Weekly Update for October 15

class: center, middle, inverse, title-slide

# Weekly Update for October 15
### 2018/08/15

---

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**/handwritten_logo.png"

---
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: inverse
# Amy

---
class: primary
# Fall Semester Goals

- **Data Collection**
 - Moving (slowly).
- **FlashID Data**
 - Update install scheduled for tomorrow morning!
- Expand Bayesian hierarchical model
 - **Formal Model Selection**
 - Cleaning up code.
 - Methods other than DIC.
 - Tests and Intervals
 - Validation
- Write and submit a paper on that modeling.
- Exploritory analysis of 'next level' features (AAFS)
 - Work with Ben and Nick
 
---
class: inverse
# Ben

---
class: primary
# Preprocessing Complete!
- Next steps: Character recognition? Other feature detection?

---
class: inverse:
<img src = "images/ben/goodwordmessage_9oct.png" width="100%"/>

---
class: inverse
# Nick

---
class: secondary

- Worked with Ben on preprocessing -- we saw his slides.
- No update to making node counts more intuitive

---
class: inverse
# Ganesh

---
class: primary   
# UI

- Shiny App Demo

---
class: inverse
# Nate

---
class: primary
# Groove ID

- Documented functions necessary to run changepoint analysis
- Working on a vignette for the R package

---
class: primary
# SLR

- Documenting what I've tried so far
- Small examples to compare SLRs and LRs
- Do well-calibrated SLRs have properties that imply they are close to true LRs with high probability?

---
class: inverse
# Sam

---
class: primary
# Fall semester plans:

- submit papers from thesis for publication
- writing R packages with SP
    + glass 
    + shoes
    + "uncertainty pyramid" stuff with SL & HI
- working on book funded by ROpenSci Fellowship 
- narrowing down topics & formats for new CSAFE training materials 
- writing "Ten Simple Rules for..." articles with DO
    + statisticians doing forensic science
    + forensic scientists doing statistics
    + [legal professionals encountering statistics](https://github.com/CSAFE-ISU/slides/issues/2)

---
class: inverse
# Guillermo

---
class: secondary

- **IRB** The saga continues (Applting for continuing review).

- **Database** Added function to `checkshoenames` to find a specified pattern in the file names in a directory and change the name of such files.

- **Registration** Reading papers (suggested by Susan) and testing functions to register matscan images.
    - Idea: Borrowing ideas from fMRI images to find a template image within another image.

---
class: inverse
# James

---
class: primary   
# Back at Data Collection!

---
class: inverse:

---
class: inverse:

<img src = "images/James/LightRoom.png" width="32%"/>
<img src = "images/James/LightUp.png" width="32%"/>
<img src = "images/James/LightSide.png" width="32%"/>
 
---
class: inverse
# Kiegan

---
class: primary   
# Chugging along!

- **Submit to AFTE journal**  
    - Getting close!  
- Complete writing of Chapman & Hall book  
    - On pause so we can focus on getting 2 papers ready  
- **Write a joint paper with Nate on grooves project**  
    - Integrating in to `bulletxtrctr` this week
    - Getting test sets ready
- **Submit proposal to DFSC for funding**
    - Reworking NIJ Grad Fellowship proposal
    - Incorporating peer review comments

---
class: primary 
# Update to plotting results 
 
<img src = "images/kiegan/boxplots_afte_results.png" width="65%"/>

---
class: primary 
# Update to plotting results 
 
<img src = "images/kiegan/stacked_bars_afte_results.png"/>

---
class: inverse
# Danica

---
class: primary   
# Fall Semester Update

- Write a BF vs. LR paper for LPR  
    - working on making second draft less technical
    - this is taking longer than I expected :(
- Continue NIJ grant to validate FDE conclusions  
    - poster presentation for Forensics@NIST next month
    - oral presentation accepted for AAFS Feb 2019
- New NIJ grant awarded!
    - working on award acceptance and budgets
- 10 Things article
    - writing a draft complete with paragraphs and references

---
class: inverse
# Susan

---
class: primary
# Write all the things!

- `bulletsamplr` package 
    - Determined there's not a significant change in groove height along the signature
    - Started examining null feature distribution
    - Paper describing method for resampling bullet sequences
- Paper validating bullet algorithm on different case study sets (w/ Heike)
- Revising Statistical Atlas paper

---
class: inverse
# Miranda

---
class: primary
# Neural Networks

- Working on poster for Forensics@NIST and AAFS
- How to best quantify accuracy for one-hot prediction?
    - ROC-type curve for probability threshhold
- Generating heatmaps to visualize what neural networks look for
    - Generating these for our model is difficult
    - So far, can only generate from original VGG16 (see next slide...)

---
class: primary
# VGG16 Prediction

---
class: inverse
# Heike

---
class: primary
# CSAFE persistence bullets

![](images/heike/B902-L1-20x.png)
![:scale 40%](images/heike/B902-L1-50x.png)

---
class: primary
# CSAFE persistence bullets

Signatures from 20 and 50x magnification

![](images/heike/signatures-raw.png)

---
class: primary
# CSAFE persistence bullets

Aligned Signatures from 20 and 50x magnification

![](images/heike/signatures-raw-aligned.png)

---
class: inverse
# Soyoung

---
class: primary
# Analysis result with CSAFE shoe data 
- POC is rotation invariant!
- Known mated pairs
 <img src="images/soyoung/rot_KM.png" width="60%">

---
class: primary
# Analysis result with CSAFE shoe data 
- POC is rotation invariant!
- Known non-mated pairs
 <img src="images/soyoung/rot_KNM.png" width="60%">

---
class: inverse
# Issues

---
class: secondary

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