FeatureX is disrupting the field of satellite image analysis.  We do the heavy lifting: satellite image licensing and acquisition, image enhancement, and computer vision object detection.  You focus on your business needs. (pau: people don't say "oh we need to work with featurex."  but doesn't lend itself to solving business needs)  

Our deep semantic approach enables our systems to understand the fundamental structure of images and then reimagine them in new ways, so you can obtain better insights.

We currently offer five types of image processing and analysis:  

122_3_ours.png

Pan Sharpening

 

We do an amazing job of adding color to high resolution satellite images.

sample_3_1_gan.png

Color Image Enhancement

Get more detail out of your visible spectrum images.

chip521_10504100128B6D00_image_normalized_3.png

Object Detection

Measure the size of a forest, and then count the trees.

92_b.png

Image Alignment

Watch an area of interest change over time.

AOI_2_Vegas_img1305_dg_infra.png

Infrared Image Enhancement

Understand what IR is telling you.


Pan Sharpening (need new name)

A common challenge in satellite imagery is pan sharpening, where high resolution grey scale (or "panchromatic") images are combined with lower resolution color ("RGB") images to produce high resolution color images.  We approach this problem from the inside out, by first analyzing the sementic content of an image, and then reimagining that image as if though it was taken with a better camera.

The results are stunning.  Our systems understand both natural and man-made structure in images and colors them appropriately.  Our AI colors inside the lines, something that is only possible if a system understands lines and other image structure.  Our systems also understand natural structure, such as trees and the ground.  Since we color base on semantics, we can correct for subtle errors cause by imaging through the atmosphere, resulting in near photographic quality color images. 

(change images so labels don't obscure.  also do more object 

Enhancing Resolution

Since we operate at the semantic level, we can take a grey scale image, and guided by the low resolution color information, re-imagine the image at a higher resolution. In the example here, a single color pixel is replace by 64 new color pixels. The re-imagined images are surprisely accurate— we know because we test using high resolution imagery to verify our results. 

INSERT IMAGES HERE SHOWING BEFORE AND AFTER, SIX SLIDES, THREE EXAMPLES OF BEFORE AND AFTER

Object Detection

Want to count the percentage of pine trees in a forest?  How about cars in a rental car parking lot?  FeatureX's semantic object recognition algorithms provide you with unmatched tools to recognize small objects in satellite images.  We can accurately identify objects as small as 5x5 pixels. 

 

Infrared Image Enhancement

Our ability to enhance images doesn’t stop with just visible RGB colors. We can also boost the resolution of other spectrums. For example, here is a panchromatic images along with the near infrared in low resolution, boosted to high resolution. 

AOI_2_Vegas_img4484_dg_infra copy.png

Pricing

Please contact us to discuss pricing for the services you need.  We provide deep discounts for researchers and academic institutions, and we support non-profitit organizations that monitor the health of our planet's environment.