Open dataset of RGB and thermal images of roe deer in grasslands
Identifikátory výsledku
Kód výsledku v IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F06632581%3A_____%2F25%3AN0000004" target="_blank" >RIV/06632581:_____/25:N0000004 - isvavai.cz</a>
Výsledek na webu
<a href="https://fly4future.com/TQ16000038-v7" target="_blank" >https://fly4future.com/TQ16000038-v7</a>
DOI - Digital Object Identifier
—
Alternativní jazyky
Jazyk výsledku
angličtina
Název v původním jazyce
Open dataset of RGB and thermal images of roe deer in grasslands
Popis výsledku v původním jazyce
The output V7 of the TQ16000038 project is an open dataset of image data capturing roe deer and persons involved in their rescue in permanent grasslands (meadows), acquired in both RGB and thermal spectra. The dataset was intentionally created as a supporting resource for research and development of automated roe deer detection technologies and is primarily intended for the training, validation, and testing of machine learning and neural network models within the development of a software solution. The dataset comprises images captured at various times of day and under diverse climatic conditions, thereby encompassing a broad range of real-world operational scenarios and enhancing the robustness of the developed detection algorithms. An integral part of the dataset consists of detailed annotations localizing roe deer and other relevant objects within the images, provided in the form of bounding boxes and pixel masks. The dataset is structured into two separate subsets, categorized by the type of sensor used (RGB and thermal data), and contains a total of 58,300 images.
Název v anglickém jazyce
Open dataset of RGB and thermal images of roe deer in grasslands
Popis výsledku anglicky
The output V7 of the TQ16000038 project is an open dataset of image data capturing roe deer and persons involved in their rescue in permanent grasslands (meadows), acquired in both RGB and thermal spectra. The dataset was intentionally created as a supporting resource for research and development of automated roe deer detection technologies and is primarily intended for the training, validation, and testing of machine learning and neural network models within the development of a software solution. The dataset comprises images captured at various times of day and under diverse climatic conditions, thereby encompassing a broad range of real-world operational scenarios and enhancing the robustness of the developed detection algorithms. An integral part of the dataset consists of detailed annotations localizing roe deer and other relevant objects within the images, provided in the form of bounding boxes and pixel masks. The dataset is structured into two separate subsets, categorized by the type of sensor used (RGB and thermal data), and contains a total of 58,300 images.
Klasifikace
Druh
O - Ostatní výsledky
CEP obor
—
OECD FORD obor
20204 - Robotics and automatic control
Návaznosti výsledku
Projekt
<a href="/cs/project/TQ16000038" target="_blank" >TQ16000038: Udržitelné zemědělství: Energeticky výkonné bezpilotní technologie pro precizní zemědělství</a><br>
Návaznosti
P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)
Ostatní
Rok uplatnění
2025
Kód důvěrnosti údajů
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů