Intelligent Security Solution for Restricted Bank Areas | CDL Exam Preparation

Intelligent Security Solution for Recognizing Unauthorized Access in Bank Restricted Areas

Question

You are working on proposing a solution to a prospective security service provider that wants to implement an Intelligent solution in restricted areas of bank (like locker etc.) for recognizing people who are not authorized to access the area. Since the contract will be awarded to the google cloud compatible proposal that is most effective, cost-efficient, and has minimum maintenance overhead.

Which of the below would you propose?

Answers

Explanations

Click on the arrows to vote for the correct answer

A. B. C. D.

Correct Answer: C.

Option A is Incorrect.

OpenCV could be used to write code for the given purpose, but it is an open-source library.

Hence it will be less preferred over Cloud Vision AI.

Option B is Incorrect.

Rekognition is not a Google Cloud offering but is offered by AWS.

Option C is Correct.

Face detection using Cloud Vision AI on Google cloud is:

-Effective,

-Cost-efficient and has.

-Minimum maintenance overhead.

Option D is Incorrect.

This could be implemented, but it requires high expertise and higher maintenance and hence is not preferred.

https://opencv.org https://cloud.google.com/vision

The security service provider wants to implement an intelligent solution for recognizing people who are not authorized to access restricted areas of a bank. The solution should be effective, cost-efficient, and have minimum maintenance overhead. There are four options to consider: OpenCV, Rekognition, Cloud Vision AI, and triplet loss function trained deep CNN model.

Option A: Propose to use OpenCV OpenCV (Open Source Computer Vision Library) is an open-source computer vision and machine learning software library. It can be used to develop real-time computer vision applications, image and video processing, and machine learning algorithms. OpenCV has a wide range of functions, including face detection and recognition. However, OpenCV requires significant programming expertise to implement and integrate into a security system. It may not be the most cost-efficient option due to the required development time and maintenance overhead.

Option B: Propose to implement Rekognition Rekognition is a cloud-based image and video analysis service provided by Amazon Web Services (AWS). It uses machine learning algorithms to detect, track, and recognize people and objects in images and videos. Rekognition can be used for face detection, recognition, and analysis. However, Rekognition is not compatible with Google Cloud, which is a requirement for this project.

Option C: Propose to implement Cloud Vision AI Cloud Vision AI is a cloud-based image analysis service provided by Google Cloud. It uses machine learning algorithms to detect and classify objects in images and videos. Cloud Vision AI can be used for face detection, recognition, and analysis. It is compatible with Google Cloud, which is a requirement for this project. Cloud Vision AI is also easy to integrate into a security system, making it a cost-efficient option with minimum maintenance overhead.

Option D: Propose to implement triplet loss function trained deep CNN model. A triplet loss function trained deep convolutional neural network (CNN) model is a machine learning algorithm that can be used for face recognition. It works by learning a feature representation of faces that is discriminative enough to distinguish between different people. However, developing and training a deep CNN model requires significant programming and machine learning expertise. It may not be the most cost-efficient option due to the required development time and maintenance overhead.

In conclusion, the best option for this project would be to propose to implement Cloud Vision AI. It is a cloud-based image analysis service provided by Google Cloud, which is a requirement for this project. Cloud Vision AI is easy to integrate into a security system, making it a cost-efficient option with minimum maintenance overhead. It can also be used for face detection, recognition, and analysis, which are key features required for the project.