CLOUD-DIGITAL-LEADER · Question #90
The Border Security Agency has hired your software services firm to build an application for them that will collect information about visas stamped on passports. You are given stamped images. You…
The correct answer is A. Use Cloud Vision API -write code to identify the text blocks, copy the data, and store it. Cloud Vision API provides ready-to-use OCR (Optical Character Recognition) and text detection capabilities that can read text from images without requiring custom model training. For visa stamp images, it can identify and extract text blocks (country name, validity dates) which…
Question
The Border Security Agency has hired your software services firm to build an application for them that will collect information about visas stamped on passports. You are given stamped images. You have to find out which country issued the visa and the period of validity. Pull out this data and put it into a database. Which of these applications would be suitable for that?
Options
- AUse Cloud Vision API -write code to identify the text blocks, copy the data, and store it
- BUse TensorFlow -write code that will identify the type of visa and the bounding text blocks. Copy
- CUse AutoML -upload other images of visas and run the model creation process which will
- DUse Data Labeling service -outsource the work of marking and extracting the in-formation to
How the community answered
(32 responses)- A91% (29)
- B6% (2)
- D3% (1)
Explanation
Cloud Vision API provides ready-to-use OCR (Optical Character Recognition) and text detection capabilities that can read text from images without requiring custom model training. For visa stamp images, it can identify and extract text blocks (country name, validity dates) which can then be stored in a database. TensorFlow requires building a model from scratch. AutoML requires uploading labeled training images and training a custom model. Data Labeling service only assists with annotating data, not automated extraction.
Topics
Community Discussion
No community discussion yet for this question.