Computer Vision DY0-001 Practice Questions (Page 2)

Questions 6–10

1. 6expert · hard

   A government agency needs to digitize historical documents that contain both printed text and handwritten annotations. The documents are old and have varying levels of degradation. The team must extract the printed text accurately, but the handwritten annotations are also important. Which OCR approach is most appropriate?

   • AUse a single neural-network OCR for both printed text and handwriting
   • BUse a statistical OCR for the printed text and template matching for the handwriting
   • CUse a single template-matching OCR for both printed text and handwriting
   • DUse a neural-network OCR for the printed text and a separate neural-network OCR fine-tuned for handwriting for the annotations
   Reveal Solution Discussion Select an answer first
2. 7foundation · easy

   Which tracking algorithm iteratively shifts a window toward the region of highest density in a feature space, such as color histogram?

   • AKalman Filter
   • BYOLO
   • CSSD
   • DMean Shift
   Reveal Solution Discussion Select an answer first
4. 8application · medium

   A medical imaging team is training a model to detect tumors in chest X-rays. They have only 500 labeled images, and the dataset is imbalanced with few positive examples. The team wants to improve the model's generalization. Which data augmentation strategy is most appropriate?

   • AUse only horizontal flips because X-rays are symmetric
   • BOversample the positive class by duplicating the same images in the training set
   • CApply random rotations, flips, and slight intensity shifts to the X-ray images
   • DApply aggressive random cropping and heavy color jitter
   Reveal Solution Discussion Select an answer first
5. 9foundation · easy

   How does data augmentation typically improve model performance in computer vision?

   • ABy increasing the model's inference speed
   • BBy reducing the model's memory footprint
   • CBy reducing overfitting through increased training data diversity
   • DBy eliminating the need for a validation set
   Reveal Solution Discussion Select an answer first
6. 10foundation · easy

   Which of the following is a direct benefit of using data augmentation in a computer vision model?

   • AImproved robustness to variations such as lighting and orientation
   • BElimination of the need for labeled data
   • CReduced training time
   • DGuaranteed higher accuracy on the training set
   Reveal Solution Discussion Select an answer first