ALL4LAND

We use artificial intelligence to enhance low-resolution images, such as satellite imagery or historical aerial images, that are difficult to interpret or analyze, transforming them into high-resolution images.
We possess AI models that can improve existing images by 2x to 4x in resolution.

National and local governments often compare past and present images visually for tasks such as urban planning, revenue generation, and illegal construction monitoring. This process is time-consuming and costly.
To address this, we have AI models capable of instantly detecting areas of change between past and present images.

In many fields such as administration, transportation, and geospatial information, workers manually create numerical maps or thematic maps from aerial or drone images.
We possess AI models that can automatically extract such data.
The automatic extraction targets include buildings, vehicles, people, streetlights, clouds, and more, with plans for continuous expansion.

When constructing realistic imagery or digital twins, unnecessary information is often included, such as vehicles or street trees, which are not needed for high-quality information construction. We possess AI models that can instantly remove such unnecessary information from images.

• AI-based detection of Hanzi (Chinese characters) areas in old land registry documents
• Supports natural language processing for detected Hanzi areas in both Chinese characters and Hangul
• Continuous performance improvement through automatic re-learning
• Application of AI-based Hangul digital conversion for old land (forest) registry documents

• Segmentation AI: An AI that detects objects such as buildings, change detection, vehicles, people, streetlights, clouds, light reflections, fish farms, and marine debris in aerial images (including drone photos).
• Inpainting AI: An AI that erases user-defined areas from the results of Segmentation AI inference or from images.
• SuperResolution AI: An AI that performs super-resolution on images of aerial photos, drone images, and digital twin objects to enhance their resolution.

By learning from various geospatial data and satellite big data databases, we use AI to detect forest objects.

An AI trained on orthophotos automatically detects crop types and predicts crop yields based on geospatial information.
The AI trained on forests automatically detects forest species and environmental changes to calculate carbon sequestration.

We detect models such as buildings, vehicles, and people in aerial and drone imagery and monitor changes in individual models.
We build GIS databases and use them as foundational data for digital twins across various fields.