KNIME Image Processing (trusted extension)

clicar na imagem para seguir o link

clicar na imagem para seguir o link

Apenas um exemplo das fantásticas possibilidades do KNIME

KNIME Image Processing (trusted extension)

Fri, 12/03/2010 – 13:09 — knime_admin


Overview

The KNIME Image Processing Plugin allows you to read in more than 120 different kinds of images (thanks to the Bio-Formats API) and to apply well known methods on images, like preprocessing. segmentation, feature extraction, tracking and classification in KNIME. In general these nodes operate on multi-dimensional image data (e.g. videos, 3D images, multi-channel images or even a combination of them), which is made possible by the internally used ImgLib2-API.

Several nodes are available to calculate image features (e.g. zernike-, texture- or histogram features) for segmented images (e.g. a single cell). These feature vectors can then be used to apply machine learning methods in order to train and apply a classifier.

Currently the Image Processing Plugin for KNIME provides ca. 100 nodes for (pre)-processing, filtering, segmentation, feature extraction, various views (2D, 3D), etc. and integrations for various other image processing tools are available (see used and integrated libraries)

Future directions include a full, bidirectional integration of ImageJ2. Such an integration allow the users to use directly use/update ImageJ2 Plugins inside KNIME as well as recording and running KNIME Workflows in ImageJ2. Please see ImageJ2 Integration (BETA) for more information.

For the first steps please consider the KNIME Image Processing User Manual (incomplete draft!).

Important Links

Tags: ,

Poverty and Race in America

Uma boa representação gráfica interactiva

Uma boa representação gráfica interactiva

Strategies to tackle poverty, inequality, and neighborhood distress must be informed by local data. The history, geography, and politics of individual metro regions all matter profoundly, and any serious policy strategy must be tailored to local realities.
To help take the policy conversation from the general to the specific, we offer a new mapping tool. It lets you explore changes from 1980 to 2010 in where poor people of different races and ethnicities lived, for every metropolitan region nationwide.
Understanding how the geography of poverty has changed can provide essential context for answering questions like: Are some poor neighborhoods isolated from the region’s job opportunities? What would it take to connect them? Where should family support services be targeted? Which neighborhoods should be prioritized for improvements in essential amenities and opportunities? How can poor people across the metro landscape be better connected to the services and opportunities they seek?
For metro regions to systematically reduce poverty and expand opportunity, local civic and political leaders, advocates, and practitioners should start by sitting down together to understand the evolving realities of poverty, race, and place in their communities. We hope our maps help catalyze these conversations.

Tags: , ,

Income inequality seen in satellite images from Google Earth

Uso de proxis para identificar vizinhanças pobres

Uso de proxis para identificar vizinhanças pobres

Researchers Pengyu Zhua and Yaoqi Zhang noted in their 2008 paper that “the demand for urban forests is elastic with respect to price and highly responsive to changes in income.” Poor neighborhoods tend to have fewer trees and the rate of forestry growth is slower than that of richer neighborhoods.

Tim De Chant of Per Square Mile wondered if this difference could be seen through satellite images in Google Earth. It turns out that you can see the distinct difference in a lot of places. Above, for example, shows two areas in Rio de Janeiro: Rocinha on the left and Zona Sul on the right. Notice the tree-lined streets versus the not so green.

De Chant notes:

It’s easy to see trees as a luxury when a city can barely keep its roads and sewers in working order, but that glosses over the many benefits urban trees provide. They shade houses in the summer, reducing cooling bills. They scrub the air of pollution, especially of the particulate variety, which in many poor neighborhoods is responsible for increased asthma rates and other health problems. They also reduce stress, which has its own health benefits. Large, established trees can even fight crime.

Okay, I don’t now about that last part about fighting crime. Without seeing the data, I think that sounds like a correlation more than anything else, but still. Trees. Good.

Tags: , ,

High-detail maps with Disser

Software open source para trabalhar com mapas

Software open source para trabalhar com mapas

Open data consultancy Conveyal released Disser, a command-line tool to disaggregate geographic data to show more details. For example, we’ve seen data represented with uniformly distributed dots to represent populations, which is fine for a zoomed out view. However, when you get in close, it can be useful to see distributions more accurately represented.

If the goal of disaggregation is to make a reasonable guess at the data in its pre-aggregated form, we’ve done an okay job. There’s an obvious flaw with this map, though. People aren’t evenly distributed over a block — they’re concentrated into residential buildings.

So Disser combines datasets of different granularity, so that you can see spreads and concentrations that are closer to real life.

Tags: , ,

Useful Videos on Information Visualization

Bons videos sobre visualização de dados

Bons videos sobre visualização de dados

Noah Iliinsky – Data Visualizations Done Wrong – A Beautiful Collection of Stories and Tips for Success.

The Four Pillars of Data Visualization

Designing Data Visualizations with Noah Iliinsky

Best Practices for Data Visualization

Designing Data Visualizatins

Seeing the Story in the Data and Learning to Effectively Communicate – Inspired by Stephen Few Principles, Visualization Guru

David McCandless: “The beauty of data visualization” – Data Detective Telling Stories From Visualization of Information

This also has a nice quiz about visualization principles.

As I collect more, I will consolidate this list.

Tags: , ,

selfiecity

Um estudo sobre este tipo de fotos com muito boas visualizações

Um estudo sobre este tipo de fotos com muito boas visualizações

Investigating the style of self-portraits (selfies) in five cities across the world.


Selfiecity investigates selfies using a mix of theoretic, artistic and quantitative methods:

  • We present our findings about the demographics of people taking selfies, their poses and expressions.
  • Rich media visualizations (imageplots) assemble thousands of photos to reveal interesting patterns.
  • The interactive selfiexploratory allows you to navigate the whole set of 3200 photos.
  • Finally, theoretical essays discuss selfies in the history of photography, the functions of images in social media, and methods and dataset.

Selfiecity, from Lev Manovich, Moritz Stefaner, and a small group of analysts and researchers, is a detailed visual exploration of 3,200 selfies from five major cities around the world. The project is both a broad look at demographics and trends, as well as a chance to look closer at the individual observations.

There are several components to the project, but Imageplots (which you might recognize from a couple years ago) and the exploratory section, aptly named Selfiexploratory, will be of most interest.

The two parts let you filter through cities (Bangkok, Berlin, Moscow, New York, and Sao Paulo), age, gender, pose, mood, and a number of other factors, and this information is presented in a grid layout that self-updates as you browse.

So you can get a rough sense of how facets relate. There seems to be a higher proportion of female selfies and average age seems to skew towards younger as you’d expect. The average age of females in this selfie sample seems to be younger than that of males.

However, before you jump to too many conclusions about how countries vary or differences between the sexes, etc, consider the classification process, which was a combination of manual labor via Mechanical Turk and face recognition software. Age, for example, can be though to estimate from pictures alone since you have outside factors like makeup, angles, and poses. Do these things account for the two- to three-year average difference between the sexes? Maybe. So consider the data. But that should go without saying.

That said, Selfiecity is a fun one I spent a good amount of time browsing. It’s a weird, tiny peek into 3,200 people’s lives, with a dose of quant and art. And don’t miss the theoretical component in essay format, a reflection of social media, communities, and the self.

Tags: ,