February 23rd is National Tile Day. But if you’re looking to learn about the benefits of granite versus marble, you’ve come to the wrong place. This year, we’re celebrating by taking a closer look at a different kind of tiling.

The tiling we’re discussing is a powerful computer vision approach, which sees a large image broken into many separate, smaller “tiles” and then reassembled.

What Is Tiling?

Tiling is an important process for analysis of images with computer vision and allows for a more detailed look at specific sections of an image without sacrificing resolution. The technique is typically used for detecting small objects in high-resolution images. For example, tiling can be used with satellite imagery to recognize specific fields of crops or find a golf ball after a perfect drive.

After dividing an image into tiles, computer vision models can then inference on each individual tile before stitching them back together into a composite image. This results in a more accessible image for inferencing that can ultimately be used for batch and real-time predictions helping to solve organization and industry-wide challenges.

Tiling: Example Use Cases

The tiling process is often necessary when users or businesses need to detect small objects within images or minute imperfections within objects of any size. Take finding defects in pharmaceutical products like pills, for example. For a human to do this consistently and accurately, it would require breaking out a magnifying glass or microscope. Even a standard computer vision solution won’t be especially useful without tiling. All the necessary resizing will result in enough resolution loss to make a picture useless.

By dividing an object like a pill into several dozen small tiles, vision AI solutions like Plainsight’s are capable of recreating a magnified image in high resolution. This empowers vision AI model training with highest-quality datasets, enabling models to grow faster and more accurate with time.

Tiling helps make satellite imagery more accessible for computer vision solutions too. To make sense of satellite photos and video, users typically require several different zoom and cropping levels. A solution capable of tiling can divide, analyze, and reassemble images to simplify the process, making it easier to identify and auto-zoom to fit common images like parking lots or fields.

Tiling: Addressing Challenges

Automated tiling can sometimes result in “cutoff” and double counting. When the initial image is reassembled, an object spread across multiple tiles could get counted twice, resulting in an incorrect overall tally of relevant objects.This undesired result can be corrected by implementing a sliding window that tells the model to ignore images in certain buffer areas. Simply put, the model will only count the tile where the larger portion of the item is included.

Filters can also help to eliminate monotonous, irrelevant data. Empty tiles may be a concern in instances where a small number of important visual data points are spread across an especially large frame. Imagine you were tasked with drone footage analysis to identify signs of crop loss in an expanse of farmland or survey a large forest for active fires. While a person could naturally skip over irrelevant tiles, a vision AI model needs to learn to do so. With continued data inputs, an expertly-deployed solution can quickly improve in its ability to eliminate unnecessary tiles from its analysis, saving time and addressing a range of potential use cases.

See More with Plainsight

Plainsight’s no-code, end-to-end vision AI platform makes it easy for the world’s most forward-thinking companies to do more with their visual data.

Platform Features

  • Intelligent Image and Video Data Curation: Collect visual datasets to provide a foundation for training custom vision AI models.
  • AI-Powered Data Annotation: Accelerate automation with smart polygon selection, predictive labeling, and object recognition.
  • Automated Machine Learning: Easily train models with a breakthrough process designed to reduce time to value.
  • Operationalized Solutions: Deploy and scale applications at the edge, in the cloud, or on your own premises.

Contact us today to learn more about how Plainsight can transform your computer vision pipeline to enable better strategies for tackling challenges of every size.

About the Author

Elizabeth Spears, Plainsight Co-Founder and Chief Product Officer

Elizabeth Spears is an AI technology executive who has led the productization of a series of multi-layer, compute-intensive software service platforms, usually pioneering the product management function in her companies. At Plainsight, Elizabeth has productized nascent AI & computer vision technologies to a very high functional and usability standard, creating products that help forward-thinking businesses realize the untapped potential of their visual data.

 View All Blogs