#### Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

Options

# Introduction

edited September 2018 in Chat

I would like to try to frame my research considering the applied category theory. I´m available for scientific collaboration and tech discussion in general.

• Options
1.

Hello, Pierre! It's good to see you here. It sounds like you do really interesting practical stuff. How do you relate digital image processing and computational intelligence to environmental science? I can think of many possible ways, but I'm curious about what you actually do in the realm of environmental science.

Comment Source:Hello, Pierre! It's good to see you here. It sounds like you do really interesting practical stuff. How do you relate digital image processing and computational intelligence to environmental science? I can think of many possible ways, but I'm curious about what you actually do in the realm of environmental science.
• Options
2.
edited April 2018

Hi Professor Baez!

I´ve been working on automatic classification/descriptors of images of bark of trees, for automatic tree identification purpose, with some perpective focus on future UAV acquisition of this type of images. Also, I´ve been exploring automatic classification/descriptors of aerial (optical/airplane acquisition) images to allow to map different land uses, looking for minimizing human experts efforts/define optimal set of descriptors . Additionally, I´ve been exploring possible relations between radar (ASAR) images features and the task of quantification of water pollution by surfactants resultant from cities/industry activity. The classifier I´ve been using is the Fuzzy ARTMAP one.

Best regards, Pierre

Comment Source:Hi Professor Baez! I´ve been working on automatic classification/descriptors of images of bark of trees, for automatic tree identification purpose, with some perpective focus on future UAV acquisition of this type of images. Also, I´ve been exploring automatic classification/descriptors of aerial (optical/airplane acquisition) images to allow to map different land uses, looking for minimizing human experts efforts/define optimal set of descriptors . Additionally, I´ve been exploring possible relations between radar (ASAR) images features and the task of quantification of water pollution by surfactants resultant from cities/industry activity. The classifier I´ve been using is the Fuzzy ARTMAP one. Best regards, Pierre 
• Options
3.

Thanks, that's interesting. My former student Blake Pollard did some work at Columbia classifying satellite photos of cropland, but not with sophisticated machine learning.

Comment Source:Thanks, that's interesting. My former student Blake Pollard did some work at Columbia classifying satellite photos of cropland, but not with sophisticated machine learning.
• Options
4.
edited April 2018

Thanks for your interest, this image/vision/computational intelligence field of research is really intriguing..These cropland images processing are very important . I will look for exchange some ideas with Blake Pollard. Do you see plausible some approach unifying/duality between applied category theory/networks and some analog of a QRcode/binary or a gray level texture? best

Comment Source:Thanks for your interest, this image/vision/computational intelligence field of research is really intriguing..These cropland images processing are very important . I will look for exchange some ideas with Blake Pollard. Do you see plausible some approach unifying/duality between applied category theory/networks and some analog of a QRcode/binary or a gray level texture? best
• Options
5.

I don't know enough about how people process and analyze images to know how category theory might be applied. Brendan Fong and David Spivak have a paper on category theory and supervised learning algorithms. This is just the first step in what might become a very big subject.

Comment Source:I don't know enough about how people process and analyze images to know how category theory might be applied. Brendan Fong and David Spivak have a paper on [category theory and supervised learning algorithms](https://arxiv.org/abs/1711.10455). This is just the first step in what might become a very big subject.
• Options
6.
edited April 2018

Very interesting reference on a first sight, I will study it carefully, thanks a lot. I've been working on a draft that perhaps it will be of interest of yours but I ought have to send it by email to you with the co-author (not participant in this forum) in carbon copy. best

Comment Source:Very interesting reference on a first sight, I will study it carefully, thanks a lot. I've been working on a draft that perhaps it will be of interest of yours but I ought have to send it by email to you with the co-author (not participant in this forum) in carbon copy. best
• Options
7.

Hi @Pierre, serendipity rules. I happen to want to find out more about remote sensing of soil moisture. This is for 2 friends: one working for an NGO with farmers in sub-Sahara, East Africa and Bangladesh and the other working for the UN on disaster prevention in East Timor where we're interested in landslide probabilities. I've only got as far as having a preliminary browse through some of the MODIS Terra datasets. I'll definitely check out what Blake Pollard did. It would be great if you started a new discussion on the forum if you want to chat more. :)

Comment Source:Hi @Pierre, serendipity rules. I happen to want to find out more about remote sensing of soil moisture. This is for 2 friends: one working for an NGO with farmers in sub-Sahara, East Africa and Bangladesh and the other working for the UN on disaster prevention in East Timor where we're interested in landslide probabilities. I've only got as far as having a preliminary browse through some of the MODIS Terra datasets. I'll definitely check out what Blake Pollard did. It would be great if you started a new discussion on the forum if you want to chat more. :)
• Options
8.
edited April 2018

Hi @JimStuttard, in this theme I would suggest to start by look for SMOS sensor/ TDVI index information on this links, this team has expertise in this theme.

https://www.uv.es/elopez/?24 Feel free to send messages..so far i have any clue to start a topic but let me know if you start one i could try to contribute somehow..best

Comment Source:Hi @JimStuttard, in this theme I would suggest to start by look for SMOS sensor/ TDVI index information on this links, this team has expertise in this theme. https://www.uv.es/elopez/?24 Feel free to send messages..so far i have any clue to start a topic but let me know if you start one i could try to contribute somehow..best
• Options
9.

Thanks for the link: just the sort of thing I need to know. I'll start a chat page when I've read some more. :)

Comment Source:Thanks for the link: just the sort of thing I need to know. I'll start a chat page when I've read some more. :)
• Options
10.
edited April 2018

Hi all, I did some work a number of years ago training a 2-stage classification/regression model to distinguish irrigated crops in northwestern India during the Rabi (dry) season. We used logistic regression in the first stage to basically throw out everything that was obviously not irrigated (deserts, bodies of water, cities, etc.). The second stage assigned some number between 0-1 which was meant to roughly reflect the percentage of a pixel that was irrigated. This second stage was trained to distinguish forest from farm etc. and was trained using labels I generated by staring at lots of time stamped high-res satellite images.

Our raw data sets were a bunch of MODIS NDVI measurements at 1km resolution, reanalysis rainfall data, and elevation data from which we computed a local gradient (harder to grow stuff on hillsides). We did some wavelet based feature extraction and ended up with a handful of predictors on which we built the model.

This was all really cool stuff and we ended up with a little time series video of the changes in the extent of irrigated agriculture over 10 years in NW India including the Indus Valley. All this was related to monitoring the effects of an electricity management program (Jyotigram) that had been implemented in India meant to discourage illegal groundwater pumping while increasing the reliability of household electricity.

Anyways, it's been quite a while since I've worked on that kind of stuff and surely haven't thought much about how category theory might play a role. I could see a few general aspects of this type of problem that folks are thinking about using categories for, heterogeneous data integration using sheaves (Michael Robinson, Cliff Joslyn, Emilie Purvine) and the machine learning paper by Spivak, Fong, and Tuyeras, that John pointed to. There is clearly a lot more to be in done on both those fronts, particularly the machine learning front.

I'm curious though to know a little bit more about the types of problems you were hoping categories might help you tackle?

Comment Source:Hi all, I did some work a number of years ago training a 2-stage classification/regression model to distinguish irrigated crops in northwestern India during the Rabi (dry) season. We used logistic regression in the first stage to basically throw out everything that was obviously not irrigated (deserts, bodies of water, cities, etc.). The second stage assigned some number between 0-1 which was meant to roughly reflect the percentage of a pixel that was irrigated. This second stage was trained to distinguish forest from farm etc. and was trained using labels I generated by staring at lots of time stamped high-res satellite images. Our raw data sets were a bunch of MODIS NDVI measurements at 1km resolution, reanalysis rainfall data, and elevation data from which we computed a local gradient (harder to grow stuff on hillsides). We did some wavelet based feature extraction and ended up with a handful of predictors on which we built the model. This was all really cool stuff and we ended up with a little time series video of the changes in the extent of irrigated agriculture over 10 years in NW India including the Indus Valley. All this was related to monitoring the effects of an electricity management program (Jyotigram) that had been implemented in India meant to discourage illegal groundwater pumping while increasing the reliability of household electricity. Anyways, it's been quite a while since I've worked on that kind of stuff and surely haven't thought much about how category theory might play a role. I could see a few general aspects of this type of problem that folks are thinking about using categories for, heterogeneous data integration using sheaves (Michael Robinson, Cliff Joslyn, Emilie Purvine) and the machine learning paper by Spivak, Fong, and Tuyeras, that John pointed to. There is clearly a lot more to be in done on both those fronts, particularly the machine learning front. I'm curious though to know a little bit more about the types of problems you were hoping categories might help you tackle? 
• Options
11.
edited April 2018

Hello @Blake Pollard, thanks a lot for sharing your experience. Is there a paper related to it? Concerning categories, my initial intuition is that it can help me to gain a more realistic and important insight on machine learning and it would be my pleasure if I could try to contribute with some workgroup (is it already running?) in both fronts that you did cite. I have been drafting ideas but it would require some exchange by email... could you please send yours for me? best

Comment Source:Hello @Blake Pollard, thanks a lot for sharing your experience. Is there a paper related to it? Concerning categories, my initial intuition is that it can help me to gain a more realistic and important insight on machine learning and it would be my pleasure if I could try to contribute with some workgroup (is it already running?) in both fronts that you did cite. I have been drafting ideas but it would require some exchange by email... could you please send yours for me? best
• Options
12.
edited April 2018

Hi @BlakePollard, I did send you an email .On Cc others researchers and Prof Baez. If possible , please, I would like to know your thoughts about that exploratory conjectures and related temptives to frame it using the applied category approach and also count with the scientific collaboration of yours and Prof Baez groups. My guess is that the further steps on it would relies very much on the and Baez et al obtained results on framing passive eletrical circuits . Best

Comment Source:Hi @BlakePollard, I did send you an email .On Cc others researchers and Prof Baez. If possible , please, I would like to know your thoughts about that exploratory conjectures and related temptives to frame it using the applied category approach and also count with the scientific collaboration of yours and Prof Baez groups. My guess is that the further steps on it would relies very much on the and Baez et al obtained results on framing passive eletrical circuits . Best