ground truth pixels


Yesterday I came across the term ground truth pixels. It was embedded within a range of other interesting urban geotech terms such as confusion matrix and seed cities in a paper about remote sensing, that is, reading cities via satellite imagery. To me this is urban poetry. I was reading about the urban growth of a city in Western China, Chengdu which I visited earlier this year. The results of this remote sensing research had concluded that the urban agglomeration of Chengdu had increased in spatial terms by 350% between 1978 and 2002. That’s a staggering figure. I find it interesting reading these scientific papers about cities where most of the paper is dedicated to talking about the problems with the technology, that is the grey areas on the ground, fallow agricultural spaces or other urban voids that create confusion within the matrix, and so the technology has to be matched up with real field work to establish a ground truth pixel, one that is sensed remotely and then verified at ground level. That work is made even harder in places like Chengdu where there is so much atmospheric haze, air pollution and construction dust contributing to the confusion matrix. Yet this is incredible research, cities are increasingly complex, but such work can reveal the patterning of how growth occurs. Specifically in Chengdu, the inner city is already built out and so the growth happens on 7 central corridors that fan out in all directions from the city epicentre. These spaces, often activated by the presence of western corporations, operating within a high-tech research and development cluster, are the seed cities of a new emerging urban constellation. These are the cities of globalisation, remotely sensed, and more and more difficult to comprehend at ground level.

It follows of course that the expansion of the urban space corresponds with the loss of agricultural space, rural villages, etc, and so the value of green space within the city becomes increasingly important.

This map of a district of Chengdu (below) is a visual take on a confusion matrix. The left hand strip shows the smog barrier in Google Earth that hovers across the map of Chengdu. The rest of the image Google Earth have managed to clean up. (Thanks guys). I have coloured most of the city section in colours which probably express my New Zealand take on things. I did a search in the map space for ‘garden’, ‘park’, ‘temple’, ‘school’, ‘historic site’, ‘grotto’, and ‘teahouse’ and Google Earth couldn’t find anything. The magenta spaces are those spaces which are potentially confused pixels. The ground truth pixels are the green strip that flows through the city which is the Funan River and the green space is the Living Water Garden of Chengdu, a small constructed wetland that helps filter toxicity from the Funan River, and is for the local residents an extension of their living space, it’s a place to play cards, or music, to gather together even in the middle of winter…. it’s the kind of intimate, micro-urban cultural space that holds communities together, and its significance spreads beyond the pixel count of its mapping contours.

It’s tempting with urban sensing to only read the city from a satellite perspective, yet to focus on that is to give over to the global aspect of the contemporary city. There is a relationship between local and global, and the local is often hard at work to resist the homogenising impact of an economically-driven globalisation agenda. On the other hand its incredible that what happens locally registers at macro view. As anthropologist Arjun Appaudurai observes transurbanism, urbanism in the era of globalisation, provides new opportunities to develop an imagination, that finds its bearings at multiple scales. Just for the reference here is some urban ground research in Chengdu, China and another variation of ground truth pixels in the Living Water Garden…photos taken on a cold Saturday in January 2011….

Thanks to Annemarie Schneider, Karen Seto and Douglas Webster for their inspiring research…

Schneider, Annemarie, Karen Seto, and Douglas Webster. “Urban Growth in Chengdu, Western China: Application of Remote Sensing to Assess Planning and Policy Outcomes.” Environment Planning B: Planning and Design 32 (2005): 323-45.

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s