Colour Recognition with Artificial Intelligence (AI)

This ppt is about the color recognition in AI that how AI recognizes the color

1. Colour Recognition with AI Computing Grade 8 Revision Summary

2. How AI Sees the World AI sees data, not pictures like humans. For colours, AI reads RGB values (Red, Green, Blue). Each pixel = 3 numbers ranging 0–255. Example: RGB (255, 0, 0) = Pure Red.

3. AI Big Ideas (in this lesson) Perception – AI sees colours as numbers. Representation & Reasoning – AI uses rules (If–Else). Learning – AI improves using labelled data. Natural Interaction – We give inputs using sliders in Scratch.

4. Starter Discussion How does your camera detect colours? What happens when you use colour sliders in editing apps? How do computers understand colours?

5. Activity 1: Collecting RGB Data Download an ocean satellite image. Use an online colour picker to get RGB values. Record values in a simple table (Red, Green, Blue).

6. Activity 2: Testing AI Logic in Scratch Create 3 sliders (Red, Green, Blue). Screen changes colour as sliders move. Add If–Else logic: If R > 200 & G < 100 & B < 100 → say 'Red'.

7. AI Concepts Applied Representation & Reasoning: Using rules to decide colour. Learning: Training with RGB samples. Garbage In = Garbage Out: Wrong data → Wrong AI output.

8. Activity 3: Building a Colour Bot Build Scratch bot with sliders for RGB input. Display colour preview square. Use If–Else rules to predict colour name. Test bot using collected data.

9. Natural Interaction Students interact using sliders instead of typing. Similar to real AI systems: Siri (voice), Apps (touch/visual input). Scratch chatbot responds in real time with colour name.

10. Reflection Questions What did you learn about how AI sees colours? Why is data important in training AI? Can AI always be trusted to give the correct colour? Why/why not?

11. Key Takeaways Colours are represented digitally as RGB values. AI can use rules (If–Else) to classify inputs. AI improves using labelled training data. Good data = Good AI, bad data = wrong predictions.