Home Page Icon
Home Page
Table of Contents for
Title Page
Close
Title Page
by Scott Krig
Synthetic Vision
Cover
Title Page
Copyright
Contents
Preface
Chapter 1: Synthetic Vision Using Volume Learning and Visual DNA
Overview
Synthetic Visual Pathway Model
Visual Genome Model
Volume Learning
Classifier Learning and Autolearning Hulls
Visual Genome Project
Master Sequence of Visual Genomes and VDNA
VGM API and Open Source
VDNA Application Stories
Overcoming DNN Spoofing with VGM
Inspection and Inventory Using VDNA
Other Applications for VDNA
Background Trends in Synthetic Intelligence
Background Visual Pathway Neuroscience
Feature and Concept Memory Locality
Attentional Neural Memory Research
HMAX and Visual Cortex Models
Virtually Unlimited Feature Memory
Genetic Preexisting Memory
Neurogenesis, Neuron Size, and Connectivity
Bias for Learning New Memory Impressions
Synthetic Vision Pathway Architecture
Eye/LGN Model
VDNA Synthetic Neurobiological Machinery
Memory Model
Learning Centers and Reasoning Agent Models
Deep Learning vs. Volume Learning
Summary
Chapter 2: Eye/LGN Model
Overview
Eye Anatomy Model
Visual Acuity, Attentional Region (AR)
LGN Model
LGN Image Assembly
LGN Image Enhancements
LGN Magno and Parvo Channel Model
Magno and Parvo Feature Metric Details
Scene Scanning Model
Eye/LGN Visual Genome Sequencing Phases
Magno Image Preparation
Parvo Image Preparation
Segmentation Rationale
Saccadic Segmentation Details
Processing Pipeline Flow
Processing Pipeline Output Files and Unique IDs
Feature Metrics Generation
Summary
Chapter 3: Memory Model and Visual Cortex
Overview
Visual Cortex Feedback to LGN
Memory Impressions and Photographic Memory
CAM Neurons, CAM Features, and CAM Neural Clusters
Visual Cortex and Memory Model Architecture
CAM and Associative Memory
Multivariate Features
Primal Shapes, Colors, Textures, and Glyphs
Feature VDNA
Volume Feature Space, Metrics, Learning
Visual DNA Compared to Human DNA
Spatial Relationship Processing Centers
Strand and Bundle Models
Strand Feature Topology
Strand Learning Example
Bundles
Visual Genome Sequencing
Visual Genome Format and Encodings
Summary
Chapter 4: Learning and Reasoning Agents
Overview
Machine Learning and AI Background Survey
Learning Models
Training Protocols
Reasoning and Inference
Synthetic Learning and Reasoning Model Overview
Conscious Proxy Agents in the PFC
Volume Learning
VGM Classifier Learning
Qualifier Metrics Tuning
Genetically Preexisting Learnings and Memory
Continuous Learning
Associative Learning
Object Learning vs. Category Learning
Agents as Dedicated Proxy Learning Centers
Agent Learning and Reasoning Styles
Autolearning Hull Threshold Learning
Correspondence Permutations and Autolearning Hull Families
Hull Learning and Classifier Family Learning
Autolearning Hull Reference/Target Differences
Structured Classifiers Using MCC Classifiers
VDNA Sequencing and Unique Genome IDs
Correspondence Signature Vectors (CSV)
Alignment Spaces and Invariance
Agent Architecture and Agent Types
Custom Agents
Master Learning Controller: Autogenerated C++ Agents
Default CSV Agents
Agent Ecosystem
Summary
Chapter 5: VGM Platform Overview
Overview
Feature Metrics, Old and New
Invariance
Visual Genomes Database
Global Unique File ID and Genome ID
Neuron Encoder and QoS Profiles
Agent Registry
Image Registry
Strand Registry
Segmenter Intermediate Files
Visual Genome Metrics Files
Base Genome Metrics
Genome Compare Scores
Agent Management
Sequencer Controller
Correspondence Controller
Master Learning Controller (MLC)
CSV Agents
Correspondence Signature Vectors (CSVs)
Group Metric Classifiers (GMCs)
Strand Topological Distance
Interactive Training and Strand Editing
Metric Combination Classifiers (MCCs)
MCC Function Names
MCC Best Metric Search
Metric Combination Classifier (MCC) Summary
VGM Platform Controllers
Image Pre-Processing and Segmenter: lgn
Genome Image Splitter: gis
Compute Visual Genomes: vg
Comparing and Viewing Metrics: vgc
Agent Testing and Strand Management: vgv
Summary
Chapter 6: Volume Projection Metrics
Overview
Memory Structure: 3x3 vs. 3x1
CAM Feature Spaces
CAM Neural Clusters
Volume Projection Metrics
Quantization Space Pyramids
Strand CAM Cluster Pyramids
Volume Metric Details
Volume Impression Recording
Volume Metrics Functions
Volume Metrics Memory Size Discussion
Magno and Parvo Low-Level Feature Tiles
Realistic Values for Volume Projections
Quantized Volume Projection Metric Renderings
Summary
Chapter 7: Color 2D Region Metrics
Overview
Background Research
Color Spaces
RGB Color
LUMA, RGBI, CIELab Intensity
HSL Hue and Saturation
Eye Model Color Ranging
Squinting Model and Sliding Histograms
Sliding Contrast over Cumulative Histograms
Sliding Lightness over Normal Histograms
Sliding Metrics, Centroid, and Best Match
Static Color Histogram Metrics
LGN Model Color Leveling
Color Level Raw
Color Level Centered
Color Level CIELab Constant
Color Level HSL Saturation Boosting
LGN Model Dominant Colors
Leveled Histogram Distance, Moments
Popularity Colors
Standard Colors
Color Metrics Functions
Summary
Chapter 8: Shape Metrics
Overview
Strand Topological Shape Metrics
Single-Image vs. Multiple-Image Strands
Strand Local Vector Coordinate System
Strand Vector Metrics
Strand Set Metrics
Strand Shape Metrics: Ellipse and Fourier Descriptors
Volume Projection Shape Metrics
Statistical Metrics
Ratio Metrics
Genome Structure Shape Metrics
Genome Structure Local Feature Tensor Space
Genome Structure Correspondence Metrics
Shape Metric Function List
Summary
Chapter 9: Texture Metrics
Overview
Volume Projection Metrics for CAM Clusters
3x1 RGBI Component Textures
3x3 RGB Textures
Volume Metric Distance Functions
Haralick Features
Haralick Metrics
SDMX Features
SDMX Metrics
Haralick and SDMX Metric Comparison Graphs
Texture Similarity Graphs (Match < 1.0)
Texture Dissimilarity Graphs (Nonmatch > 1.0)
MCC Texture Functions
CSV Texture Functions
Summary
Chapter 10: Region Glyph Metrics
Overview
Color SIFT
Color Component R,G,B,I G-SURF
Color Component R,G,B,I ORB
RGB DNN
MCC Functions for Glyph Bases
Glyph Base CSV Agent Function
Summary
Chapter 11: Applications, Training, Results
Overview
Test Application Outline
Strands and Genome Segmentations
Building Strands
Parvo Strand Example
Magno Strand Example
Discussion on Segmentation Problems and Work-arounds
Strand Alternatives: Single-image vs. Multi-image
Testing and Interactive Reinforcement Learning
Hierarchical Parallel Ensemble Classifier
Reinforcement Learning Process
Test Genomes and Correspondence Results
Selected Uniform Baseline Test Metrics
Test Genome Pairs
Compare Leaf : Head (Lo-res) Genomes
Compare ront Squirrel : Stucco Genomes
Compare Rotated Back : Brush Genomes
Compare Enhanced Back : Rotated Back Genomes
Compare Left Head : Right Head Genomes
Test Genome Correspondence Scoring Results
Scoring Results Discussion
Scoring Strategies and Scoring Criteria
Unit Test for First Order Metric Evaluations
Unit Test Groups
Unit Test Scoring Methodology
MATCH Unit Test Group Results
NOMATCH Unit Test Group Results
CLOSE Unit Test Group Results
Agent Coding
Summary
Chapter 12: Visual Genome Project
Overview
VGM Model and API Futures
VGM Cloud Server, API, and iOS App
Licensing, Sponsors, and Partners
Bibliography
Index
Search in book...
Toggle Font Controls
Playlists
Add To
Create new playlist
Name your new playlist
Playlist description (optional)
Cancel
Create playlist
Sign In
Email address
Password
Forgot Password?
Create account
Login
or
Continue with Facebook
Continue with Google
Sign Up
Full Name
Email address
Confirm Email Address
Password
Login
Create account
or
Continue with Facebook
Continue with Google
Prev
Previous Chapter
Halftitle Page
Next
Next Chapter
Copyright
Add Highlight
No Comment
..................Content has been hidden....................
You can't read the all page of ebook, please click
here
login for view all page.
Day Mode
Cloud Mode
Night Mode
Reset