Page 361 - AI Ver 3.0 class 10_Flipbook
P. 361
10. Which of the following is NOT considered a low-level image feature?
a. Colour b. Intensity
c. Objects d. Textures
11. Teachable Machine is built on which JavaScript library?
a. TensorFlow.js b. Game development
c. Web development d. Hardware programming
12. What is an image feature in Computer Vision?
a. A programming function b. A specific element extracted from an image
c. A type of AI model d. A random pixel
13. Which of the following is a common type of pooling used in CNN?
a. Max Pooling b. Min Pooling
c. Average Pooling d. Both a and c
14. In which layer are the final outputs of a CNN predicted?
a. Convolution Layer b. Rectified Linear Unit (ReLU) Layer
c. Pooling Layer d. Fully Connected Layer
15. What is the result of applying ReLU to the feature map?
a. All the feature map values are doubled
b. All negative values are turned to zero and positive values are kept unchanged
c. The feature map is resized
d. The image is sharpened
SECTION B (Subjective Type Questions)
Short answer type questions.
1. Name the four important layers in CNN.
Ans. The 4 layers in CNN are:
• Convolutional Layer • Rectified Linear Unit (ReLU)
• Pooling Layer • Fully Connected Layer
2. Why is pooling important in CNN?
Ans. It helps in reducing the number of parameters and computation in the network thus making it more manageable and
improving the efficiency of the whole system.
3. What is a No-Code AI tool?
Ans. A No-Code AI tool is a software platform that allows users to build, deploy, and manage Teachable Machine is designed
to make artificial intelligence (AI), machine learning (ML), and deep learning (DL) accessible to everyone, including
beginners with no technical background.
4. What is the role of AutoML in training AI models with No-Code tools?
Ans. AutoML (Automated Machine Learning) features optimise the training process by automatically selecting the best
algorithms and parameters, ensuring the most efficient model training without manual intervention.
Computer Vision (Practical) 359

