Evaluate the impact of an AI application used in the real world. (case study) : Working with X-Ray images: Case Study -

The video tutorial covers image preprocessing techniques, focusing on scaling and rescaling images using min-max normalization instead of dividing by 255. It discusses the impact of rescaling on image quality and the importance of problem-specific considerations, such as detecting anomalies. The tutorial also explains augmentation techniques for training, emphasizing the need for computational resources. Finally, it highlights the application of learned techniques to different datasets, stressing the importance of understanding problem dependencies.

Why is the min-max scaling method preferred over dividing by 255 for the images discussed?

• Because it is the only method available

• Because it enhances the color contrast

• Because the images are not natural and may not fit the 0-255 range

• Because it is faster to compute

What is the main reason for rescaling images from 1024x1024 to 256x256?

• To improve image quality

• To reduce computational load

• To enhance the detection of small anomalies

• To make the images compatible with all models

What is a potential downside of reducing image size when detecting anomalies?

• The image might become too large to process

• The anomaly might become too small to detect

• The colors might become distorted

• The image might lose its original format

Why might augmentation not be applied to every image during training?

• Because it is not necessary for all datasets

• Because it reduces image quality

• Because it is a time-consuming process

• Due to computational constraints

How does the use of random values influence the augmentation process?

• It determines the type of augmentation to apply

• It decides whether to apply augmentation or not

• It sets the intensity of the augmentation

• It selects the images to be augmented