Question

Project : facial recognition system

use the DEEP method to reflective writing. The four steps in this method are:

• 2.1 D - Describe experience (past or present).
  1. Describe the problem
• 2.2 E - Explore or consolidate new learning
  1. Read/Research about the problem
• 2.3 E - Evaluate/analyse/self reaction
• 2.4 P - Plan future learning

Answer 1

Face recognition systems:

A facial recognition system is a technology able to identifying or verifying a person from a digital image or a video frame from a video source. There are various methods in which facial recognition systems work, but in general, they work by differentiating selected facial features from given image with faces within a database. It is also described as a Biometric Artificial Intelligence based application that can especially identify a person by analyzing patterns based on the person's facial features and shape.

2.1 D- Past and present experiences with face recogition:

Earlier facial recognition technology was considered as an idea of science invention. But in the past decade, facial recognition technology has not only become real but it’s broad. Today, people can easily read articles and news letters stories about facial recognition everywhere.

The world considers that Woodrow Wilson Bledsoe was the father of facial recognition. In the year 1960s, Bledsoe developed a system that could arrange faces  photos by hand using the RAND tablet. The tablet is a device in which people could use to enter vertical and horizontal coordinates on a grid with the help of a stylus that released electromagnetic pulses. People used that system to manually record the corelate areas of facial features like eyes, nose, mouth, and hairline.

The manually recorded metrics could be later saved within a database. And when the new photograph of an individual person was entered into the system, it was able to get the most closely related image via database. During this period, face recognition was unaffected by technology and computer processing power.

1.Social Media 2010 to present:

When 2010 started, Facebook started using a facial recognition feature that helped to detect people with featured faces in the photos updated by Facebook users. While the update created publicity in the media industry Facebook stayed very low key since there was no evident negative impact on website popularity and usage.

2. Largest biometric installation at the airport:

In 2011, the Panama government and US Secretary of Homeland Security Janet Napolitano collaborated and authorized a pilot program of the Facial Recognition platform.

3. Adopted in the military for dead bodies:

In 2011, U.S. law enforcement and military professionals used face recognition for the identification or recognition of dead bodies.

4. Law enforcement adoption of facial recognition in 2014:

Law enforcement agencies came forward to accept mobile face recognition in 2014. The technology became assured for the retail industry in 2017.

2.2 E- Explore or consolidate new learning:

It is very difficult for a recognition system to identify them. These problems can be due system defects used in face recognition, such as camera distortion, background noise, inefficient storage, improper techniques etc. More than that there can be network problems or defects due to environmental conditions.

Challenges or problems in face recognition:

1. Pose variations

2. Presence/absence of structuring elements/occlusions

3. Facial expression changes

4. Ageing of the face

5. Varying illumination conditions

6. Image resolution and modality

7. Availability and quality of face datasets

2.3 E- Evaluate/analyse/self reaction:

Solutions to the above challanges of facial recognition are:

1. Face recognition systems: Most of the AFR systems consist in a two‐step process based firstly on facial feature extraction, and second, on facial feature classification/matching against an available face database.

2. Feature extraction: Facial features are representing the face in a codified way which is computationally efficient for further processes such as matching, classification or other machine‐learning techniques, in order to perform AFR. On the other hand, computing facial features in an image could serve to find a face and to locate it within the image.

2.4 P- Future learning in Facial recognition systems:

Governments all over the world are increasingly investing their resources in facial recognition technology, especially the US and China are the leaders in the facial recognition market.

The government of the USA has decided to increase airport security with a facial recognition system for identification and registration of visitors. The US has several states that have allowed law enforcement to run searches within the database these searches involves details of a driver’s license and ID photos. The facial recognition and resulting search techniques can be also used in police checks.

The whole world is using this technology and acquiring many benefits. In India, banks are using this facial recognition technology to prevent fraud at ATM’s. It is also used for reporting duplicate voters, verification and validation of passport and visa, driving license, etc.