I know how to take a picture with my camera using the “expo-camera” module, but I don’t know how to setup a system where it takes a photo about 10 times a second, analyses the colors in the image, to be used for tracking. Expo camera can return images as base64, so I’m guessing I would have to use that, but I don’t know to efficiently take a picture constantly and analyse it.
react-native-vision-camera might be more appropriate for what you're trying to achieve. It allows you to write frame processors to analyse frame contents.
But if you wanted to use expo-camera then you could do it the way you described, but you'd need to then find a module to take the Base64 encoded image and turn it into an array or stream of pixel values. This is likely to be very slow and use a lot of memory to run on the JS thread because each image from a standard 12MP camera is going to mean you'll be looping over an array of 12 million RGB values.