Kuis terkait panduan · Sedang Tingkat

AI dalam Kuis Analisis Citra Satelit

Uji bagaimana model memproses pemandangan observasi Bumi yang besar, menggabungkan pita sensor, dan memetakan perubahan tutupan lahan dari data satelit.

Jalur panduan terkaitAI Dalam Analisis Citra Satelit
Pertanyaan 1 dari 8

What can AI help analysts do with large collections of satellite images?

Pratinjau setiap pertanyaan8 pertanyaan

Jawaban dan penjelasan muncul saat Anda mengikuti kuis.

  1. What can AI help analysts do with large collections of satellite images?
    • Find patterns and changes across imagery for human interpretation
    • Replace all field observations with certain diagnoses
    • Guarantee that every land-cover label is correct
    • Create satellite observations where no sensor data exist
  2. Apa yang dilakukan 'deteksi perubahan' dalam analisis satelit?
    • Menghitung satelit di orbit
    • Membandingkan gambar tempat yang sama dari waktu ke waktu untuk menandai perbedaan yang berarti
    • Mengubah warna satelit
    • Mendeteksi satelit baru yang diluncurkan
  3. Mengapa radar aperture sintetis (SAR) sangat berharga?
    • Ini hanya berfungsi pada hari-hari cerah
    • Dibutuhkan foto berwarna untuk wisatawan
    • Ia dapat melihat menembus awan dan beroperasi di malam hari
    • Ini menggantikan kebutuhan akan AI apa pun
  4. What does the USGS describe NDVI as measuring?
    • A complete diagnosis of plant health
    • The amount of rainfall in a watershed
    • The concentration of minerals in soil
    • The contrast between near-infrared and red reflectance used as a vegetation index
  5. Apa yang dimaksud dengan 'segmentasi semantik' dalam analisis gambar?
    • Memberi label pada setiap piksel berdasarkan kelasnya, seperti air, jalan, atau hutan
    • Menghapus bagian dari suatu gambar
    • Mengompresi gambar ke file yang lebih kecil
    • Menerjemahkan keterangan gambar ke bahasa lain
  6. What PlanetScope coverage did Planet document for its Dove fleet?
    • Continuous video of every location
    • Near-daily multispectral coverage over landmasses
    • Daily imagery of every ocean depth
    • A guaranteed clear image of every land parcel each day
  7. Why are large satellite scenes often divided into smaller image tiles before model processing?
    • It turns radar data into visible-light photographs
    • It guarantees every tile shows a different land-cover class
    • Tiling makes a very large scene manageable for computation and localized analysis
    • It removes the need to align images from different dates
  8. What can combining multiple spectral or radar bands help an analyst infer beyond visible color?
    • The exact identity of every person on the ground
    • The satellite's remaining fuel
    • A guaranteed forecast of next year's weather
    • Properties such as vegetation condition, soil moisture, or flooding