---
title: Building Segementation In An Image Of 1.225 Giga Pixels
description: Explore how DEEP BLOCK can analyze high-resolution aerial images. Check out its amazing performance in segmenting buildings from 1.225 gigapixel images. Dive into the world of remote sensing AI and machine learning.
image: https://www.deepblock.net/hubfs/Building%20Segementation%20in%20an%20image%20of%201.225%20Giga%20Pixels.png
---

[![deepblock_logo_full2](https://www.deepblock.net/hubfs/deepblock_logo_full2.png)](https://www.deepblock.net?hsLang=en)

![menu](https://f.hubspotusercontent40.net/hubfs/7479797/menu.png)

![close_24px](https://7479797.fs1.hubspotusercontent-na1.net/hubfs/7479797/close_24px.svg)

- [Product](https://www.deepblock.net/product)
  
  Show submenu for Product 
  
    - [Manual](https://www.deepblock.net/product/manual)
    - [How To Use](https://www.deepblock.net/product/how-to-use-deep-block)
    - [Pricing](https://www.deepblock.net/product/pricing)
- [Remote Sensing](https://www.deepblock.net/remote-sensing)
  
  Show submenu for Remote Sensing 
  
    - [Surveillance & IMINT](https://www.deepblock.net/remote-sensing/defense-surveillance)
    - [Precision Agriculture](https://www.deepblock.net/remote-sensing/precision-agriculture)
- [Microscopy](https://www.deepblock.net/microscopy)
  
  Show submenu for Microscopy 
  
    - [Semiconductor & MEMS](https://www.deepblock.net/microscopy/semiconductor)
    - [Life Science & Medicine](https://www.deepblock.net/microscopy/life-science-and-medicine)
- [Resources](https://www.deepblock.net/resources)
  
  Show submenu for Resources 
  
    - [Case Studies](https://www.deepblock.net/resources/case-studies)
    - [Computer Vision for High-Resolution](https://www.deepblock.net/resources/challenges-high-resolution-imagery-ml-solution)
    - [Why Deep Block?](https://www.deepblock.net/resources/why-deep-block)
    - [E-Books](https://www.deepblock.net/ebooks)
    - [About Us](https://www.deepblock.net/about-us)
- [Blog](https://www.deepblock.net/blog)

- [Product](https://www.deepblock.net/product)
  
  Show submenu for Product 
  
    - [Manual](https://www.deepblock.net/product/manual)
    - [How To Use](https://www.deepblock.net/product/how-to-use-deep-block)
    - [Pricing](https://www.deepblock.net/product/pricing)
- [Remote Sensing](https://www.deepblock.net/remote-sensing)
  
  Show submenu for Remote Sensing 
  
    - [Surveillance & IMINT](https://www.deepblock.net/remote-sensing/defense-surveillance)
    - [Precision Agriculture](https://www.deepblock.net/remote-sensing/precision-agriculture)
- [Microscopy](https://www.deepblock.net/microscopy)
  
  Show submenu for Microscopy 
  
    - [Semiconductor & MEMS](https://www.deepblock.net/microscopy/semiconductor)
    - [Life Science & Medicine](https://www.deepblock.net/microscopy/life-science-and-medicine)
- [Resources](https://www.deepblock.net/resources)
  
  Show submenu for Resources 
  
    - [Case Studies](https://www.deepblock.net/resources/case-studies)
    - [Computer Vision for High-Resolution](https://www.deepblock.net/resources/challenges-high-resolution-imagery-ml-solution)
    - [Why Deep Block?](https://www.deepblock.net/resources/why-deep-block)
    - [E-Books](https://www.deepblock.net/ebooks)
    - [About Us](https://www.deepblock.net/about-us)
- [Blog](https://www.deepblock.net/blog)

[Contact Us](https://www.deepblock.net/contact?hsLang=en)

[geoAI](https://www.deepblock.net/blog/tag/geoai) [geospatial AI](https://www.deepblock.net/blog/tag/geospatial-ai) [remote sensing AI](https://www.deepblock.net/blog/tag/remote-sensing-ai) [computer vision for remote sensing](https://www.deepblock.net/blog/tag/computer-vision-for-remote-sensing) [high resolution image machine learning](https://www.deepblock.net/blog/tag/high-resolution-image-machine-learning) [high resolution image deep learning](https://www.deepblock.net/blog/tag/high-resolution-image-deep-learning) [satellite image image segmentation](https://www.deepblock.net/blog/tag/satellite-image-image-segmentation) [large image deep learning](https://www.deepblock.net/blog/tag/large-image-deep-learning) [object detection large image](https://www.deepblock.net/blog/tag/object-detection-large-image) [deep learning high resolution](https://www.deepblock.net/blog/tag/deep-learning-high-resolution) [remote sensing computer vision](https://www.deepblock.net/blog/tag/remote-sensing-computer-vision) [image segmentation](https://www.deepblock.net/blog/tag/image-segmentation) [computer vision large image](https://www.deepblock.net/blog/tag/computer-vision-large-image) [object detection satellite image](https://www.deepblock.net/blog/tag/object-detection-satellite-image) [object detection big image](https://www.deepblock.net/blog/tag/object-detection-big-image) [big image object detection](https://www.deepblock.net/blog/tag/big-image-object-detection) [satellite image deep learning](https://www.deepblock.net/blog/tag/satellite-image-deep-learning) [remote sensing deep learning](https://www.deepblock.net/blog/tag/remote-sensing-deep-learning) [deep learning satellite image](https://www.deepblock.net/blog/tag/deep-learning-satellite-image) [satellite image object detection](https://www.deepblock.net/blog/tag/satellite-image-object-detection) [deep learning remote sensing](https://www.deepblock.net/blog/tag/deep-learning-remote-sensing) [satellite image machine learning](https://www.deepblock.net/blog/tag/satellite-image-machine-learning) [big image deep learning](https://www.deepblock.net/blog/tag/big-image-deep-learning) [object detection drone image](https://www.deepblock.net/blog/tag/object-detection-drone-image) [satellite imagery deep learning](https://www.deepblock.net/blog/tag/satellite-imagery-deep-learning) [deep learning satellite imagery](https://www.deepblock.net/blog/tag/deep-learning-satellite-imagery) [satellite imagery machine learning](https://www.deepblock.net/blog/tag/satellite-imagery-machine-learning) [object detection satellite imagery](https://www.deepblock.net/blog/tag/object-detection-satellite-imagery) [machine learning satellite imagery](https://www.deepblock.net/blog/tag/machine-learning-satellite-imagery)

# Building Segementation In An Image Of 1.225 Giga Pixels

[![Gwihwan Moon](https://www.deepblock.net/hs-fs/hubfs/Brand%20Assets/Team%20members/Deep%20Block_Gwihwan%20Moon_Blog.png?width=50&height=50&name=Deep%20Block_Gwihwan%20Moon_Blog.png)](https://www.deepblock.net/blog/author/gwihwan-moon)

[Gwihwan Moon](https://www.deepblock.net/blog/author/gwihwan-moon)  14 June, 2024

![](https://www.deepblock.net/hubfs/Building%20Segementation%20in%20an%20image%20of%201.225%20Giga%20Pixels.png)

<https://twitter.com/intent/tweet?text=I+found+this+interesting+blog+post&url=https://www.deepblock.net/blog/building-segementation> <http://www.facebook.com/share.php?u=https://www.deepblock.net/blog/building-segementation> <http://www.linkedin.com/shareArticle?mini=true&url=https://www.deepblock.net/blog/building-segementation>

In this video, we will build an AI model to extract buildings in aerial photos using DEEP BLOCK, and based on this model, we will check the model's inference results on validation data. (1.225GP image)

 

In this project, we used the aerial image dataset published by INRIA.

The resolution of the training image is 5000x5000 pixels, and we will divide it into 5x5 to train the model.

![](https://www.deepblock.net/hs-fs/hubfs/training%20data-png.png?width=1483&height=640&name=training%20data-png.png)

Click the DIVIDE & TRAIN button.

![](https://www.deepblock.net/hs-fs/hubfs/training-png-1.png?width=1476&height=749&name=training-png-1.png)

After some time, the model will start training and a graph will be displayed showing the training progress.

Training the model on the training data takes approximately 20 minutes.

![](https://www.deepblock.net/hs-fs/hubfs/training%20progress-png.png?width=1446&height=733&name=training%20progress-png.png)

Over time, training is complete.

Now, we will analyze the super-resolution images with the trained model.

Click the PREDICT button and start running the model with the 35x35 division option for our validation image.

![](https://www.deepblock.net/hs-fs/hubfs/predict-png.png?width=1435&height=695&name=predict-png.png)

After a certain period of time, the inference of the model is completed.

![](https://www.deepblock.net/hs-fs/hubfs/result-png.png?width=1479&height=758&name=result-png.png)

We stitched together several validation images to create a 35000x35000 pixel image.

A single image is several gigabytes in size.

There are more than 30,000 buildings in the image.

![](https://www.deepblock.net/hs-fs/hubfs/building-png.png?width=1405&height=680&name=building-png.png)

The remote sensing images we have to deal in the real world are very large, and powerful software technologies are needed to analyze these images.

Get started with DEEP BLOCK, a very easy and seamless remote sensing image analysis AI model development tool now!

Previous Post

###### [TEM vs SEM: A Detailed Comparison](https://www.deepblock.net/blog/tem-vs-sem?hsLang=en)

Next Post

###### [Fixing Invalid Polygon Geometry Errors](https://www.deepblock.net/blog/fixing-invalid-polygon-geometry-errors?hsLang=en)

### Similar Blog Posts

[![](https://www.deepblock.net/hs-fs/hubfs/cytoplasm%20segmentation.png?width=352&name=cytoplasm%20segmentation.png)](https://www.deepblock.net/blog/cytoplasm-segmentation?hsLang=en)

#### [Cytoplasm Segmentation in High-Resolution Microscopy with Deep Block](https://www.deepblock.net/blog/cytoplasm-segmentation?hsLang=en)

[![](https://www.deepblock.net/hs-fs/hubfs/micro%201.png?width=352&name=micro%201.png)](https://www.deepblock.net/blog/difficulties-in-analyzing-high-resolution-micrographs?hsLang=en)

#### [Difficulties in analyzing HIGH-resolution micrographs](https://www.deepblock.net/blog/difficulties-in-analyzing-high-resolution-micrographs?hsLang=en)

[![](https://www.deepblock.net/hs-fs/hubfs/sar%20image.png?width=352&name=sar%20image.png)](https://www.deepblock.net/blog/unveiling-the-power-of-synthetic-aperture-radar-images?hsLang=en)

#### [Unveiling the Power of Synthetic Aperture Radar (SAR) Images](https://www.deepblock.net/blog/unveiling-the-power-of-synthetic-aperture-radar-images?hsLang=en)

[![Deep Block Logo White](https://www.deepblock.net/hubfs/Deep%20Block%20Logo%20White.svg)](https://www.deepblock.net?hsLang=en)

Deep Block is a NO CODE AI-powered platform where you can create and use AI models without any of the frustration of coding. Deep Block allows you to use a library of pre-trained models but also lets anyone train their own AI data sets, no expertise required.

[linkedin icon](https://www.linkedin.com/company/deep-block-net/) [youtube icon](https://www.youtube.com/@deepblock)

###### Get In Touch

![Deep Block_email_icon_B](https://www.deepblock.net/hubfs/Deep%20Block_email_icon_B.svg)

[Contact us](https://www.deepblock.net/contact?hsLang=en)

![Deep Block_address_icon_B](https://www.deepblock.net/hubfs/Deep%20Block_address_icon_B.svg)

43, Changeop-ro, Seongnam City, Republic of Korea, 13449

Copyright © 2025 OMNIS Labs Company All Rights Reserved.

[Privacy Policy |](https://www.deepblock.net/terms-and-privacy-policy?hsLang=en) [Terms of Use](https://www.deepblock.net/terms-and-privacy-policy?hsLang=en)

```json
{
  "@context" : "https://schema.org",
  "@type" : "BlogPosting",
  "author" : {
    "@type" : "Person",
    "name" : "Gwihwan Moon",
    "url" : "https://www.deepblock.net/blog/author/gwihwan-moon"
  },
  "dateModified" : "2024-06-17T12:28:11.537Z",
  "datePublished" : "2024-06-14T08:26:39.000Z",
  "headline" : "Building Segementation In An Image Of 1.225 Giga Pixels",
  "image" : [ "https://www.deepblock.net/hubfs/Building%20Segementation%20in%20an%20image%20of%201.225%20Giga%20Pixels.png" ],
  "mainEntityOfPage" : {
    "@id" : "https://www.deepblock.net/blog/building-segementation",
    "@type" : "WebPage"
  },
  "publisher" : {
    "@type" : "Organization",
    "logo" : {
      "@type" : "ImageObject",
      "url" : "https://www.deepblock.net/hubfs/deepblock_logo_full2.png"
    },
    "name" : "Omnis Labs Company"
  }
}
```