Weekly Update for October 22

class: center, middle, inverse, title-slide

# Weekly Update for October 22
### 2018/10/22

---

class: inverse
# About Slides

---
class: primary 
# New CSAFE slide template

We're now using [`xaringan`](https://github.com/yihui/xaringan)

What's changed:

- New person slide:

````
---
class: inverse
# Your Name
````

---
class: secondary

- New content slide with title:

````
---
class: primary
# Title of slide

Slide content
````

- New content slide without title:

````
---
class: secondary

Slide content with no title on slide
````

---
class: inverse
# Sample User

---
class: primary
# Sample Slide

- Sample Table. Sam Tyner talked about the three must-haves of your summary:

---
class: secondary

- If you are going to add an image, **create a directory** with your name within "images" folder. For example, "images/**guillermo**/sample_image.JPG"

<img src = "images/guillermo/sample_image.JPG" width="35%"/>
 
---
class: inverse
# Susan

---
class: primary
# Last Week

- Set up participant recruitment for truthiness
- Writing `bulletsamplr` paper
    - Figured out how to bootstrap matching bullet signatures
- Submitted revised Statistical Atlas paper
- Added crosscut to `image_x3p` in x3ptools package

---
class: inverse
# Miranda

---
class: primary
# Neural Networks

- Working on poster for AAFS and Forensics@NIST
- Heatmaps to visualize important image features
    - Track predictions across sections of a whole shoe image
    - Restructuring model training method to connect pre-trained base to our classifier
- More visualizations...

---
class: primary
# ROC

---
class: inverse
# Soyoung

---
class: primary
# Fall semester plans: 
 - Shoes
    + Wrap up shoe analysis using edges and SURF 
    + Compare the performance with other methods such as phase only correlations, Fourier-Mellin transformation correlation
    + Do analysis with CSAFE shoe data
    + Submit papers regarding shoe analysis soon!
   
- Research on "uncertainty pyramid" with Steve and Hari at NIST   
 - Glass analysis with Sam
    + Exploratory analysis 
    + Various covariance estimations
    + Submit papers!
   
- Working on book chapters of glass and shoes funded by ROpenSci Fellowship

---
class: primary
# Analysis with CSAFE shoe data 
- POC is rotation variant!
- Try to find ways to adjust rotation bewteen two images
 <img src="images/soyoung/ex_rotation.PNG" width="85%">

---
class: primary
# Glass Analysis  
- Comparison between Standard method vs. Random forest 
- When two methods disgree,

|   | False neg.  | False pos.| True neg.  | True  pos.  |
|:-:|:-:|:-:|:-:|:-:|
|std| 92  |7635   |2299   |9   |
|RF | 9  |2299   |  7635 |92   |

- For false positives, std uses K (27.81%), Li (18.62%), Zr (14.43%), Ba (8.98%), Ti (6.72%)

---
class: inverse
# Amy

---
class: primary
# Works in progress..

- Data
 - FlashID - Updates to data dump
 - Hoping this will help with model performance. 
 
- Hierarchical model evaluation
 - Coding: ROC/AUC (multi-class extensions, one vs. all)
 - Coding: ABC for model selection 
 - Fscore (micro and macro averaging)
 - Log-loss for multi-class classification
 
 
---
class: inverse
# Ben

---
class: primary
# Semester Intentions
<img src = "images/ben/stack.jpg" width="35%"/>
- Collaborate w/Nic & Amy to improve handwriting project
- Develop improvements / features part of algorithm
- Eventually port algorithm through an application portal
- Improve file structure of project
- Working on preprocessing binary image
 - Prototyping w/ Numba & Python
 - Testing performance relative to Cpp
 - Hope to expedite development

---
class: inverse
# Nick

---
class: secondary

- Working on preprocessing binary image
  - Prototyping w/ Numba & Python
  - Testing performance relative to Cpp
  - Hope to expedite development

---
class: inverse
# Ganesh

---
class: secondary

- The Chumbley Land-to-land comparisons Paper got accepted!!
- Now setting Bullet-to-Bullet Comparisons and writing about the UI..

---
class: inverse
# Nate

---
class: primary
# Groove ID

- Writing vignette/drafting methodological description
- Work on integrating changepoint functions with bulletxtrctr(?) this week

---
class: primary
# SLR

---
class: secondary

---
class: secondary

---
class: inverse
# Sam

---
class: primary
# Fall semester plans:

- submit papers from thesis for publication
    + **Chapter One submitted!**
- glass work with SP 
    + **NIST poster accepted**
    + working on a paper for Journal of Forensic Sciences **(template?)**
- working on book funded by ROpenSci Fellowship 
- writing "Ten Simple Rules for..." articles with DO. **meeting tomorrow**
    + statisticians doing forensic science
    + forensic scientists doing statistics
    + **legal professionals encountering statistics - have a very rough draft of rules.** 
    + **Chatted with SP from ASA about it last Friday. He is very supportive.**
    
---
class: primary
# Last week

Went to the Committee on National Statistics offices and 137th meetings.

What I learned: 
  - Turns out there's more to federal statistics than the Census Bureau! 
  - Current administration's talk of reorganizing the federal statistical agencies is causing [uproar](http://magazine.amstat.org/blog/2018/08/01/usdafoot/) in the community
  - Bureau of Transportation Statistics has an Office of Spatial Analysis and Visualization
  - The government has A LOT of data out there. Just have to know where to look! (e.g. The U.S. Energy Information Administration (EIA) API is the source of gasbuddy.com)
    
Inspired to give a talk to Graphics group on Nov. 16th @ 2:10pm about the many data sources available from the US federal government.

---
class: inverse
# Guillermo

---
class: primary
# Plans for this Fall

- Complete adaptation of registration of 3D scans in R and perform analysis

- Complete shiny app for 2D outsole scans registration

- Write paper of the database which must include some data analysis

- Write paper on speaker recognition (w/Vianey Leos)

---
class: inverse
# James

---
class: primary
# Data Collection

* Estimate 10-13 hours per- 3 pairs
   + 2D Scanning
   + 3D Scanning (High Definition)
   + High Resolution Photos
   + 3D Mud Scans
   + Mud Photos
   + Casting 
   + Baby Powder Photos
   
* We are contacting an expert for input
* estimate >17 weeks to finish all shoes in their current condition

---
class: primary
# Data Collection

<img src = "images/James/Path.png" width="45%"/>
<img src = "images/James/CastStudy.png" width="45%"/>
* Baby Powder Prints - 342
* Casting - 6 
 + 334 Casts total 
* Cast Photos - 4
* 3D Mud Scans - 15
* Mud Photos - 21

---
class: inverse
# Kiegan

---
class: primary   
# Fall Semester Goals

- Submit to AFTE journal  
- Complete writing of Chapman & Hall book  
    - Working on this from home T/Th  
- Write a joint paper with Nate on grooves project  
    - Probably aim for Journal of Forensic Sciences  
    - Compare/contrast of methodologies
    - Downstream impact on the algorithm!  
- Taking one class (Stat 643) and research credits

---
class: inverse
# Danica

---
class: primary   
# Fall Semester Goals

- Submit draft BF vs. LR paper to AOS  
    - Write a follow-up paper to LPR  
- Continue NIJ grant to validate FDE conclusions  
- Work with Amy on the CSAFE Handwriting project  
- Work with Dan Spitzner from UVA on the CSAFE Statistical Foundations project
- Write a paper for LPR on approximations to BFs
- Write a paper on Fiducial Factors with UNC

---
class: inverse
# Heike

---
class: primary   
# x3p images

Adding helper lines on x3p images

![](images/heike/x3p-helper.png)

Might help with creating a mask for different areas of the file
![](images/heike/x3p-mask.png)

---
class: inverse
# Issues

---
class: secondary

- [Issues!!](https://github.com/CSAFE-ISU/slides/issues)
- Two issue down, two to go.