GE103 - Human vs AI systems

Computer vision systems

Dr Robert Batzinger
Instructor Emeritus

Payap University
Chiang Mai, Thailand
3-Sep-2022

1 Regression analysis of COVID19

11-09-22 Prediction

2 Creativity

Where in the brain do creative ideas come from?
Why do you think individuals who make it a habit to read lots of books, watch a wide range of movies or documentaries, and play music or sports, can generally offer a wider range of creative solutions?
What defines a really good creative idea? Where do such ideas come from?
How do creative ideas of a person compare with those generated by a computer? How are they similar and how are they different?
How can you prevent creative ideas from becoming subject to accusation of plagarism and copyright infringement?

3 Computer Vision Systems

3.1 Agenda

Segmentation of visual elements
Object recognition
Individual recognition
Augmented reality

4 Segmentation

Break a scene into regions of interest
Simplifies the computing by identifying areas
Allows for analysis and simplifies data entry
Human are often distracted in projects requiring detail

5 Object analysis

6 Product labeling without AI

Requires a sticker and a reader
Fails if the sticker becomes detached

6.1 QR Coding

6.2 Text entry

6.3 Commercial computer vision applications

6.4 Some facts about potato chip sorting in Lumphun

Manual sorting lines

2 bags per sec, miss rate: 1%
requires 20 sorters, 1 engineer
sorters make 120-200 baht/day depending on experience

Automated lines

45 bags per sec, miss rate 0.02%
requires 4 sorters, 1 engineer
sorting machine costs around 1.8 million baht
Required more workers (especially in preparation, packaging and quality control)

6.5 Peanut sorting

Peanut sorting

7 Virtual Reality

Started with the concept of a video wall
Controls the full visual experience often using googles
Uses sensors to detect motion, location and orientation
Adjusts the world to simulate reality
Often leads to accidents

8 Augmented reality - AR

Merging sensory input from our environment with other visual stimuli and data
The purpose is to extend the human experience (making the most of it)
Sixth sense
AR in Real estate
Future of shopping: Microsoft vision clip

8.1 Differentiate between these terms

Augmenting human intellect
Amplifying perception
Assisted cognition

8.2 Augmenting human intellect

http://app.rid.go.th/reservoir/rsvmiddle

8.3 Water levels

14 Sept

8.4 Corresponding data table

Date	Max ht	Sched ht	Curr ht	Intake	Outflow
8 Sept	170	131.80	138.00	2.51	0.48
9 Sept	170	133.80	140.00	1.84	0.21
10 Sept	170	143.80	150.00	10.33	0.21
11 Sept	170	149.80	156.00	6.95	1.18
12 Sept	170	151.80	158.00	5.83	3.55
13 Sept	170	154.80	161.00	6.91	3.82
14 Sept	170	165.80	172.00	18.54	7.27

8.5 Amplifying perception

8.6 Assisted cognition

Bushfire Risk is based on:

Potential Fuel Load (tonnes/ha),
Maximum Landscape Slope (degrees)
Fire Weather Severity (Forest Fire Danger Index)

8.7 Augmenting Cognitive Limitations

Memory and learning
Attention
Creativity
Decision making
Effective and sensitive communication
Motivation
Self and situational awareness
Emotional regulation