AI in Handwriting Recognition Quiz

Answers and explanations appear as you take the quiz.

1. What is the main difference between online and offline handwriting recognition?
   • Online captures stroke order and timing while offline works from a static image
   • Online needs internet while offline does not
   • Online only works on phones while offline works on paper
   • There is no real difference between them
2. Why is recognizing cursive handwriting especially difficult for AI?
   • Cursive uses different colors of ink
   • Cursive letters connect and blur together, making them hard to separate
   • Cursive is always written faster than print
   • Cursive only exists in old documents
3. What does Connectionist Temporal Classification (CTC) allow a handwriting model to do?
   • Translate handwriting into another language
   • Detect the color of the pen used
   • Output text without needing every letter pre-segmented and labeled
   • Compress the image to save storage
4. How does USPS handle some mail pieces when automated equipment cannot read the address?
   • It asks the sender to resubmit every letter
   • It discards the mail piece immediately
   • It infers the destination from the envelope color
   • It photographs the address and sends the image to a Remote Encoding Center for human encoding
5. In a typical handwriting recognition pipeline, what role does a language model play?
   • It re-scores candidate words using context so unlikely strings become real words
   • It scans the paper for stains
   • It chooses the font for output
   • It measures the pressure of the pen
6. Which type of neural network is commonly used first to extract visual features from the ink image?
   • A decision tree
   • A convolutional neural network (CNN)
   • A simple lookup table
   • A random number generator
7. In Connectionist Temporal Classification, what is the blank symbol used for?
   • It marks the end of every document image
   • It identifies the writer’s name from the page margin
   • It allows the model to align frame-level outputs to text without fixed character boundaries
   • It converts handwritten words into a different language
8. After a CNN extracts visual features from an ink image, what does an LSTM or Transformer contribute?
   • It measures the scanner’s battery level
   • It changes the paper’s physical stroke order
   • It replaces the need to learn from text examples
   • It models the sequence of features so they can be decoded into ordered text