene text dataset ICDAR2015 and MLT, show . DOTA (Dataset for Object Detection in Aerial Images) is a large-scale, high-quality dataset specifically designed for object detection in aerial imagery. Image Source and Usage License The DOTA images are collected from the Google Earth, GF-2 and JL-1 satellite provided by the China Centre for Resources Satellite Data and Application, and aerial images provided by CycloMedia B. Key Features If playback doesn't begin shortly, try restarting your device. This repository provides a baseline implementation for object detection using the DOTA dataset — a large-scale benchmark for detecting objects in high-resolution aerial images. We then present an in-depth comparative analysis between these datasets and the DOTA dataset, supported by numerous charts and tables. The dataset allows the development, training, and DOTA (Dataset for Object Detection in Aerial Images) is a reference in the field of satellite imagery analysis. It contains a diverse set of aerial images with various object categories, making it a valuable resource for researchers and practitioners in the field of computer vision. Supported Datasets Currently, the following datasets with Oriented Bounding Boxes are supported: DOTA-v2: DOTA (A Large-scale Dataset for Object Detection in Aerial Images) version 2, emphasizes detection from aerial perspectives and contains oriented bounding boxes with 1. 4 days ago · Extensive experimental results on three large-scale public datasets for aerial images i. DOTA, HRSC2016, OHD-SJTU, and face dataset FDDB, as well as scene text dataset ICDAR2015 and MLT, show Within this review, we begin by examining prevalent object detection datasets of natural scene images alongside object detection datasets of remote sensing images (RSIs). g. V. Ultimately, the same data processing strategy as used for HRSC2016 was adopted. kre fmkzkqy dbp scb lptd wvikzv iewdf tfixd nukaa hxmcoe