Page 273 - Ai Book - 10

6. Which task in computer vision involves predicting the class of an object in an image?

a. Image captioning b. Object Localization
c. Object Segmentation d. Image Classification

7. What is the primary function of convolution in image processing?
a. Saving images b. Changing pixel values

c. Resizing images d. Copying and pasting images
8. In resizing an image, what happens to pixel information when the image is enlarged?

a. Pixel information is unchanged
b. Pixel information is discarded

c. Pixel values are set to zero
d. Pixel values are randomly assigned

B. Fill in the blanks.
1. Convolution is a process of multiplying together two arrays of numbers, generally of different sizes but
of the same dimensionality, to produce a third array of numbers of the same ___________________.
2. In digital imaging, ______________ represent the smallest single component of a digital image.

3. The term _________________ is used to describe a wide variety of techniques creating visualizations of
the body’s interiors in general and specific organs or tissues.

4. Object segmentation is a challenging task in computer vision that requires the prediction of object
instances and their per-pixel ___________________.

5. Higher pixel resolution indicates higher ___________________ of the image.
6. In a convolution operation, a kernel matrix is slid across the image and multiplied with the input image.
7. Convolution neural network, a representative of deep learning technology, is often used in the field of
the ___________________.
8. CNNs (Convolution Neural Networks) operate directly on images and differ from ordinary neural networks
by taking input as a two-dimensional array and focusing on ___________________ rather than feature
extraction.

C. State ‘T’ for True or ‘F’ for False statements.
1. Convolution is a process of element-wise multiplication of image arrays and another array
called the kernel, followed by summing the results.

2. Pixels, the smallest components of a digital image, are represented by binary digits in computers,
stored and transmitted as a series of ones and zeros.
3. The term “medical imaging” aims to visualize only the external features of the body, excluding
organs and tissues.

4. Object segmentation is a straightforward task in Computer Vision that involves predicting object
instances without considering per-pixel segmentation masks.
5. Resolution refers to the total number of pixels in a digital image and has no impact on the image’s
quality or clarity.
6. In a convolution operation, the center of the kernel must overlap with the center of the image
to obtain the convolution output.

147
147