Machine Vision

As Teresa and I speak at conferences, we often share about the progress being made in Machine Vision and how it applies in SO many ways in our life (today), and our life (in the future).

On Monday’s blog, we asked you to think about how you recognized the crop below as wheat. The challenge of machine vision is how to help a computer to learn to recognize what wheat is (colors, kernel size, still standing, harvested etc.).

To make these problems approachable to kids, we have leveraged the framework Amazon Web Services (AWS) has created to help beginning developers to understand machine vision. One of the sample projects is ‘hot dog’ or ‘not a hot dog’ which seems simple enough until you think of all of the variations that need to be accounted for (ketchup on it, mustard, relish, bratwurst, casing, pre-smoked, footlong, bun). Who knew a hot dog could be so complex! Then think about how that applies to agriculture, and it can be pretty overwhelming, so we started our kid-friendly project with telling crops apart.

Being able to tell crops apart are key to many different opportunities in agriculture:

  • Ariel imagery (acres planted in each crop)

  • Sorting of commodities as they are delivered (green soybeans)

  • Recognizing ‘wheat’ or ‘not wheat’ (weed detection)

  • Recognizing ‘insect’ or ‘not insect’ (and counting the number of insects detected)

We were not sure at the start of this project if a 4th grader could figure this out, with limited adult guidance, and we are happy to say he has! So, if a 4th grader can do it, the opportunities in agriculture are limitless. Interested in your kids (ages 8-15) learning how Machine Vision works in a 1 day Farm Femmes camp? Send us an e-mail at or and we will add you to the pre-public camp dates e-mail list!