Image Fusion

Mapping millions of buried landmines rapidly and removing them cost-effectively is supremely important to avoid their potential risks and ease this labour-intensive task. Deploying uninhabited vehicles equipped with multiple remote sensing modalities seems to be an ideal option for performing this task in a non-invasive fashion. This report provides researchers with vision-based remote sensing imagery datasets obtained from a real landmine field in Croatia that incorporated an autonomous uninhabited aerial vehicle (UAV), the so-called LMUAV.


Blade damage inspection without stopping the normal operation of wind turbines has significant economic value. This study proposes an AI-based method AQUADA-Seg to segment the images of blades from complex backgrounds by fusing optical and thermal videos taken from normal operating wind turbines. The method follows an encoder-decoder architecture and uses both optical and thermal videos to overcome the challenges associated with field application.


The dataset contains thermal and visible images of volunteers taken in different places to develop face detection algorithms. By providing both thermal and visible face images in a single dataset, our dataset empowers researchers, scientists, and developers to leverage the strengths of each image type. The dataset can be utilized for tasks like biometric authentication, emotion recognition, facial expression analysis, age estimation, and gender classification.


This dataset has 32,000 remote sensing images in UAV scenes of tiny objects with labels.


The Udacity Autonomous Vehicle Dataset is a widely used dataset that contains a large number of images and corresponding steering angle information. In IA-Udacity, we added lane marking annotation information to further improve the accuracy and reliability of the model, making it more suitable for lane detection and steering decisions in autonomous driving scenarios.


Buildings are essential components of urban areas. While research on the extraction and 3D reconstruction of buildings is widely conducted, information on fine-grained roof types of buildings is usually ignored. This limits the potential of further analysis, e.g., in the context of urban planning applications. The fine-grained classification of building roof type from satellite images is a highly challenging task due to ambiguous visual features within the satellite imagery.

Last Updated On: 
Tue, 03/07/2023 - 11:58

In order to contribute to the development of automatic methods for the detection of bacilli, TBimages is an image dataset composed of two subsets: TbImages_SS1 contains 10 images per field, of different focal depths, and aims to support the definition of autofocus metrics and also the development of extended focus imaging methods to facilitate the detection of bacilli in smear microscopy imaging.   TbImages_SS2  aims to support the development of automatic bacilli detection.


The Paddy Doctor dataset contains 16,225 labeled paddy leaf images across 13 classes (12 different paddy diseases and healthy leaves). It is the largest expert-annotated visual image dataset to experiment with and benchmark computer vision algorithms. The paddy leaf images were collected from real paddy fields using a high-resolution (1,080 x 1,440 pixels) smartphone camera. The collected images were carefully cleaned and annotated with the help of an agronomist.


Drone based wildfire detection and modeling methods enable high-precision, real-time fire monitoring that is not provided by traditional remote fire monitoring systems, such as satellite imaging. Precise, real-time information enables rapid, effective wildfire intervention and management strategies. Drone systems’ ease of deployment, omnidirectional maneuverability, and robust sensing capabilities make them effective tools for early wildfire detection and evaluation, particularly so in environments that are inconvenient for humans and/or terrestrial vehicles.


A Chinese character gesture dataset for 8 Chinese characters ( “国”,“图”,“木”,“工”,“口”,“中”,“国”,“人” ) .Datasets were collected in three different environments.