FRSC-3100: Exam Review

LECTURE 1 - INTRODUCTION AND DISRUPTIVE TECHNOLOGIES

Disruptive Technologies:

• Disruptive technology is an innovation that significantly alters how consumers, industries, or businesses operate. Disruptive technology superseded an older established process, product, or habit with recognizably superior attributes.
• Recent disruptive technology examples include e-commerce, online news sites, ride-sharing apps, and GPS systems.
• Clayton Christensen introduced the idea of disruptive technologies in a 1995 Harvard Business Review article.

Disruptive Technologies Covered:

• Artificial Intelligence/Machine Learning
  • Text generation
  • Image compilation
  • Pattern recognition
• Autonomous Vehicles
  • UAV
  • Terrestrial
• 3D Imaging
  • TLS
  • Photo-g
  • NeRF
• eXtended Reality
  • AR, VR, MR Metaverse
• Neural-Computer Interface
• Nanotech
• Genetic Engineering and DNA

Artificial Intelligence:

• AI is based on algorithms that mimic human neural networks.
• When coupled with extremely high computing speed, large volume dataset processing, and the ability to self-learn.
• On its most basic level it can identify patterns, predict sequences and/or generate data in a fashion that can appear life-like in its speed and sophistication.
• Generative AI: Focuses on understanding patterns and structure in data and using that to create new data that looks like it.
  • Writing blocks of text
  • Writing lines of code
  • Creating photo-realistic images
• Predictive AI: Focuses mainly on classification, learning the difference between “things”.
  • Recommendation engines used by Netflix or Amazon to distinguish between things you might want to watch/buy and things you’re unlikely to be interested in.
  • Navigation apps to distinguish between good routes from A to B and one you should avoid.

Levels of Artificial Intelligence:

• Artificial Narrow Intelligence (ANI): Specialized to the function in which it has been developed.
  • The technologies running our smartphones, online purchases, and social media apps.
  • Artificial General Intelligence (AGI): Referred to as “human-level AI” because it describes the capacity of a computer that is as smart as a human (Singularity).
  • These computers possess the ability to plan, reason, problem-solve and comprehend abstract and complex ideas, learn from experiences and develop intelligent conclusions as fast as, or perhaps faster than, the human brain.
• Artificial Super Intelligence (ASI): The point at which computers possess intellectual capacity far greater than that of human beings with the capacity for social skills and general knowledge that would increase exponentially over time.
  • Does not exist yet but worries many of today’s leading thinkers.

Examples of AI:

• ChatGPT
• Lumalabs.AI (3D Models)
• Luna
• Soundstorm (Audio Generation)
• Synthesia IO
• Chicago Crime Prediction
  • 90% accuracy and identified police bias
• Facial Recognition
• AFIS
• Corsight AI (DNA Phenotyping)

Robotics:

• SPOT terrestrial robot equipped with:
  • SPOT CAM captures spherical images and comes with 30X optical zoom for detailed inspections.
  • SPOT CORE provides dedicated processing for applications requiring on-robot computation.
  • SPOT ARM enables mobile manipulation for tasks like opening doors and grasping objects.
  • SPOT GXP provides regulated power and an Ethernet port
• Multi-Modal Mobility Morphobot (M4)
  • Terrestrial mobility on wheels
  • Walking, crouching, wheel rolling, and tumbling capabilities
  • UAS flying capability

NeRFs:

• Neural Radiance Field: Computer-generated 3D models that mimic the dynamics of lighting on an object or scene to create a near photo-realistic representation that can in turn be viewed from any viewpoint as either a static image or simulated video.
• ClimateNeRF: Physically-based neural rendering for extreme climate synthesis

SLAM:

• A handheld device that scans and maps the dimensions and layout of a scene in real time via a rotating laser.

Photogrammetry:

• Compilation of photos and videos to map and visualize a scene
• Crowdsourced post-blast forensic analysis of the Beirut explosion

Terrestrial Laser Scanning:

• Stationary device that scans the area using a laser

Crime Scene Documentation:

• Capture:
  • Station points and targets
  • Individual scans
  • Resolution and quality
  • Traversing and control
• Process:
  • Filters, registration, colourizing
• Present:
  • Documentation and analysis
  • Visualization
• Documentation purposes:
  • Dimensional data
  • Screenshots
  • Diagrams/maps
  • Virtual video/ortho imaging
  • 3D printed models
  • Virtual Reality
  • Animations incorporating additional dynamic vehicles, people, objects
• Analysis purposes:
  • Blood-spatter
  • Ballistics
  • Height/velocity determination
  • Line of sight
  • Arson
  • Accident reconstruction

- The Campbell Report:

• 1996 regarding R v Paul Bernardo
• Issues: Gaps in the investigation/CFS processes
• Result: Major Case Management

The Kaufman Report:

• 1998 regarding R v G.P. Morin
• Issues: gaps in the investigation/CFS bias, junk science
• Result: Call for forensic scientists to be objective, independent, and accurate

The Goudge Report:

• 2008 regarding Dr. Charles Smith
• Issue: Faulty forensic pediatric pathology with erroneous conclusions
• Result: Commitment to training, collaboration and peer review

The Hart House Report:

• 2013 Forensic Science in Canada: A Report of Multidisciplinary Discussion
• Issue: A patchwork of contributions
• Result: Call for improved higher training, guidelines, standards, certification, accreditation and ethics covering nine forensic disciplines
  • Pathology, Anthropology, Odontology, Nursing, Entomology, Physica; sciences, Toxicology, Biology, and Psychiatry

The Overall Theme:

• Over the past three decades, forensic science has come under increasing scrutiny for shortcomings, errors, and bias.
• Studies have shown that certain forensic techniques are unreliable and have led to wrongful convictions:
  • Bite mark analysis
  • Hair microscopy
• There have been concerns about bias in interpreting evidence, especially regarding racial and socioeconomic disparities.
• In response to these issues, various commissions and reports along with organizations such as the National Academy of Sciences (NAS) and the President’s Council of Advisors on Science and Technology (PCAST) have called for improvements in the quality and standardization of forensic science.
• Efforts are being made to increase transparency and accountability through:
  • Independent scientific reviews
  • An emphasis on scientific rigour
  • Creating and aligning standards
  • Oversight boards
  • Improved training and best practices
  • Certification programs and accreditation for forensic practices
• There is also an increasing focus on multidisciplinary approaches to forensic science
  • This includes collaboration between scientists, statisticians, legal experts, and others

LECTURE 2 - AI AND FACIAL RECOGNITION TREVOR LAUGHLIN

Narrow AI:

• AI designed to do one specific task
• Virtually all AI today
• Areas of use:
  • Chatbots
  • Cancer detection
  • Protein analysis
  • Roccomendations for shopping/viewing
  • Text-to-speech or speech-to-text
  • Google Maps

Generative AI:

• AI that creates some kind of “new” content
• Areas of use:
  • Stable diffusion Art AI
  • Chat GPT

Large Language Models:

• AI program designed to understand, generate, and work with human language on a large scale
• Generate coherent and contextually relevant text based on the input they receive
• Fed large datasets containing a wide array of text, from which they learn language patterns, structures, and nuances
• Areas of use:
  • Translation
  • Summarization
  • Answering questions
  • Creative writing
• Found in:
  • Chatbots
  • Writing assistants
  • Search engines
  • Predictive text

Natural language Processing (NLP):

• Focuses on the interaction between computers and human language
• Involves enabling computers to understand, interpret, and respond to human language in a way that is both meaningful and useful.
• Algorithm: A simple set of rules.
• Areas of use:
  • Chatbots and virtual assistants
  • Hey Google voice interfaces
  • Customer service automation
  • Content categorization in media
  • Email filtering
  • Language translation services

Application Programming Interface (API):

• Protocols that allow different software programs to communicate with each other/AI
• Areas of use:
  • Text analysis: Language detection, text summarization
  • Image recognition: Object detection, facial recognition
  • Natural Language Processing
  • Voice interface
  • Captchas

Machine Learning:

• Uses data and algorithms to mimic human learning
• Statistical methods to train algorithms to classify or predict and even provide insights into data mining projects
• “Simplistic” = Requiring manual feature selection and engineering
• Areas of use:
  • Simple facial recognition
  • Medical diagnosis

Deep Learning:

• A subset of machine learning
• Uses neural networks to analyze and learn from data
• Well suited for processing unstructured data like images and text
• “Complex” = Have multiple layers that automatically detect and learn hierarchical feature representations
• Areas of use:
  • Image and speech recognition
  • Advanced facial recognition from social media data and video surveillance

How Dumb is AI?

• The monkey in the theorem:
  • Based on generating text randomly, without understanding
  • Output mostly random gibberish with the occasional coherent sentence
  • No training
• AI
  • Learning from the previous inputs
  • Outputs are not random
  • Based on learned patterns and feedback
  • Do not truly understand the text they generate but mimic patterns (predictive text)

Human Understanding:

• Sensory memories: Visual, tactile, auditory
• Semantic understanding
• Episodic memories
• Emotional responses

AI Training Data Issues:

• Which pictures does AI think are cancerous cells?
  • Rulers in the picture are signs of malignant skin cancer
  • AI learned that a ruler = cancer
• WHich pedestrians are obstacles?
  • Not those outside crosswalks
  • Bias for people of colour
  • Rules of the road
• AI has no semantic understanding of words
• AI is only as smart as the data it is trained on and the feedback it receives
• Biased data in language and photos

Feature-Based Facial Recognition:

• Identify fiducial points (specific facial landmarks)
  • Measures relationship and distance between facial features (eyes, nose, mouth…)
  • Creates geometric relationships
• Involves deep learning algorithms which improve the accuracy
• Advantages:
  • Generally more robust to variations in lighting and facial expressions since it relies on stable and distinctive facial landmarks
• Applications:
  • Often used in systems where high precision is required and in situations where the facial expressions might change
  • 3D systems can also measure depth information, which helps in distinguishing a real face from a photograph
  • Automated border control systems, passports and smartphones

Appearance-Based Facial Recognition:

• Use holistic information from the face
• These techniques analyze the entire facial image as a whole
  • Principal Component Analysis
  • Linear Discriminant Analysis
• Advantages:
  • Can be more effective in capturing more detailed and subtle facial characteristics
  • Generally easier to implement since they don't require the detection of specific facial landmarks
• Applications:
  • Commonly used in environments where facial conditions are more controlled
    • Passport or ID verification systems
    • Social media photo tagging

Knowledge-based Facial Recognition:

• Designed to identify suspicious or abnormal behaviours.
• Based on predefined rules or knowledge about facial expressions and behaviours.
• Relies on a database of facial expressions and behaviours corresponding with specific intentions/emotions.
• Analyzes facial expressions in real-time and compares them to ID potential security threats.

Pros of Facial Recognition:

• Enhanced security and safety:
  • Can significantly increase security in public spaces and prevent crime by identifying suspects in real-time
• Efficiency in policing:
  • Allow for quick identification of individuals in criminal investigations, leading to faster case resolutions
• Convenience in consumer applications:
  • Streamlines processes like automated check-ins
• Support in forensic work:
  • Helps with the identification of victims in disaster situations and contributes to forensic analysis

Cons of Facial Recognition:

• Privacy concerns:
  • Collection and storage of facial data often occurs without consent.
• Potential for abuse:
  • Risk of misuse by authorities, constant monitoring of citizens.
• Bias and inaccuracy:
  • Studies show that facial recognition systems may have radial and gender biases, leading to higher error rates for certain groups.
• Legal and regulatory challenges:
  • Lack of comprehensive legislation governing the use and limits of facial recognition.
• Chilling effect on civil liberties:
  • The fear of being constantly watched can deter individuals from exercising rights to freedom of assembly and expression.

Clearview AI in Canada:

• Utilizes a database of 3+ billion images scraped from the internet, including social media, to aid in facial recognition for law enforcement
• Employed by various police departments across Canada, including the RCMP
  • Software compares photos taken by police to the database of images scraped from the internet
• The RCMP acknowledged their use of Clearview AI in several investigations before ceasing its use after privacy concerns were raised in 2020
• Privacy and consent issues:
  • Investigations by federal and provincial privacy commissioners in Canada found the mass AI and scraping practices to be illegal.
• Potential for bias and misuse:
  • Concerns about the accuracy and bias of facial recognition technology, particularly in the potential misidentification of minority groups.
  • The ethical debate has led to broader discussions about the need for regulatory oversight and the establishment of clear guidelines for the use of facial recognition in Canadian law enforcement.

Traffic Jam:

• Developed to help police stop human trafficking and find missing people
• Used by the RCMP for investigations related to human trafficking and child sexual exploitation
• Uses Amazon’s Rekognition technology
• RCMP paused the use of this technology due to public concern

Spotlight:

• Developed by Thorn and co-founded by Demi Moore and Ashton Kutcher
• Provided to law enforcement free of charge
• Uses Amazon’s Rekognition technology
• RCMP paused the use of this technology due to public concern

LECTURE 3 - FINGERPRINTING AI AND AFIS SHANE TURNIDGE

The Data:

• AFIS is the Automated Fingerprint Identification System
• The Canadian AFIS database contains approximately 4 million records
• Each record contains:
  • 10 rolled images
  • 10 plain images
  • 6 palm images
• Each record is stored in two files:
  • 1 image file
  • 1 minutiae map

How to Interpret the Data:

• AFIS is a biometric computer system, it measures
• It attempts to locate and measure the reliable and persistent features within fingerprint and palm print images
  • Ridge endings
  • Bifurcating ridges
  • Large ridge dots
• It then notes their location in X-Y coordinates and their theta vectors, showing the orientation of those features
• Using the X-Y coordinates and theta vectors, every level 2 feature (minutiae) in the image is reduced to a mathematical value
• The physical distances between features are measured through AI technology
• Early AFIS systems used:
  • Digit
  • Patterns
  • Core + delta distances
  • Nearby minutiae values
• The next generation AFIS systems used minutiae only
• Modern systems use third level features to assist in a database search
  • Sweat pores
  • Edge features

The Problems AI Tries to Solve:

• AFIS uses advanced algorithms to search the data in each fingerprint and/or palm print record saved to the database
• No two complete fingerprints have been found to be the same, not even from the same person
• There are subtle differences in each friction ridge record, including those from the same person
• The AI uses advanced mathematics to anticipate these differences while at the same time searching the constellation of features within various large databases
  • Growth
  • Inquiry
  • Deviation/distortion
  • Quality of record
• Double tap/butterfly impressions occur when prints from the same person overlap
  • These two prints can line up and appear to be one trip
• AFIS must be able to accommodate injuries
  • Scar tissue is cement, rigid, and deforms ridge details
  • Cigarette burns and other deliberate injuries
• AFIS must be able to accommodate distortion
  • Ridges expand in the direction of force/pressure

What the AI in AFIS Does:

• Types of searches:
  • Tenprint to tenprint searches
  • Tenprint to latent print searches
  • Palm print to palm print searches
  • Palm print to latent palm print searches
  • Latent fingerprint to latent fingerprint searches
  • Latent palm print to palm print searches
  • Latent palm print to latent palm print searches
• All respondents are searched and sorted based on a score system.
• The user can determine how many respondents to request, or it can be pre-set by an administrator.
• A threshold can be initialized so that poorly ranking respondents are not viewed by an examiner.
• The highest-ranking score value will be placed in the top position with each other respondent placed behind it in numerical value order

Limits of AI in AFIS:

• Ethics:
  • AI does not function ethically just yet
  • AFIS allows for the search of latent friction ridge images with as few as 3 minutiae
  • Resolving a latent print with 3 minutiae is usually outside the skill set of most examiners
• Personal Privacy:
  • AFIS records must be scrutinized by people for legality of use
    • Non-conviction records
    • Young offender records
    • Record suspension

LECTURE 4 - 3D CRIME SCENE DOCUMENTATION GREG SCHOFIELD

CPAP:

• Capture → Process → Analyze → Present
• Capture with:
  • UAVS: Aerial, terrestrial, hybrids
  • Terrestrial Laser Scanning (TLS)
• Process with:
  • Videogrammetry
  • NeRF
  • Point Cloud Mesh
• Analyze with:
  • Temporal
  • Chromo
• Present with:
  • Still and video screen
  • 3DOF and 6DOF Virtual Reality

Robotic Scene Processing:

• SPOT terrestrial robot
  • On top of features previously mentioned, you can mount a laser scanner on it

M4:

• Drives and flies
  • Flying avoids contamination
• Captures video
• Forensic research with drones
  • Scan large fields and scenes
  • Identify bone material
  • Saves time, labour, and eliminates human error

UAV/ALS Remote Search System:

• Equipped a drone with a forensic light source (Crime-lite82s 420-470 blue) and a longpass filter
• Investigated whether it was possible to make human and animal bone material visible from the air (luminication)
• Comparisons were made between searching with people on a line and the drone
• The composite drone found all the bone particles and the people found only a few
• The crime scene remained virgin by the drone at a height of 10 meters

Laser Scanning:

• A laser beam is shone onto a rotating mirror that reflects the beam out towards the area being scanned
• Through the rotation of the mirror, the beam is distributed in a vertical arc of ~300 degrees
• Simultaneous to this vertical rotation, the system rotates horizontally to cover a range of 360 degrees
• The laser beam is reflected back to the scanner by objects/surfaces in its path
• The beam is recorded as a series of pulses or points as the distances as well as their relative vertical and horizontal angles are determined, recorded, and converted into X-Y-Z values
• This data is presented in the form of a point cloud that illustrates the spatial qualities of the objects and surfaces scanned
• Multiple images of the area are scanned and recorded so that specific colour values of individual pixels can be attached to the corresponding scan points

Laser Scanning Procedure:

• Capture:
  • Station points and targets
  • Individual scans - resolution and quality
  • Traversing and control
• Process:
  • Filters, registration, colourizing
• Analysis:
  • Dimensional data
  • Ballistics
  • Bloodstain pattern analysis
  • Suspect height
  • Witness POV
  • Arson/explosions
• Presentation:
  • Screenshots
  • Diagrams/maps
  • Virtual video/ortho imaging
  • 3D printed model
  • Virtual Reality
  • Animations

Virtual Reality:

• 3DOF: Model + video decides the viewpoint and what we are looking at
• 6DOF: You can virtually enter the scene, immersion creates presence

In Summary:

• UAVs offer aerial perspectives that were once inaccessible, enabling rapid data collection and comprehensive scene analysis.
• Laser scanning delivers precise point cloud data, facilitating the creation of detailed 3D models crucial for forensic examinations.
• NeRF technology, driven by deep learning algorithms, synthesizes immersive 3D reconstructions from 2D images, providing unparalleled fidelity in crime scene visualization.
• Videogrammetry extracts 3D information from video footage, aiding in event reconstruction and spatial analysis.

LECTURE 5 - HUMAN COMPUTER INTERACTIONS DAMIAN SCHOFIELD

Human Computer Interaction (HCI):

• The discipline is concerned with the design, evaluation, and implementation of interactive computing systems for human use and the study of the major phenomena surrounding them
  • How is society affected by these technological changes
• There is a direct comparison between architecture and Human Computer Interaction (HCI)
• Architecture:
  • The architects understands the needs of the specific owner and the functions they need to perform in the building
  • The architect designs the plan and shares it with the construction team
  • The owner lives in the house
• HCI:
• The HCI expert understands the needs of the specific user and they functions they need to perform with the software
• The HCI expert designs the software and sends it to a computer engineer to be built
• The computer engineer builds the software
• The user interacts with the software

Sense of Self: An Issue in HCI

• Most people agree that the core components are memories, emotions, and relationships
  • A Google Design Ethicist: The job is to manipulate more people to use google
  • Facebook taps into the instinctive need to be loved and love others. It does this through images to enhance memories, evoke emotions, and strengthen relationships
• Virtual and augmented reality are creating a constantly shifting virtual-real continuum
  • Pokemon Go

Artificial Intelligence:

• Transhumanism: The belief that the human race can evolve beyond the current physical and mental limitations, especially by means of science and technology
• Cyborg: Someone who uses technology to enhance their human ability
  • Glasses and hearing-aids
  • Has a chip in his hand, loads money on his “university card” and opens locked doors by waving his hand
• AI has been around for years:
  • Movie suggestions on Netflix
  • Amazon recommending products, Alexa
  • Washing machines that optimize power usage
• Scary article: Man resurrects childhood imaginary friend (microwave oven) using AI. Then it tried to kill him.
  • Wrote the backstory of a psychopath and “accidentally” trained it to kill him
• His article: Would You Write an AI generated Movie?
  • Reviewed “Salt”, the first fully AI generated movie
  • We are past the point of being able to determine if media is created by humans or not
• Bruce Wilis has sold his image to an AI company (recently diagnosed with dementia and is no longer acting)
• Google Research Lumiere: A Space-Time Diffusion Model for Video Generation
  • You can type in a text prompt that will generate a video
  • You can take an image, add a prompt, and create a video
• Natural language processing:
  • Open AI ChatGPT
  • AI text generators
  • OpenAI Codex is an AI system that translates natural language into code
  • It owes its understanding of natural language to its predecessor, GPT-3, but it’s also been trained on billions of lines of code
  • GitHub Inc. is an internet hosting service for software development and version control
  • GitHub Copilot suggests code and functions in real-time, right from the editor. Trained on billions of lines of code, Copilot turns natural language prompts into coding suggestions across dozens of languages.

Evidential Issues in AI:

• Fake text, images, people, videos
• Current solutions:
  • Chain of custody
  • Verification (experts and new tools to detect these fakes)

LECTURE 6 - PROBABILISTIC GENEALOGY STEVE SMITH

The Abduction and Murder of Christine Jessop:

• November 29th, 1974
• 9 years old at time of didsapearance
• Abduction scene: York Regional Police
• Homicide scene: Durham Regional Police
• Guy Paul Morin was arrested for her murder
• Why Morin?
  • Neighbour, lived close to the Jessop’s
  • Strange fella, socially inept and quiet
  • Didn’t help search for Christine
  • Didn’t attend the funeral
• There was major Tunnel Vision against him in this case
• DNA evidence excluded Morin as the offender

Results of the Kaufman Report:

• Toronto Police to take over the investigation
• 10 person task force established in 1996
• Task force disbanded in 1998

Is Genealogy a Viable Option?

• DNA:
  • How many nanograms do we have?
  • How many nanograms do we need?
  • Is the DNA degraded?
  • Is the DNA a mixture?
• Investigative considerations:
  • Do we have the support of command?
  • Are we willing to pay?
  • Is there a POI list?
  • Has the NDDB profile elimination occurred?
  • Have we done BGA testing?
  • Where was the offender born?
  • Do we have the necessary resources?
  • Do we have genealogists?
• Selecting a lab:
  • Cost
  • Location
  • Package deal?
  • Services needed vs services offered
  • Who are we dealing with?
  • Microarray vs full genome sequencing
• Geneaologists:
  • Are we paying a lab to do our genealogy?
  • Do we have trained genealogists>
  • Who are we paying?
  • 1000s of hours
  • Separation of genealogy and investigation
• The Toronto Police selected the Othram INC lab in Texas
  • Gedmatch results found two distant relatives in the area of Belleville, Ontario

Genealogy Sites:

• Gedmatch and FTDNA
  • Sites that allow law enforcement use
• Data limits
• Informed consent (privacy)
• Terms of service
• Limits on warrantless searches: no other genealogy sites allow police use

Genealogy:

• Creating family trees
• Open source data: death notices, Instagram, Facebook, newspapers
• Police information: CPIC, CNI, MTO
• Target testing : freebies and paid kits
• From 33000 people to 5000 people
• Know your centimorgans

Upload to FTDNA:

• Target tests between 90 and 150 dollars a test
• FTDNA upload $700: need to make an application
• Result: 100 familial matches over 50 centimorgans
• Huge amount of data for genealogists to work with
• Centimorgan: A unit used to measure genetic linkage
  • One centimorgan equals a one percent chance that a marker on a chromosome will become separated from a second marker on the same chromosome due to crossing over in a single generation

Cooking With Gas:

• One family tree led to two families
• One family with 8 kids
  • Father had children with a 14 year old in his late 20s
  • 4 brothers, all convicted of sexual offences
  • One brother committed suicide
• There was no forensic link, onto the first cousins
  • There are two brothers, one committed suicide
  • Calvin Hoover is the target, his brother Brian not so much
  • Police find and address that Calvin used in 1984 linked to his wife at the time, Heather
  • Heather and Calvin Hoover were listed by Janet Jessop as 2 of 5 people who had access to the Jessop house
  • Calvin Hoover was 28 in 1984 and worked with Christine’s dad
  • He committed suicide in 2015 by carbon monoxide poisoning
  • There are two vials of blood from the crime scene stored at the CFS toxicology unit that need to be moved to the biology unit
  • A warrant is needed
  • Calvin Hoover can not be excluded!
  • The autosomal STR DNA results are estimated to be greater than one trillion times more likely to originate from Calvin Hoover than an unknown person unrelated to him
  • 36 years and 6 days later, we have found and named Christine’s killer

LECTURE 7 - CRIME SCENE INVESTIGATION AND BPS IRV ALBRECHT

DNA Evidence:

• Toronto Police Statistics:
  • 2022: 1509 DNA hits
  • 2021: 1314 DNA hits
  • 2020: 1219 DNA hits
  • Convicted offender and crime scene indices
• Two main databases:
  • National DNA Databank in Ottawa run by the RCMP
  • Local DNA database at CFS
  • NDDB contains the Convicted Offender Index and Crime Scene Index
• DNA Evidence Conclusions:
  • Cannot be exlcuded as the source of the DNA profile
  • The STR DNA results are estimated to be greater than X times more likely if the profile originates from ____ than if it originates from an unknown and unrelated person
• Positive outcomes:
  • Increased sensitivity of DNA analysis (smaller samples required)
  • Faster results
  • Better interpretation of mixed profiles
  • Better results from difficult surfaces
• Forensic Genealogy tracing:
  • Killer of Christine Jessop identified
• Challenges:
  • Issues regarding secondary and tertiary transfer of DNA
  • Increased issues with collection (cross contamination)
  • Paper on cut resistant leather slash gloves worn my officers (DNA contamination)
  • Paper on multi-use fingerprint brushes in forensic casework (DNA contamination)

Future of DNA Evidence:

• Single use forensic tools
• Rapid DNA analysis
• Ande rapid DNA analysis system presently in trial and validation phase in ontario through CFS
• Biometrics and Facial Recognition:
  • NEC NeoFace Reveal software
  • Two full-time facial recognition analysts
  • FBI trained
• Future of Facial Recognition:
  • Improved software to increase detection
  • Higher quality images available to search
  • More image sources available

Fingerprint Evidence:

• Latent fingerprint development and discovery
• Collection of known fingerprint impressions
• Fingerprint identification of deceased
• FIngerprint search through AFIS
• Fingerprint comparison
• Presentation of findings
  • Move from ambiguous or misleading language to more accurate explanations
  • Blind verification of comparison process
  • Controls on exposure to contextual information by fingerprint examiners
• Fingerprint research
  • Research and training into deposition and distortion of fingerprints impressions
• Changes to fingerprint collections:
  • Electronic capture of fingerprint impressions
  • Live scan
  • Portable device

Fingerprint Development Techniques:

• Chosen for surface: porous vs non-porous
• Chosen for fingerprint matrix: blood, sweat, dust
• Chosen for contrast/photography
• Powder
• Fluorescent powder
• Cyanoacrylate fuming
  • Porous: tape, leather
  • Non-porous: firearms, plastic bags
• ALS examination:
  • Laser exam
• Forensic laser
• Forensic light source
• Detection and recovery from unusual surfaces:
  • Matte boxes: Laser
  • Blood: Hungarian Red

New Fingerprinting Methods:

• Improved powders
• Vacuum Metal Deposition (VMD):
  • Depositing very thin layers of metal on to a surface in a vacuum chamber to visualize latent fingerprints or touch marks
  • Uses zinc, gold, or silver
  • Can be used on Canadian currency
• Recover system for fingerprints on cartridge cases:
  • Disulfur dinitride fuming process
  • Developed in the UK
  • First system in Canada
  • 12% recovery rate during validation

Improved Trends:

• Improved blood search techniques
• Visual darkness adaptation
  • Crime-lite Eye
  • Modified night vision technology
• Technology-based analysis (AI):
  • Eye tracking study of bloodstain pattern analysts during pattern classification
  • Autmatic classification of bloodstains with deep learning methods
  • Observations and 3D analysis of controlled cast-off stains
  • Automated reconstructions of cast-off patterns based on Euclidean geometry and statistical likelihood
  • HemoVision
  • Blood deposition and drying time estimation
  • Blood pool drying time lapse analysis

Summary:

• Forensic investigation continues to develop and improve through continued input of scientific knowledge and research
• Overall goal remains to provide reliable information to the justice system

LECTURE 8 - VIRTUAL REALITY GREG SCHOFIELD

Basics of Virtual Reality:

• A computer-generated artificial immersive environment experienced through sensory stimuli in which one’s actions may affect what happens in the environment.
  • Primarily sight and sound stimuli
  • The technology used to create or access a virtual reality environment

The Immersive Environment:

• In training, the effectiveness of VR is a direct function of the fidelity of its immersive environment
• Fidelity:
  • To the visuals of the actual situation (how realistic are the representations)
  • To the task (how accurately is the task modeled and presented)
  • To the cognitive path (how closely do the mental processes experienced and retained reflect the real-world process)
• Perceptual Fidelity: Virtual interactions that closely mimic the physical world activate the same neural pathways in the brain.
  • Muscle memory or state-dependent retention
• Feedback fidelity: In VR, learners make decisions just as they would in the real world, and depending on the level of engagement, these decisions can have direct and immediate positive or negative impact.
• Emotional fidelity: In VR, can invoke a sense of presence that creates real emotional and empathetic responses. These sensory stimulations trigger the brain in much the same way it reacts to actual situations to release endorphins, serotonin and/or dopamine.
  • Triggers feelings of joy, fear, surprise…

The Goal of Immersive Environment:

• Agency (the degree of choice, interaction, ownership, and/or control) along with the immersive environment generates presence (the feeling of actually being there, interacting, and sharing the experience) which in turn encourages engagement.
• Engagement when combined with fidelity determines the success of the training process
  • Agency + Immersive Environment = Presence
  • Presence encourages Engagement
  • Engagement + Fidelity = Success

Additional Points About the Immersive Environment:

• Suspension of Disbelief (SoD): The first steps in immersion and the formation of Presence involve Suspension of Disbelief
  • Issues that break presence can be very jarring and make renewed SoD difficult
• Persistence: To develop a deeper investment in the virtual world, participants need to feel it is persistent and consistent.
  • In turn, this creates trust in the system, builds community to encourage social norms, and fosters co-creativity
• The Narrative: Immersion is significantly enhanced by a compelling and comprehensive narrative. It contextualizes and helps make sense of the activity and tasks being encountered.
• Decisions to be made in selecting a system:
  • Content: Off the shelf, In-house/custom
  • Degrees of Freedom: 3 DoF or 6 DoF
  • Guidance: Implicit or Explicit (self-discovery or directed learning)
• At all stages, whether planning, selecting, deploying, or evaluating, prioritize the pedagogy over the technology.

Degrees of Freedom:

• VR headsets with 3DoF only have rotational control. You cna look around but not interact with the environment or dictate where to go.
• VR headsets with 6DoF have both rotational and translation control. Subjects have control over interacting with the scenario and can use haptic technology to enhance the experience

VR for Training:

• Subjects for:
  • Recruiting
  • Screening
  • On-boarding
  • New skill development
  • Introducing new processes and techniques
  • Soft skills
  • In-service requalification
  • Testing
• Anyone except the 10% who can’t see in VR are heavily susceptible to VR sickness
• DICE Scenarios:
  • Dangerous: Low frequency, high-risk situations
  • Impossible: Microscopic, outer-space, super-human abilities, time travel
  • Counter-productive: Complex activities requiring repetitive action cycles to experience, reflect on, and correct erroneous performance
  • Expensive: Offsite travel, physically dispersed staff, saged scenarios requiring role-playing actors and resource consumption
• Role playing/rehersals
• Swapping roles
• Developing and practicing empathy
• Considering ideas from multiple perspectives
• Considering multiple alternatives, including unlikely, unpopular, and impossible scenarios
• Where:
  • At existing training centers
  • Dispersed using 5G and streaming
• Pros:
  • Flexibility in accessing training when convenient
  • Provides the ability to explore in private
  • Asynchronous or synchronous as needed
  • Together in a team setting when mutually convenient
  • Independent as time permits
  • Can be used as a preparatory stage working towards a final hands-on/real-world exam
  • As a way of experiencing a wider variety of subject situations or scenarios
• Cons:
  • Delayed feedback from instructor
  • Delayed tech help
• Why VR training?
  • Improved comprehension, internalization, retention and confidence
  • In 2018, Walmart Academy added an active shooter module to its VR employee training program and a year later the was a mass shooting in an El Paso store. Employees provided feedback that the training saved people’s lives.
• Why not VR?
  • Costs
  • Vendor issues
  • Health-related issues
  • Technical issues

TRIP PRESENTATION AND RECURRING THEMES