Page 364 - AI Ver 3.0 class 10_Flipbook
P. 364
10. What is the result of the process that highlights the important features detected by the kernels called?
a. Feature map b. Weight map
c. Edge map d. Colour map
11. What does the Fully Connected Layer do in a CNN?
a. It applies convolution operations
b. It reduces the image size
c. It classifies the image into specific labels
d. It pools the features from the image
12. What type of transformation is introduced by the ReLU function?
a. Linear transformation b. Non-linear transformation
c. Logarithmic transformation d. Exponential transformation
SECTION B (Subjective Type Questions)
Short answer type questions.
1. Why is data preparation important when using No-Code AI tools?
2. How can a user fine-tune parameters in a No-Code AI platform?
3. Explain how AI models are evaluated in No-Code platforms.
4. What is Teachable Machine, and who developed it?
5. List the types of data inputs that Teachable Machine can process.
6. What is the main purpose of Orange Data Mining?
7. What is an image feature in Computer Vision?
8. Why are corners considered good features for image processing tasks?
9. What is the role of the Convolution Layer in a Convolutional Neural Network (CNN)?
10. What is the purpose of the Fully Connected Layer in a CNN?
11. What is the difference between Max Pooling and Average Pooling in a CNN?
12. Explain the concept of feature maps and their role in the Convolution Layer.
In Life 21 st Century #Critical Thinking
#Leadership & Responsibility
Skills
Imagine a camera that can recognise objects like a cat, dog, or tree in a photo. How does it manage this? Discuss
with your class how AI's ability to process image data could affect individuals and society.
Deep Thinking 21 st Century #Critical Thinking
Skills
In computer vision, the goal is to make computers understand and interpret images, just like how humans recognize
objects. Imagine you are developing a computer vision platform that uses a convolutional neural network (CNN)
to recognize objects in images. This platform requires huge amounts of image data to train effectively, sometimes
even terabytes of images. What happens if the data is flawed? Share your thoughts with the class.
362 Touchpad Artificial Intelligence (Ver. 3.0)-X

