So taking the FFT scans from yesterday and using it to measure the stitchers gauge and doing the same with the rest of the 70+ samples I received you get the following really interesting graph showing allot of spread between the stitchers.

After looking at this, I wondered if these parameters meant something.  So I took out the samples and quickly grabbed the ones that exemplified the professional and amateur.  I then looked up their stitcher number and found which data point on the graph these represented.  What I found was that there was a zone where only amateurs lay and only professionals lay.  These zones were separated by an area of transition.

Tricia

Ok – so that was a set of terms I never thought in my life I would speak in this audience.  But there is it.  I will let those who are very interested look it up on Wikipedia.  Another good place to look is a lecture by Stanford Univeristy on-line if you want even more.  It is a reasonably layman’s lecture in the first 10 minutes.  The analog here is that our buttonhole net is like a crystal lattice.

In general, a periodic structure (in this case the buttonhole net) can be analyzed by Fourier Transforms that converts the frequency of the repetition in the image into a point on a pattern.  The distance of this point from the center of the pattern is the inverse of the distance in the real image. Its x-y position is also related to the direction of the periodicity.  Therefore, you can get precise measurements of this periodicity.  In our case, the distances between the rows.  Now if there is variability to the row spacing, the intensity of the point will reduce as the number of frequencies at that distance are less.  On the point pattern, the point will get more diffuse and less intense.  A measurement of this peak size can tell you how disordered the stitching was.

So I took the samples that you all sent in and optically scanned them.  I assigned numbers to the stitchers to protect the innocent and represent the data that way.  Here you see a series of FFT images of the stitching.  Each one shows the actual image that was processed in the bottom corner.   What you can see is that there are allot of differences between the FFTs for the stitched pieces.  In general you can say:

- If there are allot of points, the embroidery was very regular

- If the points are small and sharp, the embroidery was very regular

- If the points show us as an arc around the circumference, this often shows that the stitcher followed the curving outline of the shape but was usually pretty regular if those arcs are intense.

- If there are few points, the embroidery didn’t have allot of regularity

- the differences in distances of the peaks from one person to another show stitching gauge differences and the angle of them from each other also show net density differences in the real pieces.

When I showed about 30 of these patterns stacked on top of each other and animated to change to a roomful of diffraction experts at UofM, I had them convinced immediately that there are measurable differences between people and their stitching.  What do you think?  More tomorrow.

Tricia

Yesterday’s post came up just as I was returning from the University of Michigan from giving my talk at the 40th anniversary of the Electron Microbeam and Analysis Laboratory Symposium.  What a surreal experience.  Essentially I spoke in between talks entitled “Biomineralization of Extracellular Matrix Proteins Probed at the Nanoscale by Synchrotron and Electron Imaging Methods” and “Effect of Si-doping on the Evolution and Growth of Stress and Threading Dislocations in AlGaN Films Grown by Metalorganic Chemical Vapor Deposition”. I laughed that I was the comedic interlude – a bit of light brain candy to break up the day.

I joke.  Actually, the audience was really engaged.  It was a big risk for me to give a talk on embroidery as the group included all the scientists on my dissertation committee, my advisor, and many of my colleagues who know me as a polymer physicist.  (you didn’t know of my alter ego – did ya!).  I had to spend about 15 slides familiarizing them with 17th century costume history and the Plimoth project in order to set up the experimental data.

Essentially what I have been working on is using an analysis technique called Fast Fourier Transforms which converts periodic images into a point pattern (this is the function that describes xray patterns of crystals).  The location of the points on the pattern are related to the spacings in the image and can be measured very accurately.  In the case of the detached buttonhole embroidery, I can use the points to measure the gauge of the rows and buttonhole stitches.  Then the intensity of the peaks and the shape also tell you how perfect the detached buttonhole was worked.

My ‘thesis’ is that each individual has a characteristic guage and perfection.  Of course, as the person gains more and more experience I think that they change and approach some steady state parameters.

Since my thesis has now been vetted by a group at the MET (I gave a version of it to them in May) and a group of the formost authorities in the technique – I feel more comfortable putting the data out here in the blog.  I’ll post some of it over the next few days.

Tricia

Kris Andrews pitched in to help me out tremendously in late September.  She worked to finish off the few spangles left, any gold someone may have missed and to attach the last of the detached wings and pea pods.  But before she could do that – she helped me scan the entire jacket.  Some about why tomorrow.

Here are some pictures of her working as well as showing the detached pieces going onto the jacket pieces.

Tricia

So we convinced Siu Ying to add some spangles for us – Linda E. taught her.  And then we had this group of family members and friends come in.  They were a hoot.  Pretty funny – egging each other on to work on the jacket.  Poor teenage son who was along ended up sitting there with his phone while they all got their spangles on the jacket.

Another visitor we had was Austrian and was living in the USA.  On a cross country tour with her parents who were visiting from Austria.  Ruth (the daughter) spent awhile working on the jacket while her parents and I talked about embroideries from their country (I had been there last summer).   It was lovely to see how the embroiderers who were at each frame would invite visitors to the building to help out and get up to instruct them.  Sharing a love of embroidery all around.

Tricia