引導式測驗 · 中等 等級

人工智慧在城市垃圾收集和回收分類測驗的應用

了解人工智慧如何支援回收分類和路線,為什麼當地條件很重要,以及城市應該衡量哪些服務品質。

問題 1 的 8

What does a vision system on a sorting line classify?

預覽每個問題8 問題

當您參加測驗時,就會出現答案和解釋。

  1. What does a vision system on a sorting line classify?
    • Material or object categories presented to the sensors
    • Whether a city accepts every material
    • Whether an item will be recycled into a new product
    • Whether a resident followed local rules
  2. Why can a sorting model misclassify recyclable material?
    • A recycling logo guarantees correct classification
    • Models never see mixed materials
    • Items may be dirty, crushed, overlapping, or unfamiliar
    • Conveyor speed determines material policy
  3. What should a city measure when testing fill-level routing?
    • Only the number of installed sensors
    • Service reliability, missed pickups, and route performance
    • Only the model's predicted route length
    • The brightness of bin labels
  4. What does collection into a recycling truck establish?
    • That all items meet local recycling rules
    • That the item became a new product
    • That sorting equipment had no contamination
    • That material was collected, not that it was successfully processed
  5. Why evaluate low-volume material categories separately?
    • Aggregate accuracy can hide poor performance on rare classes
    • Every category is equally common
    • Rare categories cannot affect contamination
    • The model automatically prioritizes uncommon materials
  6. What should happen when a facility’s model is updated?
    • Assume all materials will be classified identically
    • Remove worker feedback channels
    • Monitor category behavior and document the change
    • Stop measuring contamination
  7. What information does a fill sensor contribute to routing?
    • A guarantee of the bin’s exact weight
    • An estimate of bin status that should be checked for data quality
    • The correct recycling category for each item
    • Proof that a collection occurred
  8. Why include worker feedback in evaluation?
    • Workers determine model accuracy by vote
    • Feedback replaces measured contamination
    • Safety is independent of automated sorting
    • Workers can identify safety issues and misclassifications on the line